The Reader’s Brain

Have you ever found yourself rereading the same sentence four
or five times and thought, “I should get more sleep”? Are you
clueless as to why one paragraph just seems to “flow” while you
simply can’t recall the contents of another? Guess what: you are
not alone. Even the best writers fail to grasp why their writing
works. The Reader’s Brain is the first science-based guide to writ-
ing, employing cutting-edge research on how our minds process
written language, to ensure your writing can be read quickly, as-
similated easily, and recalled precisely – exactly what we need
to transform anyone into a highly effective writer. Using the five
Cs – clarity, continuity, coherence, concision, and cadence – this
book combines irreverent humor with easy-to-follow principles
that will make readers perceive your sentences, paragraphs, and
documents to be clear, concise, and effective.

YELLOWLEES DOUGLAS is Associate Professor of Management Commu-

nication at the University of Florida.
The Reader’s Brain
How Neuroscience Can Make You a Better Writer

YELLOWLEES DOUGLAS, PhD

Clinical and Translational Science Institute and Center for
Management Communication, University of Florida
University Printing House, Cambridge CB2 8BS, United Kingdom

Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the

pursuit of education, learning and research at the highest international
levels of excellence.

www.cambridge.org
Information on this title: www.cambridge.org/thereadersbrain

© Yellowlees Douglas 2015

This publication is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.

First published 2015

Printed in the United Kingdom by TJ International Ltd. Padstow Cornwall

A catalogue record for this publication is available from the British Library

Library of Congress Cataloguing in Publication data

Douglas, J. Yellowlees, 1962–
The reader’s brain : how neuroscience can make you a better writer /
Yellowlees Douglas, Ph.D. Clinical and Translational Science Institute and
Center for Management Communication University of Florida.
pages cm
Includes bibliographical references and index.
ISBN 978-1-107-10039-8 (hardback : alk. paper) –
ISBN 978-1-107-49650-7 (paperback : alk. paper)
1. Reading – Psychology. I. Title.
BF456.R2D677 2015
808.001′9 – dc23 2015003341

ISBN 978-1-107-10039-8 Hardback

ISBN 978-1-107-49650-7 Paperback

Cambridge University Press has no responsibility for the persistence or

accuracy of URLs for external or third-party internet websites referred to
in this publication, and does not guarantee that any content on such
websites is, or will remain, accurate or appropriate.
For my tireless, endlessly proactive, and visionary acquisitions
editor – Rebecca Taylor at Cambridge University Press. Without
her interventions, this book would have been consigned to
oblivion – or Amazon’s CreateSpace.
CONTENTS

Acknowledgments .............................................................. viii

1 So much advice, so much lousy writing ............................ 1
2 The new science of writing .............................................. 10
3 Choosing words and structuring sentences
The first C: Clarity ......................................................... 29
4 Putting sentences together
The second C: Continuity ................................................ 61
5 Organizing paragraphs and documents
The third C: Coherence .................................................. 85
6 Maximizing efficiency
The fourth C: Concision ................................................ 118
7 Making music with words
The fifth C: Cadence ..................................................... 142
Supplement: Everything you ever wanted to know about
grammar, punctuation, and usage – and never learned ......... 164
Endnotes ........................................................................... 190
Select Bibliography ............................................................. 199
Index ................................................................................ 214
VIII SO MUCH ADVICE, SO MUCH LOUSY WRITING

ACKNOWLEDGMENTS

From the outset, I’ve silently thanked my initial round of anonym-

ous psychologists, linguists, and neuroscientists, who provided
spot-on advice on shifting the focus of The Reader’s Brain –
without a smidgen of the usual squeamishness academics exhibit
at a reader-friendly approach in academic writing. Dr. Kenneth
Heilman graciously provided me with a compelling narrative about
a stroke patient, as well as references to two cases that proved in-
tegral to my thinking about writing, cadence, and the brain. And
my credibility here would’ve been shot without the eagle eyes and
superb suggestions for revision I received from Professors Gordon
Pradl and Norman Holland. Both dramatically influenced the sens-
ible side of my career and saved my bacon here. Finally, I owe an
incalculable debt to Professor Sherrilene Classen, whose timely
email enquiring after the availability of this book prompted me
to submit the proposal one last time, after a decade of rejections.
Without the enthusiasm both she and Rebecca Taylor displayed
for the manuscript, this book would never have ended up in your
hands.
CHAPTER 1

So much advice, so much lousy

writing

Most people shun writing the way any chordate instinctively

shuns pain. The task of writing is inescapably labor-intensive,
no matter how facile a writer you are. Every blank page demands
its lines of coherent sentences and cohesive paragraphs that
ultimately amount to something like a rational, convincing argu-
ment. But our fear of writing reaches far beyond the hours we
know we’ll sweat over a keyboard, colonizing our blank screens
with words. Instead, most of us are less afraid of the hard work
than of grappling for hours with a complex system whose work-
ing parts we barely know. As a professor who has taught writ-
ing for more years than I’d care to publicly admit, I’ve heard
thousands of confessions that gush out of students. I’m a terrible
writer, they confess. Or, Writing’s my major weakness. Strikingly,
the majority of students who make these confessions are fairly
strong writers – just as the students who assure me that they’re
good writers tend to create fresh paragraphs whenever they feel
the reader needs to see a bit of white space, rather than from any
sense of a paragraph as a coherent entity. Obviously, some sort
of odd phenomenon must be at work here, when college students
and even seasoned professionals have no idea whether their writ-
ing skills are adequate for a stringer position on the New York
Times or barely pass muster as a child reporter writing for the
East Palatka Elementary Gazette. Try making a similar analogy
for reading or analytical skills, and you’ll discover most people
have a sound grasp of their abilities in these areas. But when the
discussion turns to writing, a disconcerting number of us find
ourselves at sea.
2 SO MUCH ADVICE, SO MUCH LOUSY WRITING

Three aspects of writing: micro, macro, middle

We struggle to even assess our writing ability because writing it-

self is inherently complex. Most forms of writing demand simul-
taneous attention to – along with at least some tenuous mastery
of – three aspects of writing: argument, correct usage, and the con-
stituents that make for clear, effective sentences and paragraphs.
Unfortunately, the first two items have reaped all the press.
Aristotle began a venerable, millennia-long history of writings on
argument that continue today in classrooms the world over, des-
pite Aristotelian notions of argument applying strictly to lengthy
orations that ran to hours and were aimed at illiterate audiences
with vastly different expectations and needs than any audience
alive today. And, of course, the usage and correctness mavens are,
as Christ described the poor, always with us – from the likes of
H. W. Fowler through to William Bennett. But Fowler was a pub-
lic school master, Bennett, a former Secretary of Education, and
John Simon, another outspoken grammar maven, is a film critic.
All of which proves you don’t need any bona fide credentials as
a linguist or researcher dedicated to the study of English to be
a grammar maven – just muscular opinions about subjects like
the correct use of less as opposed to fewer. You can master the
art of using the colon correctly – one of the more recherché rules
in the grammar canon – and also be on intimate terms with the
difference between logos and pathos and even recognize an exor-
dium when you see one, yet still write about as clearly as Forrest
Gump. Why? Between the macro side of writing – the features of
argument – and the micro side with its grammar and punctuation
exists a vast middle ground, where virtually all the grunt work of
writing occurs.
Ironically, the two ends of the writing spectrum collectively
account for the majority of advice on writing and are probably
responsible for the consumption of entire forests of virgin tim-
ber over the centuries. But the vast middle ground has attracted
relatively few experts. Moreover, to worsen matters, the handful
of experts on this middle ground offer wrong-headed advice to
writers. “Imitate published writers,” advises Richard Marius in A
Writer’s Companion.1 Unfortunately, this advice could land you
in rather hot water when you channel James Joyce in your next
 THREE ASPECTS OF WRITING 3

performance evaluation. Or, even worse, as Marius puts it in Item

8 of his Fundamental Principles of Sentences: “Begin a Few Sen-
tences with the Adverb There.”
I happened to glimpse this particular gem in a writing manual
I’d opened at random during a rant to a hapless sales assistant at
a bookstore and waved the book at him, demanding to know if
he realized this advice was the single worst recommendation you
could make to any writer. Not surprisingly – given the wealth of
misleading advice out there – he didn’t. In fact, he probably con-
sidered steering me toward the Self Help section and recommend-
ing I browse the titles on anger management.
Even the most well-intentioned how-to-write manuals give
us little concrete advice on all those burning questions that lie
uneasily just below the surface as we hunch over keyboards and
churn out sentences. How can you tell a good sentence from a
bad one? What distinguishes a well-written paragraph from a
crappy one? Are some word choices better than others? And how
in the hell do you follow Principle #17 from that ever-present
bible of writing advice, Strunk and White’s Elements of Style,
which counsels you to “Omit needless words”?2 After all, how
many people – outside of a few hundred thousand freshmen des-
perate to pad out a paper to a required word limit – have ever
mused, “Oh, only a ten-word sentence – I’d better toss a couple
of needless words in there”?
The reasons most writers have been struggling for years are
actually pretty clear-cut. If you want to write well, you’ll find a lot
of contradictory advice in those helpful manuals on writing that
attempt to address writing’s middle ground and which clutter up
the shelves at Barnes & Noble or Waterstones. Our libraries and
bookstores are groaning with mostly consistent guidance on using
punctuation and grammar and on crafting a convincing argument.
But am I the only one who thinks Strunk and White’s advice in
The Elements of Style, “Find a suitable design and hold to it,” is
just a little too similar to a Buddhist koan? What about Sheridan
Baker’s exquisite description in The Practical Stylist of a paragraph
as “a single idea … Like an essay itself, it has a beginning, a mid-
dle, and an end. The beginning and the end are usually each one
sentence long, and the middle gets you smoothly from one to the
other”?3
4 SO MUCH ADVICE, SO MUCH LOUSY WRITING

Whenever I read principles like these, I can’t help recalling the

episode of Monty Python’s Flying Circus where Anne Elk defines a
brontosaurus as being small at one end, rather large in the middle,
then small again at the other end. Sad to say, Strunk and White’s
advice on building paragraphs is about as precise and helpful as
Anne Elk’s bit of wisdom. They seem to be telling us to be con-
sistent – rather helpful advice in life as in writing. But our chief
difficulty is then figuring out what the hell they mean by “a suit-
able design,” which seems as elusive as the fine art of distinguish-
ing necessary from unnecessary words. Does this strategy involve
winnowing out those pesky, insignificant words like in, to, and
of? Developing our vocabularies – or avoiding polysyllables at all
costs? Or maybe they have something else entirely in mind: a para-
graph that describes a complex topic, fat with sentences so long,
they make Henry James seem like Hemingway – or the New York
Times read like The Sun. And, while we’re at it, does my para-
graph have a beginning that falls into Baker’s one-sentence defin-
ition, followed by a rather larger middle that gets tidily wrapped
up by my concluding sentence?
If you’re not quite confused yet, we could always obey Strunk
and White’s golden rule about suitable design, then follow A
Writer’s Companion and imitate the practices of published writers.
If we just channel Henry Miller, Virginia Woolf, or ee cummings,
we might get out of this particular rut,

although this
might not be
a
good
idea, if u
think about
it.

The science of writing

Ultimately, all our problems with writing have a common source:

precisely how all these experts arrived at the “principles” they
 THE SCIENCE OF WRITING 5

claim lead to good writing. If you emulate what published writ-

ers have written – or, more accurately in many cases, got away
with – you’re not exactly treading a sure path to good writing.
For instance, you run the risk of being mistaken for an ESL writer
if you imitate the likes of Herman Melville. Writing researchers
C. H. Knoblauch and Lil Brannon discovered this possibility when
they slipped a swatch of Moby-Dick in with two student essays and
asked teachers of writing to identify its strengths and weaknesses.
The teachers’ verdict: Melville’s paragraphs had “some evident
virtues, especially in [their] occasionally sophisticated word-
choices … Sentence fragments abound, and the phrasing seems
unnatural at times.” The teachers decided Melville was probably
an “advanced English Second Language student.” Melville’s mas-
terpiece earned him a berth in a remedial writing class, not the
impression one wants to make in applying for a job or writing
a legal brief.4 In addition, you will also never learn why some
particular phrasings are better than others. Imitation works beau-
tifully if you’re learning the violin via the Suzuki Method. Imita-
tion works for your average ten-year-old who can’t read music
but has parents with ambitions to produce the next Itzhak Perl-
man. However, imitation fails to work quite so well for a middle-
level manager charged with writing a proposal. Come to think of
it, unless you can locate a Warren Buffett rewrite of some thor-
oughly unintelligible bit of business prose, you’ll have difficulty
finding any stellar models of good writing in most professions, let
alone for most of the kinds of writing our everyday lives require.
Moreover, lore inherited from generations of not-particularly-helpful
reference manuals on good writing is equally suspect. If you pre-
fer active construction to passive, as so many books counsel,
your writing will become more efficient and concrete. Yet virtu-
ally none of these experts explains why active construction works
better than passive – or that this principle fails to apply in every
situation.
Nearly all books on writing’s under-explored middle ground
deal merely with the surface, with messing around with words
on the page, or with practices observed by what are usually arty
writers with some credentials and a couple of books in print. But
writing is always a transaction, a means to extend our convictions
beyond the reach of the human voice, across time and space. At
6 SO MUCH ADVICE, SO MUCH LOUSY WRITING

the receiving end: the hapless reader, confronted with the message
you’ve sent. Even if you’ve written in English so plain you could
put it on the label of a can of dog food, your readers have a fairly
horrific amount of work to do – deciphering your meaning, min-
imizing ambiguities, pinning words into grammatical categories,
filing your information away, comparing it against other informa-
tion they already possess, and deciding which details merit trans-
fer from fleeting short-term memory to more durable long-term
memory.
Reading itself is a highly complex act. Until relatively recently,
reading was what social scientists like to call a black box, a process
where we know the inputs and the outputs but not the mechanism
that translates one into the other. Back in the 1970s and 1980s,
researchers in what was once the field of Artificial Intelligence,
known familiarly as AI, pursued research on how readers under-
stand written language, primarily because they were interested
in building computers that could read. This research in psycho-
linguistics and cognitive psychology began shedding light on the
mental processes that enable us to make sense of words on a page.
By the early 1990s, however, most AI researchers were willing to
admit that the field was something of a will-o’-the-wisp – building
a machine that could think and read seemed like a good idea at
the time. But after a few decades of watching computers still strug-
gling to recognize speech or tell stories that could be handled with
aplomb by an eighteen-month-old, most researchers moved on.
Despite the oomph and funding oozing away from AI research,
research into the neuro-cognitive process of reading continued
apace, in psycholinguistics, neuroscience, and cognitive and
educational psychology, bolstered by new imaging technologies
that offered glimpses of our reading brain. But this flourishing
research on the act of reading remained utterly disconnected from
research on writing, which seems like a puzzling omission, given
just how eloquently psycholinguistic and neuro-cognitive studies
on reading speak to what defines a clear sentence or coherent
paragraph. However, in reality, this omission is hardly puzzling,
given the way academics operate in discipline-specific silos that
discourage them from venturing onto their colleagues’ turf, not
to mention the thoroughly daunting vocabulary required for the
 THE SCIENCE OF WRITING 7

average humanist to wade through research results reported in the

likes of Science or Cerebral Cortex. In addition, scientists rely on
validated tests and expensive technologies from eye-tracking de-
vices to functional magnetic resonance imaging (fMRI) to conduct
research on the reading brain. In contrast, the humanists charged
with teaching and researching writing have writing samples, ques-
tionnaires, and their own powers of observing fledgling writers
at work. In terms of resources, the scientists might as well be
Goliath, and the composition researchers and instructors, David –
without the sling-shot.
The connections seem obvious between what neuroscientists
and psycholinguists have learned about the reading brain and
what writers need to know when they sit down with a blank page.
Yet the science of reading and the teaching of writing end up as
two conversations conducted in parallel – different audiences,
tuned to entirely different channels. In fact, I dedicated a decade
to studying research on the reading brain to gauge the impact of
hypermedia environments on the act of reading, all while strug-
gling to teach students to write and generally finishing each course
feeling as though I should offer the students a tuition refund. I
only realized I could leverage the research on reading to teaching
writing when I accepted an invitation to spend a half-day teach-
ing lawyers how to write readable legislation. Lawyers, I decided,
would require hard data to change gems like the 290-word sen-
tence on wire-tapping that appears in the California Penal Code
Section 631a5 into something resembling a string of sentences you
could comprehend on the first, rather than the twelfth, reading.
And, I realized, I had hard data in spades that I could translate
into principles to guide writers.
Somewhat like music, writing is a system. Languages have rules
about structure – where you put the subject, where you put the
verb. Moreover, readers have an unconscious preference for cer-
tain types of sentences – a fact well established through decades of
research into how brains process language. Put simply, when you
write a sentence, your readers’ brains will process that sentence
in highly predictable ways, despite their blissful unconsciousness
of all the cogs and wheels whirring as they scan the page. The
same also holds true for groups of sentences, paragraphs, entire
8 SO MUCH ADVICE, SO MUCH LOUSY WRITING

documents. So you ignore these predictable processes at your

peril – and, unfortunately, also at your readers’.
This thumbnail history brings us, at last, to the good news.
You can quickly and painlessly master the art of becoming a
terrific writer. Or, at least, you, too, can be one of those hap-
less saps in the office who always gets lumbered with the job
of writing documents simply because you do it so well. And the
process is not only relatively simple; this method also works
across virtually every field, profession, and type of non-fiction
writing. In addition, this method uses a systematic, rules-based
approach, well suited to the learning styles of engineers, sci-
entists of every stripe, and everyone who prefers an approach
to writing based on tangible data, rather than on instinct and
verbal facility – or flailing around in the dark and hoping for
the best. Follow most of the principles in The Reader’s Brain,
and your readers, your colleagues, and, more important, your
superiors and clients will find your writing a model of precision
and effectiveness. Best of all, this method stems from decades
of scientific research in linguistics, cognitive psychology, and
neuroscience. Moreover, I’ve road-tested this method for nearly
a decade in writing courses for undergraduates in more than
fifty disciplines, MBA students and mid-career executives, law-
yers, engineers, and faculty in every branch of the biomedical
sciences.
To paraphrase Scott Adams, the brains behind the American
comic strip Dilbert, I’m writing from a position entirely different
from the usual consultant or professor, mercifully insulated from
the insults and challenges of workplace writing – a perspective
Adams likened to “writing a first-hand account of the experience
of the Donner party, based on the fact that you’ve eaten beef
jerky.”6 Like Adams, I’ve gnawed some ankles. In fact, I’ve de-
voured some femurs while working as a copywriter for blue-chip
clients and writing in the trenches in seventeen disciplines. I’ve
tackled everything from white papers on vaccines for C-suite ex-
ecutives to articles straddling rheumatology, genetics, and cardi-
ology. And my work tends to get accepted rapidly due substan-
tially to the ease with which editors, executives, and reviewers
can read it.
 THE SCIENCE OF WRITING 9

The pages that follow translate the research I’ve long relied on
in psycholinguistics, cognitive psychology, and neuroscience into
easy-to-follow principles that will enable you to
• construct sentences that efficiently convey your message,
regardless how complex it is;
• understand where to place important information (and where
to hide bad news to avoid unduly pissing your readers off);
• distinguish between good and poor word choices;
• create paragraphs that fit together seamlessly;
• introduce complex information without confusing the pants
off your readers;
• make your sentences just seem to “flow” like an expert
writer’s.
The Reader’s Brain uses five categories to promote clear, effective,
and efficient writing, the five Cs: clarity, continuity, coherence,
concision, and cadence. Practice most of these simple, easy-to-
follow principles, and you’ll not only become a good writer, you’ll
also become a pro at spotting – and fixing – even the worst writing
disasters.
CHAPTER 2

The new science of writing

We owe a good deal of what we now know about the reading brain
to a 1980s idée fixe – the scientific equivalent of padded-shoulder
suits and even bigger hair: that computers could be taught to
think, read, and play a mean game of chess. In retrospect, this sort
of optimism is entirely understandable, since during the eighties,
computers rapidly evolved from do-it-yourself Radio Shack-style
jalopies with a fraction of the computing power of your run-of-the-
mill modern cell phone to Maseratis capable of parallel process-
ing. During the eighties and early nineties, computers progressed
a generation in speed and capacity every two to three years. This
trend tidily observed Moore’s Law that predicted transistors and
integrated circuits would double in capacity approximately every
two years – a prediction that only proved uncannily apt, given
that Moore made his prediction in 1965, the same year the first
commercially successful mini-computer debuted.1 Small wonder,
then, that scientists in AI believed computers could also evolve in
a matter of decades into the thinking creatures humans had taken
millennia to become.
Now, not so many years later, we’re ready to concede that HAL
might not be around in 3001, let alone 2001, even though comput-
ers, it turns out, really can play a mean game of chess and even
win at Jeopardy. In 1997, IBM’s supercomputer Deep Blue won a
six-game match against world chess champion Garry Kasparov –
thrashing him so badly that the rattled Kasparov claimed the geeks
behind Deep Blue were using a human chess master to control the
computer’s gambits. Still, however, computers cannot read – at
least, not in the conventional sense of poring over lines of writ-
ten symbols and arriving at an understanding of what Anna and
Count Vronsky were up to in Anna Karenina. Nevertheless, for
more than a decade, AI became the equivalent of the Klondike
 READABILITY FORMULAS 11

Gold Rush, propelling scores of talented researchers and, more

important, streams of research funding into studies of the read-
ing brain. And the results, from our vantage point, proved more
valuable than the equivalent of an Anna Karenina-reading HAL.
The fruits of AI research into reading, as well as the neuroimaging
studies of the reading mind that followed, have yielded a rich pic-
ture of how our minds process written language. With this picture,
we can understand the characteristics that distinguish writing that
can be read quickly, assimilated easily, and recalled precisely –
exactly what we need to transform anyone into a highly effective
writer.

Readability formulas: a non-narcotic sleep aid

As any teacher of technical writing will tell you, readability for-

mulas are hardly new. Used for years in technical writing and edit-
ing, readability formulas rely on the sort of chestnuts that old-time
editors with green eye-shades tended to pass out to cub reporters.
Keep it simple. Keep it short. However, when these formulas were
embodied in software style-checkers beginning in the late 1980s,
they proved considerably less useful than a curmudgeonly editor.
The simplest of these formulas relied on the one aspect of writing
anyone could easily quantify – syllables in words and numbers of
words in sentences. However, Flesch’s Reading Ease score trans-
formed that simple act of counting with an equation that appears
more terrifying in its simplified form than the prospect of revising
the 290-word nightmare sentence from the California Penal Code
Section 631a (reproduced in its full glory in Chapter 1’s note 5):
Score = 206.835 − (1.015 × Average Sentence Length)
– (84.6 × Average Syllables per Word).
If you’re hankering after something a bit less numeric and more
user-friendly, you might turn to the Flesch–Kincaid score, which
correlates the Flesch Reading Ease score to American standards
for reading at grade levels and to the estimated percentage of the
US adult population capable of reading at those grade levels. A
score of 0–30 designated a document readable only by univer-
sity graduates, while a score in the 90–100 range was readable by
12 THE NEW SCIENCE OF WRITING

a fifth-grader, and, by extension, readable by over 93 percent of

American adults by Flesch’s reckoning. Perversely, the higher a
Flesch Reading Ease score, the easier the reading of the sentence
or passage. In contrast, lower scores on the Flesch–Kincaid indi-
cate easier reading, with Dr. Seuss’ Green Eggs and Ham scoring
an improbable −1.3. This score suggests the ideal reader of Green
Eggs and Ham is a fetus. However, if you break down the compon-
ents of the actual text of Seuss’ book, its Flesch–Kincaid score
hints that the book’s fifty words contain only eight polysyllabic
words, all of them the word anywhere. If a negative grade level
for reading is difficult to picture, American readers might prefer to
recall a member of the US Congress filibustering by reading Green
Eggs and Ham aloud at the sort of speed a teacher might adopt for
a class of not-terribly-bright kindergartners.
However, Flesch and Flesch–Kincaid scores hardly resemble
the handy formula that tells you how to compute your maximum
heart rate (220 – [your age] = maximum heart rate). As a result,
researchers tried a simpler formula, now embodied in Gunning’s
FOG Index:
Grade Level = 0.4 (average sentence length
+ hard words of more than two syllables).2
If you feel an overwhelming urge to run and drink something alco-
holic after struggling to apply these formulas to your writing, you
have plenty of company. Moreover, you’re also giving in to a thor-
oughly sane impulse. Flesch, Flesch–Kincaid, and the FOG scores
are excellent at ensuring reading prompts in an experiment are all
more or less comparable in their demands on readers’ educations
and brains. However, none of these formulas provides much in the
way of insight into the challenges your sentences throw at read-
ers. For starters, the number of syllables in a word isn’t a reliable
gauge of its difficulty. Specifically, praxis and model have the same
number of syllables, but few fifth-graders would have a nodding
acquaintance with praxis, for all its two-syllable length.
Moreover, readability formulas are deceptive, as they rely on a
relentless counting of syllables and words to arrive at a sense of
how challenging a reader might find any sentence or document.
Yet a brief sentence can prove more demanding on your read-
ers’ brains than a sentence of three or four times its length. For
 READABILITY FORMULAS 13

example, consider this sentence, drawn from The New Yorker, a

magazine that famously boasted three full-time grammarians on
staff, none of whom seems to have been on duty the day this gem
went to press:
What those of us who know Agee’s criticism almost by heart
read over and over, however, is the reviews that appeared in
The Nation.3
The sentence runs to a mere twenty-four words, of which nine are
polysyllabic, if not exactly the “hard” words of Gunning’s FOG
Index. However, the sentence’s structure or syntax is so complex
that writer David Denby himself was a bit fuzzy on which word
represented the sentence’s grammatical subject – apparently, the
same confusion that afflicted the skeleton staff of The New Yor-
ker’s vaunted grammarians. The sentence’s grammatical subject
is a noun clause, the most subtle of the clauses to detect in Eng-
lish, with a prepositional phrase and an adjective clause wedged
between the subject of the noun clause and its verb. Noun clauses
tend to fly under readers’ radars for two reasons: (1) they lack
clear-cut cue words and (2) they rely on strings of nouns and
verbs to represent a single noun. Confused? You ain’t seen noth-
ing yet. Look again at the detail the reader’s brain must grasp as
a whole to understand the grammatical subject of the sentence:
What those of us who know Agee’s criticism almost by heart read
over and over. Denby has written a sentence that begs to be re-
read and not in the savoring-the-words way one uses on the clos-
ing sentences of The Great Gatsby. Instead, we reread Denby’s
sentence because we’ve experienced what linguists call a “garden
path” moment where our initial predictions about what we’re
reading – the meaning of the words, phrases, and clauses based
on parts of speech – have led us totally astray. In this sentence
What is paired with the verb read, with the modifying adjective
clause who know Agee’s criticism rearing its head confusingly
between the subject and verb. In addition, readers process sen-
tences most easily when the verb closely follows the grammatical
subject, while Denby’s sentence chucks ten words between the
subject What and its verb read. For the record, according to the
text analytics on my writing software, Denby’s twenty-four-word
sentence scores 62.6 on Flesch, 10.5 on Flesch–Kincaid, and 12.9
14 THE NEW SCIENCE OF WRITING

on FOG. In other words, the sentence we just stumbled over

would pass most editors’ sniff tests for readability. As Dilbert
readers might say, “Go figure.”
Compare the burden of reading Denby’s sentence to this sen-
tence, drawn at random from Lynne Truss’ Eats, Shoots & Leaves,
a paean to punctuation:
Using the comma well announces that you have an ear for
sense and rhythm, confidence in your style and a proper
respect for your reader, but it does not mark you out as a
master of your craft.4

Even though Truss begins with a gerund phrase, using the comma,
acting as a grammatical subject, and the sentence runs to thirty-
eight words, the readability analytics give her sentence a thorough
drubbing, with a Flesch score of 50.3, a Flesch–Kincaid of 15.7,
and a FOG of 17.3.

Bewildered by gerunds and grammar generally?

If you find yourself equating gerunds with rare, tropical dis-

eases, consult the Supplement, “Everything You Ever Wanted
to Know about Grammar, Punctuation, and Usage – and Never
Learned,” for a thorough, demystifying discussion of gerunds
and the roles they play in English, along with the other parts of
speech and of sentences, as well as a brief guide to the vagaries
of English punctuation.

By all readability measures, Truss’ sentence is more difficult

than Denby’s, even though most readers can run through it quickly
and understand it without feeling as though they had hacked their
way through the dense verbiage characteristic of a fund prospec-
tus for your 401(k). In fact, along with its irreverent wit, the sheer
readability of Truss’ prose contributed to the surprise bestseller
status of Eats, Shoots & Leaves. The problem with the reliability of
readability formulas is similar to the challenges facing computer
scientists attempting to simulate reading with a machine. Count-
ing will only get you so far.
Pity the poor writer who actually pores over these textual ana-
lytics, attempting to make her sentences more readable. At worst,
 READABILITY FORMULAS 15

the analytics might prove as illuminating as the software in David

Lodge’s novel Small World, which helpfully told a famous work-
ing-class author of the Angry-Young-Men stripe that his most com-
monly used word in every novel was greasy. (Unsurprisingly, after
that revelation, Lodge’s fictional and formerly prolific writer found
himself utterly blocked, seeing greasy taunting him in every line.)
At their most useful, readability formulas can encourage writers to
prefer shorter sentences to longer ones. But if, God forbid, you slav-
ishly adhere to the guideline of sentences optimally containing an
average of twenty words, an entire document containing sentences
of such a uniform length will put your readers to sleep. You will end
up creating not a highly readable, compelling article or proposal
but, instead, the perfect non-narcotic sleep aid for anyone squeam-
ish about relying too heavily on Ambien to help them drift off.
In contrast, readability characteristics come to us via decades
of research into the way our minds process written language.
Studies measure eye movements or saccades, borrowing from the
French term that literally means “jerks,” an accurate description of
the way your eyes skip across a printed line. By examining readers’
saccades using eye-tracking devices or other imaging technolo-
gies, researchers can determine the efficiency with which readers
comprehend sentences. Electrical impulses recorded at the scalp
measure the brain’s reactions to lines of print in milliseconds and
microvolts. Magnetic imaging captures the brain processing para-
graphs. Even studies that fail to use any sexy imaging technology
still effectively record reading times, comprehension, and recall
of a variety of paragraphs, sentences, and choices of words. With
this rich body of research, spanning disciplines from linguistics
to neurology, we can gain a full-fledged picture of precisely how
readers transform black marks scattered on a white page into, say,
details of the hurricane-force winds forecasters are predicting to
hit your locale in the next 24 hours.

Imaging the reading brain

Scientists today use five different neuroimaging technolo-

gies as a window on the reading brain. The oldest technol-
ogy, electroencephalograms or EEGs, measure low amplitude
16 THE NEW SCIENCE OF WRITING

electrical activity at the scalp. Positron emission tomography

or PET scans display blood flow as the brain processes infor-
mation. But newer technologies like functional MRIs (fMRIs),
magnetoencephalograms (MEGs), and diffusion tensor imag-
ing (DTIs) offer far greater precision and resolution in the im-
ages of the working brain that they produce. Nevertheless,
each neuroimaging method, even fMRIs, MEGs, and DTIs, has
its drawbacks. EEGs generate such weak signals – less than
the equivalent of a flashlight battery – that researchers need
to average results over multiple trials to arrive at what they
call “event related potentials” or ERPs. In addition, ERPs only
register signals from the cerebral cortex, independent of the
involvement of deeper brain structures. Similarly, PET and
fMRI scans lack the time resolution to capture events as fleet-
ing as the brain recognizing a word. DTIs, for all they yield
in terms of high-resolution, three-dimensional imaging of the
brain, provide minutely detailed maps of neural connections
but only give us insight into the brain’s white matter (the
glial cells and axons that forge connections) and not its gray
matter (the parts of the brain current research clearly links to
cognitive processing). Currently, only MEGs have the capac-
ity to capture brain activity at the level of milliseconds. But,
like most new technologies, MEGs are prohibitively expensive
and, as a result, are currently rarely used in research. Conse-
quently, most studies tend to rely on multiple forms of neu-
roimaging to compensate for the shortcomings of individual
technologies.

The mechanisms of reading: see Jane read.

Read, Jane, read

The more researchers study reading, the more surprised they are
by its sheer complexity. Reading comprehension is a multi-stage
process, all researchers agree, although few studies have estab-
lished definitively precisely how many stages actually occur be-
tween your eye falling on an article containing an astrologer’s
prediction in The Sun and the moment you finish it and begin
 THE MECHANISMS OF READING 17

wondering whether you should book that Christmas holiday in

Mexico if the world is slated to end before October 31st.
Not surprisingly, the first stage, agreed on by all researchers,
involves recognizing words. Using eye movements, scientists have
tracked the length of time our eyes stop or fixate on individual
words and linked it with EEG activity, measuring the amount of
brain activity, or cognitive load, involved in identifying a word.
Skilled readers, the sort who down entire issues of The New Yorker
with coffee, pause on words for only 300 milliseconds, a mere
fraction of the blink of an eye. This sort of speed translates into
400–500 words per minute and, in readers trained to take in entire
sentences in a single eye fixation, up to a staggering 1600 words
per minute – or one word every 40 milliseconds.5 But the speed of
our eye movements depends entirely on the context surrounding
the word. The more specific the context, the more constraint those
surroundings put on the individual word, and the fewer meanings
we’re likely to attach to it. Since most languages tend to assign
multiple meanings to a single word, even the meanings of indi-
vidual words are never completely certain. Moreover, words that
sound like other words – the old homonyms like principle and
principal that continue to dog most of us well past our twelfth
birthdays – complicate the identification process, causing more
brain activity and slowing down the speed with which our eyes
move across the page.
While scientists measure eye movements and EEG microvolts,
other researchers use neuroimaging to pinpoint the areas respon-
sible for the two primary phases of the reading process. Here,
again, EEGs and fMRIs reveal different areas in the brain tackling
the two phases. Wernicke’s area, the area of the brain respon-
sible for processing both spoken and written language, has been
identified as central to word recognition. But evidence also points
to interaction with the bulge behind Wernicke’s area, the angu-
lar gyrus, which borders regions that handle visual, spatial, and
language skills. The angular gyrus may be involved in the sec-
ond phase of reading: relating individual words to the words that
surround them in a sentence. This process, however, must also
involve Wernicke’s area, since our recognizing individual words
hinges on our ability to narrow their range of meaning according
to the context.
18 THE NEW SCIENCE OF WRITING

In the blink of an eye

Readers take, on average, 200–300 milliseconds to identify

a printed word. But experienced readers, in studies that ex-
posed them to partially blacked-out words in brief flashes,
were able to begin the process of identifying words within
50 milliseconds.6

If you’re still not convinced that reading the gossip column in

the New York Post is the mental equivalent of doing the 100-meter
butterfly, you might want to reconsider because the second phase
of reading involves multiple types of processing. Readers rely on
syntactic processing to assign meaning to words based on their
place in the sentence’s structure or syntax. For example, many
words in English act like baseball’s utility players: the most flex-
ible of them can act as nouns or verbs or, with some additions
like -ed or -ly, adjectives or adverbs. Even words that we tend to
treat as simple verbs like throw can also easily act as nouns: That
throw was lousy. Moreover, some words, while handily inflected in
speech to indicate their status as noun or verb, remain ambiguous
on the page. Rebel can serve as a noun – He’s a rebel [REBel] –
and, of course, as a verb, He’s about to rebel [reBELL].
We make sense of words during the syntactic phase of process-
ing the same way we recognize words on the page. We rely on
surrounding words to anticipate how the sentence will structur-
ally play out. And we anticipate what’s coming next, as with so
many other operations in the brain, by projecting the most com-
mon, most predictable scenarios. Here, the brain’s tendency to
anticipate the most common scenario dovetails rather neatly with
the old chestnut beloved of professors in medical school: When
you hear hoofbeats, think horses, not zebras. In other words, an-
ticipate what you’re most likely to see in context, which, unless
you happen to be standing on a Kenyan savannah, will tend to
be horses. As a result of our brains’ continually anticipating the
familiar, readers of English expect to see the main noun of any
sentence before they encounter its main verb. So our brains expect
and process easily I dodged the bullet, but not The bullet I dodged,
 THE MECHANISMS OF READING 19

which, while actually still comprehensible, tends to sound as if

the sentence belongs more to the realms of poetry than reportage.
Still, in a simple sentence, this sort of syntactic departure is
a curveball our brains handle easily, as we can sort out the re-
lationships between subject, verb, and object without too many
machinations, especially since the content is equally straightfor-
ward. Consider, however, a sentence where the complexity of the
content is mirrored by the equally gnarly sentence structure, a sce-
nario more typical of most writing we encounter than The bullet I
dodged:
Thirteen of the 27 genes significantly up-regulated at short
reperfusion but not at long reperfusion encode for known
transcription factors or inflammatory cytokines, suggesting
roles in gene transcription and regulation at this early
reperfusion time point.

This sentence, written by a faculty member in medicine, puzzled

the other faculty in my writing course because the faculty expect-
ed to see the first likely culprit for the sentence’s main verb early
in the sentence, not far behind the sentence’s subject. Instead, the
sentence led us down a garden path similar to the one in Denby’s
sentence in The New Yorker. In my writing course, we all battened
onto up-regulated as the sentence’s verb, a tactic which played
out beautifully until we hit encode, the sentence’s real main verb.
Since up-regulated was accompanied by but not at long reperfu-
sion, our predictions seemed momentarily confirmed, as but not
tends to be paired with verbs in most sentences we encounter.
Yet, in this instance, the arrival of the real verb, encode, promptly
destroyed that prediction and made a hash of our comprehension
of the main part of the sentence: its subject and main verb. This
development required us to soldier on to the end of the sentence,
attempting to make sense of it, and then to reread the sentence. Or
we could begin the sentence over again with a different prediction
of its structure. The rereading and new prediction should reveal
that up-regulated is actually part of a modifier referring to genes –
not a verb at all.
While not terribly common, garden path sentences reveal sev-
eral key features of our word- and sentence-level processing. First,
20 THE NEW SCIENCE OF WRITING

the meanings of words are inextricably linked with their situation

within any sentence’s structure. Up-regulated as a verb is a rather
different animal than up-regulated the adjective, despite the verb’s
seemingly narrow and highly technical use in an article on genes
and lung transplants. Second, when your readers hear hoofbeats
and expect horses, introducing them to a herd of zebras might be
a handy means of snagging their attention when you’re creating a
work of art but is generally disastrous in any other form of com-
munication. Efficiency and clarity in writing, as in most forms of
communication, result when writers and readers rely on the same
sets of expectations. We process sentences most efficiently, requir-
ing less brain activity and less time for reading, with only brief eye
fixations, when the sentence conforms to our predictions. In these
instances, when words look like verbs, they also act like them.
Prediction is, after all, not only a key element in the process
of reading but also a typically uni-directional affair. Backtracking
on predictions means only one thing: you were wrong. Reading
requires us to generate hypotheses, then test them, sentence by
sentence. This process remains subconscious only as long as our
predictions play out, and we keep moving forward on the page.
Similarly, readers must backtrack in sentences whenever their pre-
dictions about word meaning and sentence structure fail to play
out. But reading, like prediction, should also be uni-directional.
Reading requires a substantial cognitive load merely in complet-
ing the process of identifying, predicting, and confirming. Reading
is only efficient when the process is looking forward, not bogged
down in backtracking, especially since the process also requires a
further step that links word- and sentence-level predictions.

Inference building: see alligator chomp. See Jane run

Obviously, recognizing the meanings of words and understanding

the roles they play within a sentence’s structure represent only
two steps in the reading process. To comprehend, assimilate, and
recall written language, however, we need a further phase in the
reading process: inference building. Beginning readers, like the
first-graders who run their fingers under the words as they read,
focus on word and syntax processing. But once you leave behind
 INFERENCE BUILDING 21

the See Jane read. Read, Jane, read stage to start reading about
hobbits, the NYSE, and deficit spending, your reading process
must encompass inference building to convert nouns and verbs
into actions, abstractions, and theories. Consider, for example, the
demands placed on us when we read even the short sentence,
Horace knew he was going to have to break the lock to get free.
To make sense of this sentence, we have to envision a scenario,
one that involves identifying exactly what kind of lock Horace is
struggling with. If we already know Horace is doing time at San
Quentin or Wormwood Scrubs, we immediately perceive lock as a
metal object that prevents inmates from escaping their cells. But
if we know that Horace is a wrestler, struggling to qualify for the
Olympic team, the word lock takes on a very different meaning –
signifying the sort of hold an opponent has on Horace.
As even this simple example illustrates, all three phases of pro-
cessing work in concert. First, we identify the words and their
meanings, relative to their role in the sentence’s structure. Then
we confirm their accuracy by also measuring them against the
contents of our long-term memories. Specifically, we access our
long-term memories to verify what we read against what we know
of the world. According to some researchers, we build inferences
by drawing on our long-term memory to compare the sentences
we read to what we know of the world around us, relying on as
many as five categories to turn sentences into scenarios: time,
space, actor, cause, and intention. Or, put more simply: who,
when, where, why, and how. In addition to being staples of jour-
nalistic enquiry, the words who, what, when, where, how, and
why enable readers to see sentences as tightly connected. Readers
use inference-level processing in making sense of sentences, but
they also tend to rely on explicit inferences more heavily in see-
ing sentences as connected. In addition, we read faster and recall
content better when sentences use transitions that explicitly flag
causation.7 For example, consider this simple, concrete scenario:
Suddenly, John fell to his knees and began retching violently,
as if the pizza were poisoned. Mary screamed.

Nearly everyone would automatically assume that Mary screams

in reaction to John retching on the floor – either in horror at the
22 THE NEW SCIENCE OF WRITING

possibility he’s been poisoned or in genuine fear at the damage

he’s about to inflict on her precious Persian rug. But, in fact, caus-
ation is the unseen glue that holds most sentences together. We
assume that sentences that follow each other contain events that
also follow each other. John retches first, then Mary screams. Our
brains, once again, are opting for the obvious, the usual connec-
tion. This bias stems from the rarity with which we encounter
sentences from two entirely different scenarios, involving totally
different casts of characters, that nevertheless follow one another
sequentially. The same holds equally true for nearly any sequence.
If I begin to count 1, 2, 3, 4, 5, you’re certain to assume the next
number will be 6 rather than 79 or 4586.

Space and time

Researchers have established that readers assume that sen-

tences that follow one another inevitably reflect events that
follow one another, what linguists have dubbed the iconicity
assumption.8 In other words, we unconsciously assume, as we
read, that order on the page reflects the order in which events
unfolded. This assumption probably reflects our brains’ pre-
dilections for perceiving events in sequences that play out in
terms of causes and effect.

If you ask cognitive psychologists like Jerome Bruner, caus-

ation is not simply glue for binding sentences together. Causation,
in his view, enables stories to exist.9 After all, stories happen be-
cause something – a misplaced letter, a lottery win, a car wreck,
a loaded glance in an elevator – throws the status quo seriously
out of whack. What happens next is invariably a reaction to that
cause, which causes a long train wreck of reactions until the story
finally reaches a new status quo and The End. Moreover, for forty
years psychologists have demonstrated just how centrally cause
and effect figure in human perception, observing the ability to
detect cause and effect in infants as young as six months.10 When
you think about it, the ability to detect cause and effect is not only
central for learning – touch the stove and toast your fingers – but
 SCHEMAS 23

also central for survival. Get too close to the alligator, and you
might end up serving as lunch.
But cause and effect never exist in a vacuum. Instead, they
are always embedded in a scenario that enables us to actually
see cause and effect in action. After all, John retching and Mary
screaming, even in a scene as minimal as those two sentences
sketch out, needs to be embedded in a setting that includes time,
place, actor, and intention. For us to link John heaving with Mary
shrieking, we have to assume they’re both in the same place, re-
acting in roughly the same time frame. We need to understand that
John is the main actor and that John’s holding his sides and gasp-
ing theatrically are the product of having eaten bad pizza rather
than, say, the product of his seeing just how effectively he can tor-
ment Mary. Without those assumptions about time, space, actors,
and reasons for acting, we cannot begin to see cause. In fact, we
cannot begin to understand how the sentences relate to one an-
other or perhaps even what they mean. But even if someone had
supplied instructions to us about time and space and intentions
in asides as we read – “Assume John’s actually got salmonella,
and Mary’s an incurable hypochondriac” – we’d fail to grasp what
the sentence meant unless we had some knowledge about food
poisoning or what retching looks like. In other words, we need
something cognitive psychologists call schemas, to comprehend
just about anything we read.

Schemas: no such thing as an immaculate perception

Both syntactic and inferential levels of processing rely on frame-

works for understanding, known to cognitive psychologists as
schemas. Schemas are the building blocks of comprehension, pat-
terns that enable us to make sense of what we see and hear, even
directions on how we should act in specific situations.11 Every
day, we use dozens of schemas to make our way in the world. For
example, whenever we enter a restaurant and sit down, we rely on
a schema that tells us that waiters will arrive at our tables bearing
menus of food available for order and await the nitty-gritty on how
we want our prime rib cooked. Restaurant schemas subtly inform
us that asking about blood work on a sick cat is not standard
24 THE NEW SCIENCE OF WRITING

restaurant behavior and would, in any case, net zero results, since
waiters tend to know the condition of the day’s specials but never
the shape Fluffy was in when you left home. Moreover, schemas
seamlessly underlie our everyday comprehension. Schemas tell us
that objects that appear smaller are, in fact, farther away from
us. In the same way, a schema also reassures us that the railway
tracks before our train really do remain parallel, despite our eyes
telling us that the tracks converge in the distance. Without sche-
mas, we’re unable to recognize what we see. In fact, as art histor-
ian E. H. Gombrich argued, we’re unable to see anything without
having a schema that enables us to perceive and process what our
eyes tell us.12
If this concept sounds far-fetched, consider the case of a man
who had something close to Gombrich’s waggish immaculate per-
ception, Virgil, the blind massage therapist in Oliver Sacks’ An
Anthropologist on Mars.13 At age fifty, Virgil regained his sight but
relied on his sense of touch to decipher the bewildering images
he encountered. Without a visual schema, Virgil remained what
Sacks calls “mentally blind” because he failed to understand the
visual inputs his eyes and brain sent him. Ironically, even once
Virgil regained his sight, he continued to use his fingers and hands
to orient himself and to make sense of the world.
But even mild mismatches with existing schemas can derail
our comprehension of what we think we already understand. In
one instance, a New Yorker recounted passing a security guard’s
radio on September 11, 2001, on his way into work: “I thought my
Spanish was passable, but that dude’s radio just said a plane flew
right through the World Trade Center.” He calmly went about his
work for the rest of the day, convinced his Spanish was at fault,
rather than the possibility that a plane had actually flown through
one of the towers at the World Trade Center, an event which con-
founded any conventional schema.
Like our brains, our schemas seize on default conditions – the
standard configurations, characteristics, and outcomes that char-
acterize an object or scenario. For all their helpfulness in enabling
us to perceive the world, understand what we see, and act on
it, schemas nevertheless remain fairly blunt instruments that re-
quire us to make inferences. In a sense, schemas act like inference-
generators that establish our expectations for how a situation or
 SCHEMAS 25

transaction will play out. Schemas lay out possibilities for actions
and interpretations, and we make inferences about the way sen-
tences will play out based on our understanding of the larger con-
text. But schemas also provide us with a feedback loop. We can
confirm the accuracy of the schema we’ve unconsciously chosen
to guide our interpretation of an article in the New England Jour-
nal of Medicine by checking the way it fits with the local details
we encounter at sentence level.
On the written page, cues for schemas abound. For example, a
Wall Street Journal article generally promises a focus on business
and finance. So we’d be jarred to instead stumble across, even on
the Op-Ed pages, a breathless exposé worthy of the US supermar-
ket tabloid The Sun on a woman giving birth to a black lamb after
accidentally being inseminated with ram semen. Similarly, the
titles of articles or books, their author names, and other details,
as we’ll see in Chapter 3, also cue schemas that help us narrow
considerably the range of inferences we need to make as we read.
Not surprisingly, individual articles or chapters themselves
need to cue schemas as early as possible to provide readers with a
clear blueprint of exactly where all those printed pages are head-
ed. Even before the days of online publishing and fine-toothed
search engines, abstracts served as an excellent look-ahead to
readers, providing thumbnail sketches of the territory covered by
the article. Ditto introductions to articles and books, which gener-
ally provide detailed road maps to readers, and even, as we’ll see
in Chapter 5, the beginnings of paragraphs. Ultimately, schemas
even intervene in our syntax-level processing, guiding us to latch
onto the first suitable candidate for a verb – as in the up-regulated
garden path sentence earlier – based on our experiences with mil-
lions of previous sentences. As a result, a familiar schema can
speed our comprehension at all three levels of the reading process:
lexical, syntax, and inference processing.
Similarly, departures from schemas can be the readerly equiva-
lent of a knuckle-ball, as I. A. Richards once discovered when
he required his Cambridge undergraduates, supposedly seasoned
scholars of literature, to interpret poems from which Richards had
removed both titles and authors’ names. The result: his students
floundered, thrashed around in their interpretations, and, gener-
ally, behaved like total incompetents. Disgusted, Richards used
26 THE NEW SCIENCE OF WRITING

this experiment as the pretext for developing the approach to un-

derstanding literature he introduced in Practical Criticism, a work
which gave rise to one of the major schools of literary criticism in
the twentieth century, the so-called New Criticism. But Richards,
had he known anything about the cognitive process of reading,
would have found a simpler explanation right under his nose. By
removing the titles and names of famous poets, Richards also re-
moved valuable schematic clues that his readers needed for syn-
tax- and inference-level processing, setting them the task of in-
terpreting complex pieces of writing without knowing what they
were reading.
Imagine trying to understand any complex piece of writing – a
feasibility study, a shareholders’ report, an article from Nature, a
blog entry – without having any knowledge of who wrote it or for
what purpose. You’d expend a good deal of energy casting around
for likely schemas to throw into the breach, then hastily revising
your guesses and trying on other schemas. Or imagine going to the
cinema to watch a film without knowing its title, director, or even
the type of film you’re watching. A character fleeing pursuit by
rattling down the stairs into a murky basement is one matter in a
slasher flick, another in a horror film, quite another in a thriller –
and an entirely different animal in a romantic comedy. No
immaculate perception, indeed.

Putting it all together: readability outcomes

Reading, as the phases in this complex process reveal, is bloody

hard work, even if we perform most of it unconsciously. Gener-
ally, when we read something that’s well written, the complexity
of the process and its cognitive demands are invisible. If, how-
ever, you’re reading something that’s poorly written – and this
list can include everything from a featured article in Nature to the
documentation for Microsoft Word, and the writings of too many
philosophers – you suddenly become aware of the cognitive over-
head. Reading becomes difficult. Recent research on the reading
process is particularly valuable because, from it, we can easily
separate good writing from the difficult and the truly execrable.
Well-written documents lend themselves to easy reading, swift
processing, and good memorability. In other words, good writing
 PUTTING IT ALL TOGETHER 27

is writing that follows the contours of the processes our minds use
to interpret and absorb it. In contrast, lousy writing bucks them –
at their writers’ and readers’ peril.
By using cognitive research on the reading process, we can
arrive at readability outcomes that enable us to make concrete
decisions in our writing that lead to papers, memos, briefs, in fact,
any kind of non-fiction document, that our audience reads quickly
with apparently little awareness of effort and recalls promptly and
easily. For starters, we know that reading always relies on pro-
jection and verification, a process that constantly looks forward.
As a result, clear writing requires word choices and sentence
structures that avoid our needing to backtrack when we read. So
clarity involves our choosing, as we’ll see in detail in Chapter 3,
concrete words over abstractions, words that have definite, rela-
tively fixed meanings. Ideally, we should choose words that make
clear their role in both the structure and action of the sentence.
The sentence about Horace and the lock, for example, would read
more clearly if it read head-lock or even hold, rather than simply
lock, since those other two terms make immediately apparent
exactly the sort of situation Horace finds himself in. Similarly, the
more concrete the terms and the more apparent their roles in the
sentence structure, the more efficiently readers can fix their sights
on an appropriate schema, and the easier the task of inference
building.
The chapters that follow leverage neuroscience research into
the workings of the reading brain to supply writers with five easy-
to-follow categories to make informed decisions about which
words to use, how to structure sentences, how to jam readers’
tendencies to skim overly dense paragraphs, and even where to
stash information they need to disclose but would rather their
readers forgot. In the chapters that follow, you’ll find simple, con-
crete principles that work across a myriad of writing tasks and situ-
ations, all summarized at the end of each chapter.

Takeaways for good writing: what you need to remember about

readability outcomes
For those of you who like to flash your science at doubting
colleagues or scoffing friends, you can summarize readability
outcomes pretty easily and still come across like a PhD:
28 THE NEW SCIENCE OF WRITING

• Clarity: Words are concrete and boast relatively fixed mean-

ings that are immediately obvious from their position in the
sentence structure.
• Efficiency: Sentences rely on common sentence structures,
introducing grammatical subjects (see Chapter 3) and verbs
close to the beginnings of sentences, to aid in immediate infer-
ence building. Sentences also provide clear, linguistic cues that
make causation and continuity between sentences apparent.
• Effectiveness: Paragraphs and entire documents front-load
overviews of contents at their beginnings, enabling readers to
fix on helpful schemas that speed the comprehension process.
Paragraphs and documents also provide as many cues as pos-
sible about the contents that follow, using concrete titles and
subheadings to provide further schematic cues.
CHAPTER 3

Choosing words and structuring

sentences
The first C: Clarity

In this chapter you will learn how to

• create sentences readers process most easily
• choose the best possible subjects
• ensure your verbs work for you
• make your sentences both lean and mean – shorter and
clearer.

According to neuroscience research, we’re hard-wired to register

cause and effect. In the 1940s, researchers showed volunteers ani-
mated films of circles, squares, and rectangles shuffling around
the screen. The film was hardly the equivalent of an animated
Gone with the Wind, but, when researchers asked their subjects to
describe what they had seen, the men and women related charged
scenarios involving not only causation but intention and even, for
multiple participants, conflict between two men pursuing the same
woman.1 In a later version of the same experiment, study par-
ticipants repeatedly reported that the circle had been chasing the
square – or the rectangle was pursuing the circle. More surprising,
the subjects seemed inordinately hung up on cause and effect –
what they described was invariably a full-fledged story.2 Later,
studies with infants as young as six months similarly revealed the
same fixation with cause and effect.3 Obviously, an inborn ability
to perceive cause and effect would be rather handy, especially to
early hunter–gatherers out scrounging around for dinner amid lions,
tigers, or bears. That same ability today stops us from lovingly lay-
ing our hands on a stove burner or fondly running a finger along
the sharp edge of a whacking great knife.
30 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

This same innate tendency can also explain our attraction to

stories – even our hard news is delivered in the form of stories –
and our ability to recall small, throwaway details when they’re
embedded in a story. Similarly, readers devour and recall easily
sentences that include cause and effect. Make a sentence into a
miniature narrative, and your readers will easily remember what
you’ve written. In addition, they’ll also mentally process your
writing more quickly and with little awareness of the hard work
that goes into reading.
In fact, reading itself is the equivalent of hard labor. Even
someone caught up in the most formulaic scene of the schlock-
iest Harlequin Romance is busily multitasking. As we read, we
ferret out the meanings of words and create connections between
sentences, then compare what we read to our own experience of
the world and of other novels. We also squirrel away juicy details
for later recall. And we relate the meaning of the immediate detail
to the overall story – say, the significance of that near-grope in a
packed elevator to the plight of a hapless, single fledgling attorney
with a serious thing for her boss. All this effort also explains why
reading the instructions for programming a universal remote con-
trol for your flat-screen TV can feel like the equivalent of digging
a grave. Or why you feel as if you’re in dire need of a brain trans-
plant if, God forbid, you need to actually read any of those helpful
instructional manuals that rely on triple negatives and sentences
you’d need to recruit a linguist to interpret. And you’d also need
to resign yourself to the linguist throwing up her hands in defeat.
Clarity invites us to read – and keeps us happily reading. Clar-
ity always distinguishes good writing from the jungly thickets of
dead sentences that you couldn’t recall five minutes after reading
them, if someone held a loaded gun to your head. I once took
only a single, skinny book by an academic hottie – who shall
remain nameless – for a flight from London to New York, con-
vinced that surely the lack of any other form of reading matter at
39,000 feet would force me to digest this little academic treatise
on modern life. In mid-flight, I staggered to the john, then re-
turned to my book, which I hadn’t bookmarked. For the next 90
minutes, I reread the sixty pages I’d just plowed through, without
the slightest jolt of recognition … until I stumbled across my own
scrawl in the margins. Ten years before this moment, I would’ve
 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES 31

blamed my education (Michigan is a state school, after all!), booze

(I shouldn’t have had that fourth glass of cheap champagne!), or
my lack of attention or sleep or brain-power. But now I knew bet-
ter. My blood-alcohol was zero, and years of research had taught
me that my total lack of comprehension – let alone bloody recall
– was the author’s fault, not mine. Nearly every sentence was pas-
sive, stuffed with abstractions, chocked with memorable subjects
like it. Worse, virtually every sentence was so long that, by the
time I reached its end, I wondered what the hell the beginning of
the sentence had been about. In short, the academic hottie had
committed every imaginable sin against clarity.

An extremely short history of English

English is a mongrel, a Johnny-come-lately to world languages

descended from Old Norse, Anglo-Saxon, Latin, and French,
a result of the 1066 Norman Conquest of what later became
England. Typically, English tends to happily mug whatever
scraps of languages it happens across and, as one wag put it,
rummage around in the pockets of its victim for loose bits of
grammar. Like a hustler sporting an array of diamond Rolexes
and stainless Tag Heuers all massed together on one wrist,
English also traditionally makes little attempt to integrate the
words and phrases from the languages it absorbs into language
that obeys the rules of standard written English. In France, you
get words like le blue-jean; in English, you get schadenfreude.
As a result, even to this day, English’s Latin and French
vocabulary is clearly distinct from the words Anglo-Saxons
used before 1066. Words that come to English from Latin and
French tend to begin with prefixes like de-, dis-, in-, im-, non-,
and un- and to end with suffixes like –ion, -ate, -ive, and -ity.
Moreover, both Latin- and French-based terms are more ab-
stract and longer than words with Anglo-Saxon counterparts.
In contrast, nearly all Anglo-Saxon words refer to concrete
objects and have a single or few syllables. And, perhaps not
surprisingly, all the four-letter words US broadcast networks
still charmingly bleep out, lest they pollute American family
values, come from good old Anglo-Saxon.
32 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

For decades, expert writers have been exhorting readers to

stick to that Anglo-Saxon vocabulary. In “Politics and the Eng-
lish Language,” George Orwell indulged himself in a lengthy har-
angue on avoiding the corrupting influence of Frenchified Eng-
lish. Orwell argued that, for example, the public found easier
to swallow elimination of unreliable elements than its concrete,
Anglo-Saxon counterpart: “People are imprisoned for years with-
out trial, or shot in the back of the neck or sent to die of scurvy
in Arctic lumber camps.” Unfortunately, Orwell phrased much of
his jeremiad in the same abstract, polysyllabic Latinate language
he was urging readers to avoid at all costs, in addition to falling
back on the passive construction he advised writers to shun in
favor of active construction. But Orwell had a point. The more
concrete your language, the more easily readers can picture,
grasp, and recall your meaning. In addition, concrete language is
infinitely better at telling mini-stories than Latin-heavy, abstract
words can ever be. And the more complex your topic, the greater
your need to use concrete language to describe it.

TERMS YOU’LL NEED TO KNOW

While diagramming sentences belongs among the practices

banned by the Geneva Conventions, if you want to write any-
thing well, you actually do need to identify the function of
words in a sentence.
A noun refers to a person, place, thing, or idea. Nouns in-
clude words like thoughts, dogs, skies, actors, and Mormons.
Pronouns, on the other hand, stand in for nouns and always
refer to another noun in a sentence or in the preceding sen-
tence. Pronouns include I, you, me, they, us, we, him, and it.
Grammatical subjects are the main noun or pronoun in a
clause or sentence, always paired with the main verb.
A verb portrays action or describes a state of being. Action
verbs include launched, imitated, and gawped, while non-action
verbs – known as intransitive verbs – include all forms of to be.
Adverbs tend to end in –ly and usually modify verbs, although
they can sometimes modify another adverb. Examples of adverbs
include quickly, reluctantly, crudely, deviously, and winningly.
 READING IS A THREE-STEP DANCE 33

Prepositions are words that can fit in the formulaic sentence

teachers bludgeoned into students in grade school: The mouse
ran _______ the box, plus the word during. Prepositions in-
clude in, on top of, through, without, around, beneath, over,
and throughout.
Not surprisingly, prepositions make up the beginning of
every prepositional phrase, little clumps of words that indi-
cate relations between things in space or time. Each prepos-
itional phrase includes a preposition, followed by a noun, often
paired with a couple of descriptive words that modify the noun,
known as adjectives.
Think of nominalizations as castrated verbs, sapped of their
livelihood by serving as nouns. Every nominalization once had
its origins in a verb before it acquired the Latin-sounding end-
ings that transformed it into a noun. Nominalizations include
evaluation, dispensation, utilization, and creation. Generally
speaking, if you can turn the noun into a verb – and even solid
old Anglo-Saxon words like life and freedom can be turned into
verbs, if you think about it – you’re looking at a nominalization.
Together, a noun and a verb make up a clause. Major clauses
are complete sentences that can stand by themselves and still
make sense, even in isolation from the rest of the sentence. A
minor clause, however, will sound incomplete or like a sen-
tence fragment if you subtract it from the rest of the sentence.
In the sentence, Do it because I said so! Do it is the major
clause, since it makes sense even when isolated from the rest
of the sentence. On the other hand, because I said so sounds
incomplete when hacked off from the complete sentence, mak-
ing it a minor clause.
For a complete guide to everything you never learned about
grammar, in addition to punctuation, see the Supplement on
page 164.

Reading is a three-step dance

Reading is actually a heavy slog, as we saw in Chapter 2, the

reason why the passenger in the seat next to you on your long-
haul flight is more likely to be working away at Sudoku or star-
ing at a film rather than reading even that cheesy mass-market
34 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

paperback destined to end up in the seat-back pocket when you

all disembark. Assisted by eye-tracking devices, EEG, and fMRI
technologies, researchers identified three distinct levels of cogni-
tive processing involved in the act of reading even a paragraph in
People or Hello! magazines, not exactly the equivalent of reading
Finnegans Wake: lexical, syntactic, and inferential, all of which
are relevant to clarity.
The first level, lexical processing, begins when readers recog-
nize individual words and assign them a fixed meaning, based on
familiarity with the word from prior encounters. Using saccades,
scientists have tracked the length of time our eyes linger on indi-
vidual words and linked these saccades to EEG activity, measuring
the amount of brain activity involved in identifying a word. But in
English the speed of our eye movements depends entirely on the
context surrounding a word. In English, neither the position of the
word in the sentence nor the way the word ends clearly tells us
whether hope is a noun (Please do not flush your unrealized hopes
and dreams down the toilet, to quote the instructions on the toilet
on a Virgin train in the UK) or a verb and a noun (I hope I didn’t
flush my unrealized hopes and dreams down the toilet, as one pas-
senger told me, on emerging from the same loo). And, if you’re
reading the work of a writer untutored in the correct use of the
hyphen – which includes a substantial chunk of the US popula-
tion – “hope” might even be an adjective, as in “hope inducing,”
rather than “hope-inducing,” where the hyphen helpfully tells the
reader that at least the verb and noun forms of the word are out of
the running. Even the word writing itself can act as a noun, verb,
or even an adjective, as in I’m writing [verb] this writing [noun]
on a writing [adjective] tablet, a sentence demanding readers dis-
ambiguate the three different parts of speech occupied by the same
word, even with correct punctuation.
As a result, the interdependence of the meanings of words and
their positions in a sentence inextricably links the first and second
levels of processing. Readers make sense of words during the sec-
ond, or syntactic, phase of processing the same way they recog-
nize words on the page. We use surrounding words to anticipate
how the sentence will structurally play out.4
Generally, these first two levels of processing occur together,
with readers relying on the third level, inferential processing, only
 READING IS A THREE-STEP DANCE 35

when the sentence’s contents fail to make sense. As we saw in

Chapter 2, we make inferences based on schemas we’ve acquired
to make sense of the world. When sentences correspond to famil-
iar schemas, we read their contents more rapidly and recall them
more easily than when we encounter an altogether unfamiliar
schema.5 Ultimately, we unconsciously rely on all three levels of
cognitive processing as we read: lexical, syntactic, and inferential.
And, given the roles cause and effect play in our perceptions of
the world – as we saw in Chapter 2 – when you write sentences
that embody cause and effect, readers read them more swiftly and
easily than sentences that fail to explicitly link actors and actions
causally. This complex, three-step dance explains the first prin-
ciple to writing clearly, why readers will perceive your writing as
clear and easy to read if you rely on active, rather than passive,
construction, along with the other three clarity principles we’ll
explore in detail below.
Try to visualize the concrete details in the following examples:
Example A:
Panting from the pursuit, Tucker ducked around a corner
and huddled under a honeysuckle bush, while his would-be
captors raced past.
Example B:
The would-be captors were evaded by Tucker, who ducked
around a corner and huddled under a bunch of vegetative
assemblages.

While you might find both sentences to be fairly similar, researchers

have repeatedly proved that you’d be more likely to recall Example
A than Example B. For starters, have you ever tried to picture a veg-
etative assemblage? Multi-syllable words, a legacy of the good old
Norman Conquest, entered English either directly from Latin itself
or in a watered-down Latinate form via French. Crack open your
Webster’s, put your finger on any polysyllabic word, and you’ll
discover a Latinate term that is generally an abstraction. Vegetative
assemblages will send your average reader scrabbling for a diction-
ary. And that nice little scrap of terminology will also force your
readers’ brains into twists and turns worthy of a contortionist. Is a
vegetative assemblage a honeysuckle bush? What about multiple
36 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

honeysuckle bushes? Or do honeysuckle grouped together with

guavas, trumpet vines, and juniper count as a bona fide set of
vegetative assemblages? Incidentally, if you’re convinced this term
is so patently ridiculous that no one would ever be caught dead
using it, think again. Several years ago, a group of engineering stu-
dents submitted a proposal for the design of a campus conference
center, a plan heavily dotted with what the proposal identified as
vegetative assemblages. The students meant shrubs.

Clarity Principle #1:

Prefer active to passive construction

If you glance again at Example A, you’ll notice that Tucker is

performing the actions in the sentence in chronological order.
First he runs, then he hides while his pursuers lope past. In this
sentence, action originates with an actor at the beginning of
the sentence, Tucker. The sentence then dramatizes the action
through the verbs ducked and huddled, with the result that
his would-be captors run past him, arriving at the end of the
sentence. Note how this sentence neatly conforms to readers’
assumptions that the order in which events occur linguistically in
a sentence mirrors the order in which they unfold in the world.
But in Example B, the sentence distorts the order in which the
actions occurred. The action flows back-asswards, so to speak,
with the outcome of Tucker’s action – evasion – conveyed passive-
ly. In this version, even though Tucker clearly initiates the main
action in the sentence, in grammatical terms, he’s relegated to the
equivalent of walk-on Second Spear Carrier.
If you actually learned a smattering of English grammar – a
process that, in America, usually means you were taught by (a)
nuns, (b) teachers born before 1934, or (c) parents who were
high-school English teachers – you’d know that Tucker in Example
B is actually the object of a preposition, one of those annoying lit-
tle throwaway bits of grammar that clutter up sentences. Preposi-
tions and their full-blown counterparts, prepositional phrases, are
grammatically insignificant. In fact, if the nuns insisted you parse
sentences into parts of sentence, as well as parts of speech – a pro-
cess that still strikes terror in the hearts of even arrogant English
 ACTIVE VS. PASSIVE CONSTRUCTION [CLARITY PRINCIPLE #1] 37

PhDs with a bit of dust on their credentials – you’d discard entire

prepositional phrases, crossing them out to distinguish them from
the meatier parts of the sentence.
In Example B, the verb is the ever-dynamic was evaded. Note
how the presence of forms of to be (is, was, was being, was seen
to be) shoves the action into the wings. You’re no longer witness-
ing an event unfolding. Instead, some unseen being is narrating
the event ex post facto, hardly edge-of-the-seat stuff, let alone the
things stories are made of. But, while non-action verbs sap the vigor
from sentences, they don’t always signify a passive sentence or
what’s known as passive construction. Non-action verbs are merely
a symptom of a sentence that might be passively constructed. You
can have a non-action verb and an actively constructed sentence,
but not an action verb and a passively constructed sentence. For
instance, Tucker was furious at his pursuers relies on a form of
to be but is nevertheless an actively constructed sentence. Why?
Tucker, the grammatical subject, clearly is also experiencing fury.
In contrast, Tucker, an overweight Jack Russell, was pursued by a
pack of feral cats, is passively constructed, despite action seeming-
ly promised by the verb pursued. If you decouple the sentence’s
meaning from its structure, Tucker isn’t an actor, despite his being
the grammatical subject. Instead, he is the object of pursuit by
feral cats. As in all passively constructed sentences, the sentence
fails to mirror the unfolding chronology of events.

To root out passive construction, try using zombies

Confused? Try this acid test. Insert the phrase by zombies into
the sentence after the verb: The would-be captors were evaded
by zombies. If the sentence makes sense with zombies in it –
with zombies representing the implied actors – then you’re
looking at a passively constructed sentence. In contrast, Pant-
ing from the pursuit, Tucker ducked around a corner by zombies
makes no sense, not merely because we’ve yet to encounter a
zombie outside a film or television. The zombie, vampire, or
superhero acid test, I’ve discovered, works nearly instantly
with everyone, including students who’ve never had to think
about clauses and grammatical subjects.
38 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

Why is active construction such a big deal?

Our brains process active sentences more efficiently than passive
sentences for four reasons. First, most psychologists agree that
cause and effect – exemplified in the sequences of images that
made babies’ heart rates and blood pressure jump – power hu-
man perception. In study after study, researchers have found that
readers process sentences containing causal relationships more
quickly than they do other kinds of sentences.6 Second, causal
order involves active construction, which reproduces the chrono-
logical order of events. If I say I lobbed the cookies at his head,
the sentence preserves the order of events in which I chucked
a batch of chocolate chip cookies at my victim’s head. But if I
say The cookies were lobbed at his head by me – virtually the
same sentence, different word order – the result of the tossing
precedes me, the actor who actually tossed those lumps of baked
dough. Moreover, readers implicitly expect the order of items in
sentences to replicate the order in which events unfolded, as we
saw earlier in this chapter. And this inverted order requires read-
ers to understand the action by reassembling the pieces in order.
Third, readers’ brains take longer to process passive construction
but tend to speed through active. Fourth, readers misinterpreted
implausible sentences one-quarter of the time when sentences
relied on passive construction like The dog was bitten by the man.
Readers committed this error presumably not because they could
visualize a scenario in which a neighbor assaulted, say, a corgi
minding its own business, but because the passive construction
obscured agency. Or readers made this error because they re-
quired substantially more time to make sense of the passive, im-
plausible sentence than either the active, implausible sentences
or the passive, plausible sentences. And, finally, active construc-
tion builds on the default order of sentences in English – subject–
verb–object.7
In fact, in nearly every language in the world, subjects normal-
ly precede objects and nearly always come before verbs. Psycholo-
gist Steven Pinker notes in his popular book The Language Instinct:
How the Mind Creates Language that most languages have either
a subject–verb–object (SVO) order or a subject–object–verb (SOV)
order, both natural orders for depicting cause and effect. In fact,
the majority of the world’s languages have either an SVO or SOV
 ACTIVE VS. PASSIVE CONSTRUCTION [CLARITY PRINCIPLE #1] 39

order.8 But English’s subject–verb–object order reflects causation

so strongly that it may hijack newly developing languages. In the
1970s, when linguists in South America discovered people in Guy-
ana speaking a new, weird kind of Dutch, they discovered the
speakers of the language, now known as Berbice Dutch, formed
sentences with subject–verb–object order. Yet the languages Ber-
bice Dutch evolved from – Dutch and an African language called
Ijo – both place verbs at the ends of their sentences. In spite of
its origins, Berbice Dutch evolved from two subject–object–verb
languages into a subject–verb–object language order most likely
because this order more closely replicates the way we perceive
events in the world.9 First we encounter the actor, then the action,
then the result. Just like life – at least most of the time.

Exceptions to the rule

“Aha,” you think, “so much for giving writers some hard-and-fast
rules to cling to.” Most of us not-so-secretly long for rules, especial-
ly if we’ve put off the dreaded act of writing a paper or grant until
the deadline looms mere hours away. Rules, after all, constrain
the myriad decisions that plague us when we write, making them
particularly handy during the small hours when you feel as though
you need lashings of caffeine and toothpicks to prop up your eye-
lids. However, the exceptions to the rule on passive construction
are mercifully few, based on studies of readers’ brains and recall
of sentences.
1. Use passive construction when entire sections of
documents, like the methods sections of studies, employ
it, as our brains can process substantive chunks of
continuous passive construction nearly as efficiently as
we can actively constructed sentences.10 In addition,
passive construction in this setting conveys the gist
of a methodology more efficiently than continually
reiterating, “We excluded … We tested … We evaluated …”
and a barrage of sentences all beginning with the same
grammatical subject, I or we.
2. Use passive construction when the actor performing the
action is unknown or less important than the outcome
you’re describing.
40 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

Why these two particular exceptions? Readers read and interpret

passively constructed sentences efficiently when the emphasis
shifts to the outcome of an action, a condition common to both of
these two exceptions. For methods sections in reports of experi-
ments, the focus is uniformly on the outcome, not on the actors.
Similarly, in the second condition, the outcome assumes greater
importance than the actor – or the actor is irrelevant or even un-
known.11 In making memorable speeches, politicians make exem-
plary use of this exception, Franklin Delano Roosevelt did in his
famous speech following the Japanese attack on Pearl Harbor on
December 7, 1941.12
3. Use passive construction when you wish to avoid assigning
agency to anyone in particular, to avoid legal liability.
If you’re an American, you will likely have thought of this ex-
ception already. In the United States, I’ve heard a seven-year-old
threaten a school administrator with a lawsuit for pulling him off
the unfortunate child he was busily pummeling during recess. I’ve
also encountered an undergraduate who used his litigious father –
a lawyer, of course – to sue a university instructor because she
failed to point out the student left his $250 textbook behind in
a classroom heaving with other students. In contrast, in the UK,
when I once threatened to sue the company that left my house
without a source of heat or hot water for three weeks in the chilly,
damp dead of winter, I received only icy silence in response. In
countries outside the United States, most writers are less mindful
of litigious agency and methods for avoiding liability by falling
back on passive construction.
If you’re a solicitor, barrister, or attorney, however, you might
want to reconsider this particular exemption, as courts have some-
times found in favor of plaintiffs because passive construction ob-
scured who did what.13

Corollary: Beginning a sentence with There is or There are

is ALWAYS a bad idea
Unless you’re angling for a Booker, Pulitzer, or Nobel in Litera-
ture, writing poetry, song lyrics, or graffiti, the same handy There
is or Adverb + [ANY FORM OF TO BE] is thoroughly lethal to read-
ers’ comprehension. While ubiquitous in speech and writing, this
 ACTIVE VS. PASSIVE CONSTRUCTION [CLARITY PRINCIPLE #1] 41

construction belongs strictly in speech, which is notoriously rife

with inefficiencies and redundancies. In speech, we rely on There is
and other less-than-efficient ways of phrasing things to ensure our
audience gets the point of the sentences we hear, despite weird
accents, rapid-fire speech, and wandering attention spans. And
There is comes in handy for stretching out rhythmic lines in song
lyrics to match the music. Take, for example, that timeless Ani-
mals’ “House of the Rising Sun” lyric which begins, “There is a
house in New Orleans, they call the Rising Sun,” that stems from
an old oral tradition of blues songs. The song, probably created
in Kentucky mining camps, eventually found its way to the de-
termined musicologist Alan Lomax, who, armed with a primitive
tape recorder, first captured the lyric from a backwoods teenager.
Woody Guthrie later heard it and recorded his rendition before The
Animals created the version Baby Boomers tend to recall fondly.14
There is-type constructions most likely come to us from
French, courtesy of the Norman Conquest, where literate Normans
swarmed over the illiterate Anglo-Saxons in what is now England.
Unsurprisingly, Latin and French trumped Anglo-Saxon, especial-
ly since writing enabled the Normans to claim written titles to
English property formerly held by the Anglo-Saxons, who need-
ed trusty men who swore oaths to attest to their ownership.15 In
French, il y a is a useful construction for representing shifts in
time or setting. In oral English, that old Norman phrasing persists
stubbornly. Artifacts like there is peskily hang around, long after
their utility has vanished, because recording language in writing
slows its evolution. Bring in the printing press and mass-produced
copies of a standardized language, and you petrify it.16
Why is there is and its relatives so detrimental to clarity? When
you begin a sentence with There is, you turn the natural order of
English sentences upside down. Since there is a lowly adverb – the
sort of word nuns would’ve insisted you cross out on your home-
work – there hogs the spot where your readers would normally
expect to see the all-important subject of the sentence. Worse, the
verb is shoves its way in front of the actual subject of the sentence,
which really jams your readers’ radar. Readers begin making pre-
dictions about the way a sentence will play out beginning with
its first words, enabling them to make sense of whether rebel is
a someone upsetting the apple cart or an action that upsets the
42 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

apple cart. Those predictions sharpen when we identify the sen-

tence’s grammatical subject and end after we identify the main
verb. When you insert an adverb in the slot where readers un-
consciously expect to spot the subject and you place the verb be-
fore the grammatical subject, you flummox readers’ expectations.
You also create atypical sentence structure, since English’s default
order is subject–verb–object and only verb–subject–object in the
rare, inverted question form. Atypical sentence structure slows
down readers’ processing of the sentence’s components and their
meaning.17 In the meantime, you also lumber your readers with
passive construction – as we’ve just seen, not a brilliant strategy –
and a non-action verb, which we’ll explore in detail shortly.
In short, there is nothing redeeming in there is constructions
on the page, unless you’re aiming for a conversational style, aping
Hemingway, or tweeting about something. Or aiming to piss read-
ers off by committing the selfsame sin you’ve just exhorted them
to avoid on pain of death.
Once, when I was teaching a writing course for clinicians in a
college of medicine, one of my students announced proudly that
she’d eliminated every last instance of There is and There are from
her article on pediatric diabetes. My eye fell on the first page of the
article, first paragraph: There are. By the time the rest of the class
finished prospecting for those pesky little constructions on her
first page alone, we’d counted eight. And yet the endocrinologist
had proofread her article diligently. Because we’re used to seeing
these buggers littering up sentences, most of us have difficulty
spotting them. So, before you finish any piece of writing, use the
“Find” command in your writing software and type in There is or
There are. When you happen across one of these constructions,
revamp the sentence. Instead of saying There are three ways we
can think of this, write, instead, We can think of this dilemma in
three ways. As we’ll see in the next chapter, this reshuffled order
will also help focus your readers’ attention on the most important
content in the sentence.

Clarity Principle #2:

Make your verbs portray action whenever possible

Reading through sentence after sentence of non-action verbs

is about as interesting as curling up with the instructions for
 ACTION VERBS [CLARITY PRINCIPLE #2] 43

programming a universal remote to operate your TV, the kind that

require you to try fifty different codes to pair the remote with
your flat-screen, after your dog chewed the original that handily
came with the TV. Reading non-action verbs is the sort of exer-
cise that only my ex-father-in-law could appreciate, a man who
thoughtfully videotaped every page of his Australian tax return
and generously sent the tape, complete with monotone voiceover
narration, to his grandchildren as his first communication with
them in years. Yet, page through just about any document that
bores you stiff – academic journals, instructions for connecting
your TiVo, pages of legalese from your divorce proceedings – and
you’ll stumble across the mother lode of non-action verbs. De-
prive lawyers and professors and researchers of non-action verbs,
and you’d condemn these hordes to decades of Hoover Dam-sized
writers’ block.
Non-action verbs, remember, are not of the same magnitude
of crime as passive construction, not exactly a lethal crime against
writing. Nevertheless, non-action verbs can be nearly as deadly
to the clarity – not to mention the liveliness – of any sentence as
passive construction.
From a psychological perspective, action verbs considerably
simplify the act of reading by performing two valuable functions.
First, they clarify relationships within the sentence in a who-is-
doing-what-to-whom sort of way, clearly pinning causation to
a verb and thus clarifying which noun is performing the action
and which noun, receiving it. For those inclined to slog through
fMRI or EEG studies of the reading brain, this clarification speeds
reading and also proves less taxing on readers’ brains.18 Second,
action verbs immediately help readers grapple with the lexical de-
mands reading places on our brains. By clearly fixing the action
and identifying actor and outcome, a sentence using action verbs
eases readers’ identification of the parts of speech each word rep-
resents. This fixing of action remedies the challenges of reading
English, a language in which neither the order of words nor their
form dictates their meaning absolutely. Instead, in English, con-
text, syntax, and plausibility constrain the array of meanings we
could attribute to each word. As a result, the action verb alone
helps readers identify causation and, as we’ve seen earlier, more
easily determine the meaning of each word. When you use action
verbs, you handily tackle the demands of both lexical (word) and
44 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

syntactic (the part of speech or role played by each word) process-

ing simultaneously.19
Moreover, non-action verbs have other notable drawbacks. For
starters, non-action verbs act as mere place markers – bookmarks
rather than sentences telling the story. Use a non-action verb like
is, was, appears, constitutes, represents, or has been, and your sen-
tence no longer tells a story. In fact, your sentence will probably
read about as dynamically and memorably as instructions for whip-
ping up tomato soup from a can of Campbell’s. Your sentences,
propped up by non-action verbs, simply convey static states, which
can prove about as interesting as watching paint dry. These non-
action verbs also result in readers spending more time pondering
the relationships between nouns in your sentences, partly demon-
strated in a study of transitive (which we’ve called action) versus
intransitive (which we’ve called non-action) verbs.20 Non-action
verbs leave us wondering who did what to whom. Researchers
explored the challenges to readers’ brains in identifying lexical and
syntactic meaning via neuroimaging studies where subjects read
sentences with explicit versus implicit causation.21 If you fall back
on ye olde non-action verbes, your readers are going to spend more
time wrestling with the issue of who did what to whom than if
you use action verbs, which clearly attribute agency (the actor) by
embodying action.
Second, non-action verbs are also dull and unmemorable.
In contrast, sneezed, yawned, barfed, and chundered might
be intransitive, merely reporting a state of being, rather than
portraying an action with an outcome. Nevertheless, shrugged,
chuckled, and sank remain colorful and action-oriented verbs,
whereas non-action verbs fail to capture action concretely. If you
want to lull your readers to sleep, salt your sentences liberally
with verbs like has been and watch your readers’ eyelids start
drooping. Non-action verbs in a string of sentences read mon-
otonously and also tend to create abstract sentences, making
them difficult for readers to remember. In a 1976 study of 200
action verbs, researchers discovered that the more concrete the
verb, the easier readers’ comprehension and more precise their
recall of the sentences.22 Concrete verbs help clarify the mean-
ing of even abstract nouns within the same sentence, while also
helping readers fix the meaning of words and the roles they play
 ACTION VERBS [CLARITY PRINCIPLE #2] 45

within the sentence, the old dance between lexical and syntactic
processing that reading inevitably entails at the sentence level.
Moreover, even when verbs seem to represent action, they can
fail to embody the sentence’s primary action. When writers use
“light” verbs, they convey action only via a direct or indirect
object. In English, the verbs made, give, and take become light
verbs when they appear in made a deal, give me a break, or take
a hike.23
Finally, and most importantly, non-action verbs are the equiva-
lent of couch potatoes who call for someone to hand them the
remote if it ends up being more than two feet away. They simply
occupy space. They never carry their weight. And they need other
forces to prop them up and keep everything together. Because non-
action verbs merely indicate a state of being, rather than drama-
tizing action, writers must insert other words to carry the load
the verbs should be shouldering. As a result, writers must fes-
toon sentences with non-action verbs with more words: adverbs
to pump some color into the action and prepositions to indicate
relationships between things in space and time. In fact, non-action
verbs necessitate the use of prepositions so consistently that, if
you spot a short sentence sporting more than three or four prepos-
itional phrases, you can be certain you’re dealing with either a
non-action verb or passive construction – or both.
Scads of prepositional phrases are nearly always a symptom of
an inefficient and unclear sentence. When you’re writing your sen-
tences, think about making your verbs pull their weight. Use action
verbs like powered, catered, provided, annoyed, acted, protested,
convinced. Your sentences will not only read more concretely and
be more interesting to your readers, they’ll also be much shorter
and more efficient than sentences that rely on non-action verbs.
Compare, for instance, Examples A and B below:

Example A:
Chess Aviation has an excellent concept. The idea of
catering any service to the extremely wealthy has always
been a viable idea. Airline service is necessary for most
high-level executives of corporations, businessmen, and
wealthy individuals. The airline service that is provided by
commercial carriers is extremely poor.
46 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

Example B:
The management team has built Chess Aviation around an
excellent concept: catering to the extremely wealthy. Many
high-net-worth individuals fly frequently, including wealthy
businessmen and high-level executives of corporations.
However, commercial air carriers tend to provide poor
service, falling far short of what affluent people expect.

While the topic is hardly edge-of-the-seat stuff – particularly

given the nosedive the aviation industry suffered in the wake
of 9/11 – Example A, written in a business plan by an MBA
student, is vague and relies entirely on the verbs is and has.
In Example A, catering any service to the extremely wealthy
remains an idea, an abstraction. But, when readers encounter
Example B, they won’t have any difficulty in picturing execu-
tives boarding planes and flying. Example B is both more con-
crete and clearer. This example is also more interesting to read
and easier to recall than Example A, fixing actor and action
easily in readers’ brains, even though the two examples convey
nearly identical content.

Corollary: Avoid nominalizations!

Aside from being an obnoxious, difficult-to-pronounce word that
can make you want to run for your dictionaries – or run, period –
nominalizations are the villains of sentences because they steal
the action from the verb. You might think of nominalizations as
neutered verbs that convert concrete action into an abstract noun:
utilization (from use), development (from develop), communica-
tion (from communicate), punishment (from punish). As we’ve
already seen, the more concrete and causal your sentences, the
more easily readers move through lexical and syntactic levels of
reading. Moreover, as scholars have noted, nominalizations – the
word itself is a hideous transformation of the already-Latinate
verb to nominalize – obscure action, agency, and even meaning
in sentences.24 Take action away from the verb, and, most of the
time, you can kiss goodbye your chances of writing a sentence
that anyone other than a doting mother would want to read. More-
over, if you’re a basic scientist writing about heatshock protein 70,
for God’s sake, avoid mirroring the complexity of your topic with
 ACTION VERBS [CLARITY PRINCIPLE #2] 47

linguistic complexity. And nominalizations tend to be complex,

abstract, and difficult to picture, with the notable exceptions of
words like life, hope, and love, all of which are nominalizations of
the verbs live, hope, and love.
Unfortunately, nominalizations are sneaky, flying stealthily
beneath our radar because they crop up in so much of what we
read. Just because other writers routinely get away with using
nominalizations, however, doesn’t mean that using them is a good
idea. If you’ve written a sentence that seems sluggish or even dull,
check for nominalizations that rob your verbs of get-up-and-go.
On the other hand, if you stumble across one of those rare nomin-
alizations – like transformation in the examples below – that’s
simply lying innocuously in your sentence, letting the verb do its
thing, leave it alone. That sort of nominalization isn’t harming
anything.25

Spot the nominalization

How do you spot a nominalization? Simply ask yourself if the

noun can also be changed to a verb. Sometimes, as with the
nominalization liberty, the words themselves look pretty innoc-
uous – no Frenchified polysyllables there. But not all nomin-
alizations have multiple syllables or end in the telltale -ion. So
ask yourself if you can squeeze a verb from the noun: liberate.
If the answer is “yes,” you’re staring squarely at a nominaliza-
tion. Now look at the verb closest to the nominalization. Is the
verb also just lying there, failing to depict any action whatso-
ever? If the nominalization occurs near verbs like has been or
any form of the verb to be, you need to get rid of the nominal-
ization.

How do you get rid of nominalizations?

Fortunately, getting rid of nominalizations is less difficult than
pronouncing the word itself. Getting rid of nominalizations,
however, requires you to also think about active construction
and action verbs. And you were probably hoping for a quick fix.
Unfortunately, you also must observe our next principle, which
48 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

involves choosing the right subject for your sentence. First,

though, you have to spot the nominalization, as in the following
sentence:
There was first a review of the transformation of the market
for mouthwash.

Now, if you’ve been paying any attention to this chapter – as op-

posed to reading it while mouthing the lyrics to a song on your
playlist or half-watching TV, you should’ve spotted three clarity
issues that virtually scream out from the page. Give up? First, the
sentence begins with There was, a violation of what should be
the First Commandment of Writing: Thou Shalt Not Use There
is, There was, or There are in any sentence. Second, the short
sentence also boasts no fewer than three prepositional phrases, a
dead giveaway of a sentence in dire need of a clarity tourniquet.
And, third, the sentence also contains the word review. Review,
in case you’d merely skimmed over this seemingly innocuous
word, is one of those dreaded nominalizations – in this instance,
one that’s sucking the life out of the verb. While review is a
noun, you can also turn it into a verb, review, as in we reviewed
the market.
Is review just an innocent bystander in the sentence? After all,
the word doesn’t even end in one of those nasty-sounding -ion
endings, right? Okay, proceed to Step 2 and look at the verb. Is it
depicting action?
In fact, the verb in the example is non-active, and the sentence
omits any explicit actor, let alone including one helpfully in the
grammatical subject. Flummoxed? Try the by zombies test: There
was first a review of the transformation of the market for mouth-
wash by zombies. Now you know you really have to overhaul the
sentence. This patient needs a tourniquet and sutures, so stop
thinking you can simply put a bandage on the sucker and sidle off.
For starters, you have to ditch There was and the passive construc-
tion. To achieve this change, however, you need to cast around for
a likely suspect for your subject. A good candidate in this instance
would be something like department or that other likely suspect,
researchers. Both are good candidates because both words rep-
resent actors – groups of people capable of making something
 ACTION VERBS [CLARITY PRINCIPLE #2] 49

happen. Now change the nominalization review into your verb,

which makes both the sentence and verb active:
First, the department reviewed the transformation of the
market for mouthwash.

Personally, I’d be fairly happy with this version. But the anally re-
tentive among you might have spotted yet another nominalization
skulking in even our revised sentence: transformation. If you want
to ensure your readers come away from this sentence with a highly
concrete picture of exactly what occurred, you might want to change
this nominalization back to a verb, like changed, and also scrounge
around for another likely actor-subject, such as companies:
First, the department reviewed how companies had drastically
changed the market for mouthwash.

I’d lay serious money, however, on no one grousing over the qual-
ity of your writing if you leave transformation intact. In fact, as
you can see, the third version of the sentence is slightly longer
than the second. When nominalizations aren’t bleeding the life
out of your verbs and making your sentences passive, they’re not
entirely bad things. Sometimes, they even let you say things more
efficiently than you otherwise could, as with the slight difference
in length between the second and third versions of the sample
sentence above.

Turning passive sentences into active involves choosing

the right subject
All right, we’re ready for a trial run at observing Clarity Principles
1 and 2. Let’s say you’re struggling over an irate letter to City Hall,
complaining about the sudden, nosebleed-inducing hike in your
property assessment. As you reread your magnum opus, however,
you stumble across the following sentence:
My property’s assessed value has been doubled in the past
year, despite the fact that it got walloped by hurricanes
Frances and Jeanne.

“Uh, oh,” you think, feeling a vague realization that the writing’s
not quite as stellar as it could be. Okay, has been could be a
50 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

non-action verb. But how can you really be certain that the entire
sentence is passive? Then you remember the by zombies test. So
you plug the phrase by zombies in here, so the sentence becomes
My property’s assessed value has been doubled in the past year by
zombies. Oops. Well, presumably, your local officials would like to
attribute anything nasty to hordes of zombies, rather than them-
selves, humble civil servants who can get the sack or chucked
out of elected office, but they rather inconveniently can’t blame
anything on zombies. At least, not yet.
Now you need to locate a possible culprit – uh, actor: the prop-
erty appraiser. Your property cannot really do the doubling. If it
did, you would just have to resign yourself to a fate at the hands
of some obscure market forces, rather than insinuating that, say,
the property assessor has a serious case of the envies where your
house is concerned. You’re writing to City Hall because you feel
someone has made a human error, right? So the actor performing
the doubling is a property assessor, or someone in the tax collect-
ors’ office, not your property or zombies.
Now we need an action verb:
The property appraiser has doubled the assessed value of my
home in the past year, despite the fact that it got walloped by
hurricanes Frances and Jeanne.

Not bad. However, that last clause, everything after past year, is
still pretty kludgy, especially the fact that it got, a passive con-
struction that doesn’t exactly trip off the tongue like iambic pen-
tameter. Try to locate a suitable actor, in this case, the hurricanes,
which are unfortunately capable of actions that produce effects,
as the residents of Florida, New Orleans, and New York can attest.
Then pair the actors with an action verb:
The property appraiser has doubled the assessed value of my
home in the past year, despite hurricanes Frances and Jeanne
damaging the house.

Your sentence is now not only City Hall-ready, you might even
get approached for a job writing for your local bureaucrats. Most
people, after all, can recognize good writing, even if they can’t
quite pull it off themselves.
 GOOD GRAMMATICAL SUBJECTS [CLARITY PRINCIPLE #3] 51

Not entirely coincidentally, our revision of the City Hall sen-

tence brings us rather nicely to our third clarity principle.

Clarity Principle #3:

Use actors or concrete objects as your grammatical subjects

Actors are individuals, groups, even abstractions that are capable

of action. Words like I, she, we, he, and they all make for solid
actors, as do researchers, Verizon, Maggie Smith, Ford, Special
Forces, the marketing department, my Jack Russell, the football
team, our investment club, and even abstractions like The White
House. In fact, even words like a recent study, the site proposal,
my letter, and the New York Times all also qualify as good actors
because these terms are all capable of effecting a change or of
acting on a situation or influencing people’s opinions.

Actors make the best subjects

Good actors make for the strongest possible subjects for your sen-
tences. Why? Actors always clearly perform actions, which have
results. When you begin sentences with actors, you also rely on
action verbs – and active construction. As a result, your sentences
will read like miniature stories, mirroring the cause–effect bias our
minds display in perceiving the world around us.26 If you fail to
find a likely actor lurking somewhere toward the tail end of your
sentence and you can’t invent a likely suspect, try casting around
for a concrete object, something your readers can easily picture.
Avoid, at all costs, abstractions as your grammatical subjects, un-
less you absolutely must focus your sentence on an abstract con-
cept. In fact, when you turn the page, you’ll discover that I’ve
used abstractions like language and ambiguity as subjects, even
though neither word is an actor, because these concepts represent
the main ideas in each sentence. Later, in our next chapter, we’ll
discover how to turn this occasional difficulty with abstract sub-
jects into a strategy that can help bind your sentences together.

Actors avoid ambiguity

Language can be wretchedly ambiguous. Take, for example, the
time I was relating to my best friend the news that another friend’s
52 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

mother, who could slip comfortably into a Size 0, had appeared

with D-sized breast implants.
“It was awful,” I told my friend. “He said they felt horribly
fake.”
My friend was aghast. “He felt his mother’s breasts?”
Fortunately, since we were chatting via mobile phone, I could
quickly correct her, telling her that my friend said he could feel
the implants when he hugged his mother, not by playing with her
chest.
But, if I had, instead, put those details into writing, that ambi-
guity would go uncorrected, could even queer my friend’s entire
reading of the event, let alone of my other pal’s character. Spoken
language seldom remains ambiguous for long because we can al-
ways halt the conversation and swiftly correct our conversational
partners. But writing enables us to extend our thoughts and experi-
ences. Writing extends our ability to persuade others, beyond the
reach of the human voice, across space, across time. But writing
also ensures we’re seldom around to correct someone when what
seemed patently transparent to us seems about as clear to our read-
ers as mud.
Hence, to write is to tango with ambiguity at every step. Your
readers may shade a word like inconsequential with a connota-
tion that never occurred to you. Or your reader might mistake
your meaning completely, mistaking the word jazz, for example,
for a men’s cologne and not a form of music. I discovered this
particular bit of ambiguity when I asked my ex-husband, while
he was watching the Harlem Gospel Choir on TV, “Is that Jazz?”
I meant the cologne he was wearing. Instead, he replied, exasper-
ated at my presumed ignorance, “No, it’s gospel.” Ambiguity is
why lawyers can turn a transaction as simple as buying a house
into a sixty-page nightmare, then earn 200 bucks an hour for read-
ing through the same sort of contract. Writing is a constant battle
against ambiguity, a struggle to cement clear meaning into place
on the page, regardless of who reads it.
So avoiding ambiguity is paramount in writing – another rea-
son why choosing an actor or concrete object for your subjects is
such a good move. For example, even a sentence as simple as This
is a good idea is rife with ambiguity. To begin with, what on earth
does this refer to, anyway?
 GOOD GRAMMATICAL SUBJECTS [CLARITY PRINCIPLE #3] 53

In case you dozed through that day in grade school English –

or, worse, you were burdened with a student teacher – pronouns
ostensibly refer to a noun, situated either in the same sentence
or in an earlier sentence. Since, however, this is the first word in
the sentence, you need to backtrack into the preceding sentence
to happen on the noun to which this refers. Unfortunately, pro-
nouns like this seldom refer to specific nouns. Instead, they tend
to reference the content of a part of a sentence, sometimes even
the entire sentence. This situation saddles your readers with two
difficulties. First, they have to determine exactly what you mean
by this, seldom an easy task. Second, and worse, your readers
must backtrack from the sentence they’ve just read to coax out
the true identity of this so they can figure out what you’re say-
ing. And any time you require readers to backtrack, as a writer,
you’re dead. Why? Prediction is the engine that enables lexical
and syntactic comprehension. Moreover, readers take more time
identifying pronouns – even definite pronouns like she – when the
pronoun referent was situated two sentences earlier than when
the culprit, er, referent, occurred in the end of the sentence im-
mediately preceding it.27
Remember, readers generally work their tails off to make sense
of even the instructions on a box of Jell-O. They’re busily sorting
out the meanings of words, putting content together and compar-
ing it with their store of knowledge, recalling what they’ve just
read, figuring out where your drift is taking them. As a writer,
you must make that task seem as effortless as you possibly can –
which means you should avoid anything that might ever force a
reader to backtrack. As a result, you should avoid using isolated
pronouns as your subjects – words like this, that, these, it, and
those, which always rely on other words to express their meaning
explicitly.

EXPERT TIP: Avoid using isolated pronouns as your grammatical subjects

Avoid using isolated pronouns like this, that, these, those, or it as your
grammatical subjects. If you must use a pronoun at the beginning of a
sentence, ensure you pair it with a noun. Turn This seems like a mistake
into This tactic seems like a mistake.
54 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

Corollary: Avoid using phrases and clauses as your

grammatical subjects
An MBA student once raised his hand toward the close of class
and got straight to the issue that had been making him writhe in
his seat for an hour.
“I’m really concerned,” he said, “about Devon’s over-reliance
on gerundial phrases as his grammatical subjects.”
While the rest of the class wanted to bash him, I had to
fight down the impulse to high-five him. His classmate, Dev-
on, should have been concerned. In fact, a surplus of gerund
phrases like listening to Devon’s endless gerundial phrases and
watching market fluctuations put Devon in the same company
as ham-fisted writers, researchers who completed PhD degrees,
and analytic thinkers who actually read something other than
the crawl at the bottom of the news. Devon saw ideas as con-
nected in an endless chain, precisely the sort of thing that grad-
uate programs encourage in their students. Unfortunately, that
kind of thinking, so handy in writing a grant, can make your
writing read like those tortuous sentences written by German
philosophers. Worse, your sentences slow down your readers’
speed, trigger needless cortical activation – the equivalent of
your car’s engine roaring as you labor up a steep incline – and
can even cause your hapless reader to slog through the entire
bloody mess of a sentence again.
Why? Gerunds, in particular, pose three challenges to readers.
Gerunds are words inviting us to interpret them as verbs, since
they are essentially verb forms that end in -ing. But, instead of
conveying action, gerunds sneakily occupy the role of a noun in
a sentence. As a result, gerunds first invite us to misinterpret the
role they play in the sentence, since readers often need to read
to the end of the gerund phrase to identify what roles the ger-
und (and gerund phrase) perform in the sentence: Interpreting the
gerund phrase interpolated between the subject and verb made us
misread the sentence. Second, gerunds seldom do anything as in-
nocuous as occupying a simple part of sentence, as in the sentence
I like swimming. In this example, swimming is acting as a noun
(the part of speech) and also playing the role of the direct object of
like. Instead, gerunds tend to insinuate themselves into sentences
at the head of noun clauses. And, third, gerunds also have a pesky
 GOOD GRAMMATICAL SUBJECTS [CLARITY PRINCIPLE #3] 55

way of encouraging writers to lard those noun clauses with em-

bedded phrases, or, worse, clauses:

Integrating the fragments gleaned from a sentence into a vast

mental database to be recalled at a later point when they
become useful could tax even a brain like Einstein’s.

Note, before you begin tearing out bits of hair, that this sentence
both represents and accurately describes a phenomenon linguists
have established in multiple studies. To make sense of the sen-
tence, the reader’s brain must place each element in the sentence
into an existing framework. In a subject–verb–object ordered lan-
guage like English, the put-upon reader of this sentence must hold
Integrating the fragments gleaned from a sentence into a vast men-
tal database to be recalled at a later point when they become useful
in suspension until she reaches the verb could tax. Along the way,
she must avoid seizing on potential verbs like gleaned, recalled,
and useful, all of which will eventually make a hash of the sen-
tence’s meaning. In addition, she has to wend her way through
thickets of phrases and clauses worthy of late Henry James’ syn-
tax. First, the sentence unfurls with the gerund phrase, integrating
the fragments, which would be easily comprehensible if only the
writer had inserted the major clause’s main verb directly after the
gerundial phrase – in addition to crafting a sentence that made
sense. Instead, though, he plugged in a participle, gleaned, then
two prepositional phrases, from a sentence and into a vast mental
database. To inflict still further pain and suffering on his readers,
he followed these phrases with an infinitive phrase, to be recalled,
then another prepositional phrase for good measure, at a later
point, and closed with an adjective clause when they become use-
ful. This final clause is particularly knotty, as adjective clauses
generally begin with that or which, rather than when, usually as-
sociated with adverb clauses. However, in this sentence, when
they become useful clearly modifies point, making the clause an
adjective clause.
Confused? You should be. After all, the reader has to hold all
those complex parts of sentence in a comprehensible framework
prior to reaching the verb. To make comprehension still more chal-
lenging, the sentence strings together different types of phrases
56 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

and clauses.28 Unsurprisingly, when readers slog through sentenc-

es like that gerund-led gem, they display multiple signs of cogni-
tive overload. In studies, readers were slower at detecting blips
flashed on a screen while reading the sentence and also struggled
to keep lists of words accurately in their memories. In addition,
their EEGs showed the high cortical activation emblematic of cog-
nitive heavy lifting.29

EXPERT TIP: If you are unable to avoid using a phrase or clause as your
grammatical subject, keep it short
Readers can avoid garden path moments and cognitive overload if you
make brief that gerund or infinitive phrase or noun clause which you
insist on using as your grammatical subject.

Clarity Principle #4:

Place grammatical subjects close to the beginnings of
sentences and the verb as close as possible to the subject

Okay, now you’ve nailed your actor or object for your subject, I’m
going to tell you where to put them. Politely.
Readers are impatient suckers – they need to be. Guessing what
comes next is actually the mechanism that propels reading along,
enabling us to interpret and comprehend what we read. Once we
know the context of an article or a novel, we can begin weighing
everything we read. Contexts serve us like maps of unfamiliar ter-
rain: we measure what we see and guess our whereabouts based
on the larger picture the map provides. Maps help us project what
comes next. And the more detail we receive from our immediate
surroundings, the better our projections, and, the better our pro-
jections, the easier the navigation.
Back during the Artificial Intelligence craze, researcher
Roger Schank attempted to design a computer that read like a
human being – and the primary mechanism that powered the
computer’s reading was projection.30 So, unsurprisingly, the
more detail you provide readers with up front, at the outset of
any sentence, the easier the task of reading. Readers naturally
 SUBJECTS BELONG ALONGSIDE VERBS [CLARITY PRINCIPLE #4] 57

expect to encounter what a sentence is about, its main actor or

subject, in the grammatical subject. Moreover, unconsciously,
readers begin comprehending sentences only after they reach
the grammatical subject. Once they locate the subject, the com-
prehension process begins, resembling – let’s say, for the sake of
visualization – an intake of breath. After readers encounter the
subject, they hold that virtual breath until they stumble across
the verb, completing the main action, the core of any sentence.
Then they exhale, ready to take in further details, modifications,
the little niggling hedges academic writers are so fond of larding
their sentences with.31
As a result of this mechanism, readers fail to notice or recall
much detail placed either before the subject or between the sub-
ject and verb, the two things that form the nub of the context for
any sentence. Pile lots of hedges or little positioning clauses at
the outset of any sentence, and you’re almost guaranteeing your
readers will have to plod dutifully back through the sentence again
in a vain effort to tease out its meaning. Shoehorn more than a
line of detail between your subject and verb, and you ensure the
content will become fodder for instant amnesia. Moreover, you
also set your readers up for a second troll through the sentence, as
they attempt to piece together the relationship between the subject
and verb.
The next time you encounter a sentence that seems more like
a trek through the Oriente rainforest than a stroll down a side-
walk, try the following test. First, draw a line under the opening
words in the sentence. If you have to plow through more than five
or six words before you hit the subject, the writer has probably
erred by packing in too much modifying detail before she provides
you with the goods you actually need to modify. Worse still, all
that detail can derail your readers’ ability to even understand the
sentence’s meaning.
You can, of course, begin sentences with the occasional short
phrase like After that last hang-over or On my way to the Hu-
mane Society without unduly taxing your readers. Similarly, you
can even salt the beginnings of some sentences with short clauses
like Although the Lexington Avenue Local was running late, as
always. Just don’t make a habit of it. Think: subject first, then
verb, then … the deluge of details.
58 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

Consider this gem, written by a faculty member in prostho-

dontics:
The short longevity of most common restorations and
especially those of resin-based composite materials calls for
detailed studies of factors which may lead to an extension of
the lifetime of restorations.

Try picturing short longevity, then count the number of words,

many of them wretchedly Latinate, between the grammatical sub-
ject, short longevity, and the verb, calls, which readers must first
differentiate from other potential verb-candidates in the sentence,
studies and lead. You staggered through eleven words between
the subject and verb, if we count resin-based as a single unit.
Now look at the same prosthodontist’s revision, the week after
my course bludgeoned her with the clarity principles you’ve just
encountered:
Resin-based composite materials fail to last long.

In the revised sentence, materials occurs within two words of the

sentence’s beginning, while fail crops up immediately after the
subject. Note that, for all its impenetrability, the original version
relies on active construction and an action verb, a good demon-
stration of why writers should try to use most of the four clarity
principles, rather than concentrating on just active construction.
And note that, in the revised version, the sentence employs both
active construction and an action verb.

EXPERT TIP: Start sentences with a subject and follow with a verb
Start most sentences with the subject at the outset, followed closely
with a verb.

Putting clarity principles into practice

Before you start attacking the sample sentences below with your
newly discovered clarity chops, let’s pause to put them together
by examining differences between two examples. Consider the
 PUTTING CLARITY PRINCIPLES INTO PRACTICE 59

differences between Example A and Example B, below. Before

reading, prepare to time yourself as you read each one, then cover
the example with one hand and attempt to paraphrase it, keeping
track of how long your paraphrase takes. Then compare the times
for reading and paraphrasing Example A to Example B. Be certain
to tackle Example A with Example B covered, for reasons I’ll ex-
plain when you’re finished.
Example A:
The permanent neurological impairment typical of chronic
inflammatory demyelinating disorders of the central
nervous system (CNS), such as multiple sclerosis, is due
to the axonal loss resulting from recurrent episodes of
immune-mediated demyelination. So far, experimental
cell therapy for these disorders has been based mainly
on the transplantation of myelin-forming cells, or their
precursors, at the site of demyelination. Although such an
approach can trigger functional recovery and restore axonal
conduction, the limited migration of lineage-restricted,
myelin-forming cells through the brain parenchyma
highlights the beneficial effect of transplantation to the
site of the injury. This raises critical issues regarding
the therapeutic use of focal cell transplantation to treat
diseases in which multifocal demyelination is the main
pathological feature. Such issues are compounded by the
poor expansion capacity of myelin-forming cells in culture,
which greatly limits their availability and further hampers
their prospective application in clinical settings.32
Example B:
Multiple sclerosis affects nearly one million people
worldwide, subjecting people from young adulthood
onwards to repeated immunological attacks on the brain
and spinal cord. Twice as many women as men are
afflicted with the disease. The effects vary depending
on where exactly in the nervous system the attacks
occur, but paralysis, blindness, loss of sensation, and
lack of coordination are among the types of devastation
wrought by an immune system gone awry. Until now,
treatment strategies have generally been aimed at blocking
60 CLARITY: CHOOSING WORDS AND STRUCTURING SENTENCES

the autoimmune attacks and reducing the amount of

collateral damage caused. Pluchino et al. (2003) describe
a complementary approach – repairing some of the harm
already done.33

Now consider the irony behind these two passages, the first of
which took you significantly longer to read or paraphrase. They
both describe the same study. The difference? Scientists wrote Ex-
ample A – probably delegating the writing to the lowest-ranking
member of the research team, usually the least experienced mem-
ber of the lot, which seldom guarantees good writing. In contrast,
an editor at Nature, ever-mindful that the journal reaches a gen-
eralist audience of science wonks, wrote the readable Example B,
a summary of the study for Nature’s introductory pages, the ones
that make the full articles at least marginally comprehensible to
non-specialist readers.

Takeaways for clarity: what you need to make your sentences easy
to read
Whenever you can, you should
• make your sentences active
• use action verbs
• choose actors or concrete objects as your subjects and
• place both actor and verb as close to the beginning of sentences –
and to each other – as possible.
CHAPTER 4

Putting sentences together

The second C: Continuity

In this chapter you will learn

• where to place information you want readers to remember
• why long sentences are so difficult to read
• how to make your sentences hang together tightly
• how to help your readers absorb even complex information
easily.

Let’s say you’re doing what no undergraduate in his or her right

mind does: bothering to read your university’s mission statement:
Teaching – undergraduate and graduate through the
doctorate – is the fundamental purpose of the university. Research
and scholarship are integral to the education process and to
expanding humankind’s understanding of the natural world,
the mind and the senses. Service is the university’s obligation
to share the benefits of its knowledge for the public good.

You reach the second paragraph, quoted in italics in the paragraph

above. How many times do you have to read it? I’m guessing, if
you were trying to figure out what the paragraph actually meant,
you probably had to read it at least twice. Why? After all, the
paragraph’s sentences are reasonably short. And the information
the sentences convey ain’t exactly rocket science. Actually, you’re
unlikely to glean any factual information from this sort of generic,
mission-statement-ese that seems to say something while really
saying next to nothing. Moreover, the language itself isn’t terribly
complex. Actually, the sentences are even active, even if the gram-
matical subjects are all abstractions. So what’s the matter with
this paragraph?
62 CONTINUITY: PUTTING SENTENCES TOGETHER

Mind the gap!

The answer actually lies between the sentences. Look at the gaps
between each sentence carefully. Each sentence exists in splendid
isolation from the sentence that precedes or follows it. The first sen-
tence is about teaching, the second, about research and scholarship.
And the third is about service. No words act as transitions to guide
readers from one sentence to the other. And each sentence fails to
provide so much as a suggestion of what links its concepts to the
contents of neighboring sentences. In fact, only a college faculty
member could probably read this paragraph, pause, and still cor-
rectly recall its content because only faculty understand that careers
at big universities involve teaching, scholarship, research, and ser-
vice – at least in theory, on annual reports, and during usually beady
tenure and promotion performance reviews. Unfortunately, univer-
sities seldom write mission statements for people contemplating car-
eers at a university. Instead, mission statements ostensibly target
would-be students. Or, in some cases, the parents who plan on foot-
ing the bill for their kid’s tuition and want to know something about
the place that threatens to eat their entire retirement savings. As a
result, the academics who could understand this paragraph easily
are unlikely ever to read it. Ironic, isn’t it?
Look carefully at this paragraph again, a paragraph written
by a professional, probably in exchange for a paycheck. Yet this
professional writer made the same mistake that virtually all begin-
ning writers make – taking for granted what readers know. You
can provide all sorts of background information and fine details
for your readers to ensure they aren’t scratching their heads by
your second paragraph. But you also have to be constantly vigilant
about keeping them on the right track from sentence to sentence.
As many of us learned from a hoary definition of sentence, each
sentence is a “complete thought.” The problem with churning out
a series of complete thoughts is that sentences become little iso-
lated units of meaning. The challenge is that you must make these
little islands of thought seem to fit together, ideally, tightly and
logically, which is perhaps the single most difficult thing to man-
age in your writing.
The good news: continuity is far easier to achieve and requires
a whole lot less behavior modification than clarity. The bad news:
 WHY IS CONTINUITY SO IMPORTANT? 63

if your sentences don’t fit together tightly, your reader is going to

feel your document is badly written or confusing, no matter how
many pains you’ve taken over getting the clarity straight.
While continuity contains five different principles that enable
you to link your sentences together, you don’t have to use all the
principles at the same time. Unlike clarity principles, continuity
rests in your using only two or three principles at any time to tie
any sentences together. You always have to observe Continuity
Principle #1, which appears immediately in the pages that follow.
But you can use a mixture of Principles 2–5 to achieve continuity
in any paragraph. So you can tether your sentences together by
relying on either transitions or sequencing or common grammat-
ical subjects, or even a mixture of these principles, if you feel your
sentences are so complex that your readers need extra guidance to
get through a particular paragraph.

Why is continuity so important?

Continuity is, arguably, the single most important aspect of your

writing. Periods at the ends of sentences can turn them into archi-
pelagos of meaning, isolated from the adjoining sentences by both
white space and either gaps in logic or shifts in topic. When you
leave these gaps between sentences, you require readers to cross
open water to journey from one sentence to another. Actually,
however, you should be building bridges or at least providing reg-
ular ferry services between sentences to prevent your readers from
ever having to swim from one islet to another. Why? Remember,
reading is a difficult, demanding task, even when you’re not ter-
ribly aware of all the machinations you’re going through as you
read. Your readers are busily assigning meanings to words, recall-
ing what you’ve already said, guessing about what you’re going to
say next, holding sentences in short-term memory to comprehend
their meaning, then filing information away where it can be re-
trieved easily.1 As a result, your readers are grappling with a seri-
ous amount of what psychologists call cognitive load.
When you require readers to build bridges between sentences,
you’re asking them to do one thing too many. When you send
your readers into what psychologists dub cognitive overload,2 not
64 CONTINUITY: PUTTING SENTENCES TOGETHER

surprisingly, they are likely to get lost. Worse, readers can make
the wrong assumptions and string your sentences together by
making the wrong connections, which usually means your reader
might seriously mistake your meaning – or just think you have
strange ideas.
In any case, if you fail to build bridges between your sen-
tences, you’re almost certain to lose readers – and to seem like
a lousy writer. But the act of reading relies on a mechanism that
makes continuity still more crucial to good writing: prediction. In
studies in the 1970s and 1980s, researchers discovered that readers
are constantly and unconsciously making predictions every instant
they read. In fact, cognitive psychologists agree, to understand the
world around us we must constantly categorize everything we per-
ceive, an act that requires us to predict the significance of what
we’re seeing or hearing at any given moment.

Understanding and perception

In the 1970s, cognitive psychologists argued for the importance

of frameworks to our perceptions by pointing out that even
ordering a meal in a restaurant involves a rather complex set
of expectations.3 For example, you have to understand the pur-
pose of those rectangles of cardboard someone usually hands
you when you sit down. You also have to grasp that the items
printed on that menu refer to things you can order, eat, and pay
for. And you have to realize that the person standing over your
table is waiting to take your order, not to recommend treatment
for that persistent headache you get when you’ve had too much
caffeine.
Moreover, different types of restaurants require different
categories of understanding. If you waltz into a Mickey D’s and
expect to be waited on, you could wait until you’re past child-
bearing age before anyone strolls over to your table. But you’d
also be confused if you walked unsuspectingly into an antipasti
restaurant in Italy and waited for a menu to arrive. Antipasti
restaurants avoid menus. Instead, diners or waiters pass plat-
ters around, and you help yourself to whatever you fancy, then
own up to what you’ve eaten at the end of the meal. (Antipasti
 PREDICTION IS ESSENTIAL TO UNDERSTANDING 65

restaurants also have their distinct benefits. Occasionally, the

waiter toting up your bill will give you a break by refusing to
believe you’ve had six glasses of the house white and three des-
serts.) To understand the meaning of what happens to us mo-
ment by moment in any situation, we have both to understand
the entire context and be able to predict what’s coming next.

Prediction is essential to understanding

Let’s say you decide to go to the movies with some friends and
end up watching a romantic comedy featuring Julia Roberts. If,
halfway through the film, another character were to put a gun to
her head and blow her brains out, you’d most likely be floored
and wonder what sort of film you’d been watching after all.
This reaction is due entirely to our brains’ relying on established
frameworks for organizing the world. When you believe you’re
watching a romantic comedy, you unconsciously expect two char-
acters to be involved in a Two Weeks Notice sort of push–pull of
attraction and bristling character differences. So, when she ven-
tures into a dark alley or a particularly nasty-looking basement,
you don’t start looking for a serial killer lurking in the shadows.
Romantic comedies mean characters fall in love, come together,
pull apart, and, eventually, come together before the close. If,
however, a film billed as a romantic comedy involves one or both
of the main characters being murdered, you’d have great diffi-
culty figuring out exactly what’s going on from scene to scene.
Similarly, a horror film without anyone being killed, in danger
of being killed, or caught in a basement in the dark with (a)
a confirmed serial killer, (b) the latest incarnation of Jason, or
(c) ghosts, dead bodies, mummified mothers, or alien life forms
would probably have you clamoring for your money back from
the box office. Solid, conventional frameworks that organize stor-
ies, books, and papers help us to make firm predictions about
the significance of our perceptions, which, in turn, help us to in-
terpret moment-by-moment details. When you spot a knife lying
around in a teenager slasher flick, you can bet someone’s going
to use it – sooner rather than later.
66 CONTINUITY: PUTTING SENTENCES TOGETHER

Reading and prediction

As a result, reading inevitably involves prediction. As your eyes

scoot across a page, your brain is busily predicting the outcome of
the document you’re reading, as well as the content for the para-
graph – even the ending to the sentence.4 This mechanism, cen-
tral to our ability to understand what we read, has two important
implications for writers who want to create continuity between
their sentences – perhaps the most essential component to good
writing. First, you must recognize which parts of sentences, para-
graphs, and entire documents receive the greatest amount of em-
phasis in your readers’ minds. And, second, the more information
you give readers at the beginnings of sentences, the better their
predictions and their ability to process information.5
We’ll call this concept emphasis, since studies have established
that readers remember items that appear in emphasis positions
better and longer than they do items that appear in non-emphasis
positions. And we’ll use the familiar term transitions to focus on
how you can help readers to make better predictions about the
sentences they’re about to read – even as they begin reading them.

TERMS TO REMEMBER

Emphasis – The position in a sentence, paragraph, or docu-

ment that readers recall best and longest. The ends of sen-
tences, paragraphs, and documents always receive the greatest
amount of emphasis.
Transitions – Words or phrases that tell your readers how to
interpret the sentences they’re about to read relative to what
they’ve just read. Transitions include clauses and phrases, as
well as words like also, too, and, but, since, although, however,
consequently, and because.

Continuity Principle #1:

Place the most important information at the ends of
things – sentences, paragraphs, and entire papers

When given a list of items to memorize, readers always recall

the last items best, with the fewest errors and greatest number
 THE EMPHASIS POSITION [CONTINUITY PRINCIPLE #1] 67

of items recalled from the last 25–30 percent of the list. Readers
mostly remember items at the beginnings of things second best
and items in the middles of things – lists, paragraphs, papers – the
least well. In fact, if you have to give your readers bad news or
disclose a weakness in your own argument, you should sandwich
these revelations between positive items.6

How do you manage emphasis?

You should always make use of the emphasis position at the ends
of sentences to put new or important information in the last quar-
ter to one-third of the sentence. All too often, however, writers
tend to allow sentences to limp to a conclusion, like this one:
My manager tore my proposal to shreds, told me I was a waste
of space, then demanded I pack up my desk and get the hell
out of the office during last week.

Look carefully at this sentence. What’s in the end of the sentence?

My manager tore my proposal to shreds, told me I was a waste
of space, then demanded I pack up my desk and get the hell
out of the office during last week.

The end of this sentence contains the last seven words, which, if you
know your grammar, consist of two lowly prepositional phrases.
Even if you don’t know anything about grammar, however, you
can see that the least important items in the sentence are the in-
dications of where and when the incident took place. Obviously,
my audience cares more about your getting fired than when and
where you got fired. Unless, of course, you get canned every few
weeks, which could put a serious dent in your career.
To maximize the emphasis position, the sentence should read
something like this:
Last week, my manager tore my proposal to shreds, told me I was
a waste of space, then demanded I pack up my desk and leave.

As you read, pay attention to the content of the last quarter to third
of each sentence. Is the writer placing the most important informa-
tion in the emphasis positions? As you write and revise your writing,
make sure that you’re doing the same. If you find a sentence ends in
a whimper rather than a bang, like the first example above, rewrite it.
68 CONTINUITY: PUTTING SENTENCES TOGETHER

EXPERT TIP: Limit the number of items in lists

You can count on the beginnings and endings of lists to nudge your
readers’ recall – but only so far. Early studies of the number of items
recalled from the beginnings of lists ran to three or four items, while the
ends of lists benefited from a still-greater bump due to the recency ef-
fect. In studies dating back to the 1960s, readers accurately recalled
as many as eight items from the ends of lists.7 However, this recall is
higher than either you or most psychologists should expect, especially
given the broad durability of George Miller’s “seven-plus-or-minus-two”
estimate of the number of items an average person can hold in working
memory.8 As a result, try to limit items in lists to seven or fewer for bul-
leted lists, where items will stand out separately from the sentence that
contains them. For items inside a sentence, try keeping to Miller’s lower
threshold of five items, as readers tend to skim or even skip dense lists
contained within sentences.
If you’re a typical human, you doubtless remember all those times
you guiltily skimmed, scanned, or even skipped content in reading,
from the dreadfully dense descriptions in those novels you had to read
for AP English in high school to the financial prospectuses for funds
in your 403(b). From a writer’s perspective, as well as an editor’s or
teacher’s, you might hope researchers have been busily scrutinizing the
point at which your average reader bails on close reading and launches
into skimming mode. But, surprisingly, few researchers have focused
on how attentively your everyday native speaker of English reads dense
paragraphs or lengthy lists. Eye-tracking software long ago easily
established the number and length of eye movements readers use on
lines and even individual words. However, since research tends to silo
investigation into tidy categories, researchers have looked at what eye-
tracking tells us about reading and comprehension, inexpert versus
expert readers, rapid readers versus slow readers, and, most recently,
how readers handle multimedia inputs. To date, no researchers have
thought to investigate the relationship between paragraph or list dens-
ity, independent of the syntactic complexity of the sentences and items,
and readers’ shifting from close reading into skimming. One study,
however, found even the most expert readers had a reading span that
encompassed only five items, meaning that writers should think care-
fully about putting more than five items in any list.9
 THE EMPHASIS POSITION [CONTINUITY PRINCIPLE #1] 69

This principle applies even to bulleted items in a list. Writers some-

times think dumping information in a list represents the most efficient
way to get it to readers. But, since readers may recall, at best, five items
in any list, you’re virtually guaranteeing that they won’t remember most
of the contents in long lists.

Why are long sentences so difficult to read?

Wolfe certainly invokes the key figures – William James,
Ralph Waldo Emerson, and Charles Sanders Peirce (rather
sheepishly) – but he also draws upon Immanuel Kant,
Jean-Paul Sartre, René Descartes, as well as more current
figures such as Chantal Mouffe, Ernesto Laclau, and Cornel
West, if only to remind us that pragmatism, especially Rorty’s
strain, may want to express an alterity, an outside of “theory,”
but it is very much a part of the system.

Cover the sentence above with your hand and try to recall as
much of its specifics as you can. Now lift your hand and see how
much you actually managed to remember. If you read the sentence
and know something about philosophy, you might have noticed
that this sentence is dedicated to covering key theorists relevant
to the writer’s study. Not surprisingly, most of the details in the
sentence probably escaped you, especially the names of the key
theorists. Why?
For starters, the entire example is a single sentence. Long sen-
tences have three things going against them. First, they contain
few cues informing readers which information in the sentence is
particularly important – and that readers should remember. Sec-
ond, long sentences also have the same amount of emphasis as
any other sentence: the contents in the last quarter to one-third
of the sentence. As a result, readers will tend to forget virtually
everything else in the sentence. Worse, readers are likely to lose
track of relationships between elements in particularly complex,
lengthy sentences. In addition, readers may also forget the subject
of the sentence by the time they reach the end, where contents
are likeliest to lodge in their memories. And, third, long sentences
70 CONTINUITY: PUTTING SENTENCES TOGETHER

place higher demands on working memory, making them difficult

to read.10
Note, incidentally, how weak the writer’s use of the stress pos-
ition in the example above seems. Both it and system are ambigu-
ous terms, particularly since the readers have to track backward
through a lengthy sentence to try to identify the referents. Both
factors ensure that your readers will have difficulty in figuring
out how the sentence they’ve just finished reading connects with
the one they’re about to read. And this fact brings us to our next
principle: using transitions.

Continuity Principle #2:

Use transitions to tie sentences together

Readers understand what they read by projecting forward, roughly

anticipating the content of the next sentence. Your reader might
struggle to pin down the meaning of the word remit to establish
whether the word is a verb or a noun, strain to spot the verb in the
major clause, or make sense of two apparently unrelated sentenc-
es. If you were paying attention in Chapters 2 and 3, you’d identify
lexical, syntactic, and inferential processing in the string of read-
erly actions in the preceding sentence. At every stage, prediction
is the engine driving comprehension in reading. However, because
every sentence can theoretically be an island of meaning, read-
ers stumble, pause, and reread when they read a sentence about
pizza followed by one dedicated to bubonic plague.11 Unless you’re
a card-carrying hypochondriac or an unmedicated obsessive-com-
pulsive fretting about rat fleas in the kitchen of your local pizza
joint, you’re unlikely to see anything connecting the two sentences.
Transitions easily build minute bridges from sentence to sen-
tence, bolstering your readers’ ability to comprehend relation-
ships between sentences. Essentially, you’re telling your audience
how to interpret the sentence they’re about to read relative to
what they’ve already read. When you use and, also, or too to-
ward the beginning of a sentence, you’re telling your readers that
the next sentence is adding to the information they’ve just en-
countered in the preceding sentence. But when you use but, how-
ever, conversely, on the other hand, or yet as transitions, you’re
 USE TRANSITIONS [CONTINUITY PRINCIPLE #2] 71

announcing to your audience that the content of the next sen-

tence is going to contrast with what they’ve just read – the direc-
tion of the topic is about to perform an about-face.
Without these transitions, your readers will be left wondering
how sentences are related. Occasionally, your readers may even
have to backtrack and reread sentences when they discover that
they haven’t correctly guessed relationships between sentences.12
And, as we’ve already discussed in Chapter 2, any time your read-
ers have to backtrack and reread a sentence or several lines, you’ve
fallen down on your job as a writer. Remember, transitions give
you the most bang for your buck in cementing your sentences to-
gether, since transitions are generally both short and easy for read-
ers to process.

EXPERT TIP: The power of causal transitions

Transitions like as a result, therefore, because, and consequently tell
readers that the sentence they’ve just read caused the outcome in the
sentence lying just ahead. Never underestimate the power of causal
transitions, especially if you’re a lawyer, an employee arguing for a merit
pay increase, or anyone crafting a contentious bit of argument. Studies
of causal transitions established that they significantly increase reading
speeds.13 In addition, given the central role causation plays in our per-
ception generally, you’re also encouraging readers to follow your line of
thinking, rather than pausing to argue with your connecting pizza and
bubonic plague.

Transitions toolkit

Transitions can be as brief as a single word or as long as a

clause. Most transitions, however, are short and fall into ten
categories:
Continuity: also, and, besides, further, furthermore, in addition
to, likewise, similarly, too, not only . . . but also, too.
Contrast or exception: although, but, conversely, despite, how-
ever, in contrast, instead, nevertheless, nonetheless, on the other
hand, otherwise, still, though, yet.
72 CONTINUITY: PUTTING SENTENCES TOGETHER

Conditions: if, if . . . then, whenever.

Frequency or time: after, afterward, at the same time, before,
before then, during, earlier, eventually, later, meanwhile, now,
once, once again, sometimes.
Order: first, second, third, last, primarily, ultimately, finally,
more importantly, most importantly, in closing, in conclusion.
Example: for example, for instance, specifically, particularly.
Amplification: apparently, actually, especially, in fact, in par-
ticular, indeed, of course.
Cause: as, as a result, because, for, for this reason, since,
therefore.
Result: accordingly, as a result, consequently, hence, therefore,
then, thus, so.
Conclusion: then, finally, generally, in conclusion, in the end,
in short, ultimately.
Studies of the power of transitions have also focused on the
strength of transitions signaling shifts in time, particularly in
narratives, which crop up in news stories as well as in those
guilty-pleasure novels one tends to find in airport news agents.14

Conjunction junction
Remember: to be effective, transitions have to appear toward the
beginning of any sentence. You must place a transition either be-
fore the grammatical subject or before the verb for the transition
to function effectively. Once readers hit the verb in any sentence,
their most substantial work in interpreting the sentence is over.15
When writers whack a transition in at the close of a sentence,
however, they’ve completely erased any benefits the transition
might have given their readers. Compare:
Example A:
Rogers, Sanborn & Son had left the project and the site in
limbo for several years. The team discussed the liabilities of
building on the site, the difficulties in minimizing the
environmental impacts, and the necessity of driving foundations
on piles, then Sanders & Co. decided to give up on the project, too.
 USE TRANSITIONS [CONTINUITY PRINCIPLE #2] 73

Example B:
Rogers, Sanborn & Son left the project and the site in limbo
for over five years. Similarly, Sanders & Co. also decided to
give up on the project after the team discussed the project’s
numerous challenges. These challenges include (1) the
liabilities of building on the site, (2) difficulties in
minimizing environmental impacts, and (3) the necessity
of driving foundations on piles.

In Example A your readers might be prepared for the second sen-

tence to also involve Rogers, Sanborn & Son, since the first sen-
tence focuses on this company. But the second sentence switches
the focus to Sanders & Co., which disorients readers who only
encounter the change of company at the end of the sentence. And
readers are likely to miss the continuity between the two com-
panies giving up on projects on the same proposed site if they only
stumble across the transition that links them, also or too, at the
very end of the sentence in the first example. In contrast, Example
B shifts the change of companies to the beginning of the sen-
tence but also highlights the continuity between the companies
that have both found the site equally difficult to work with. Note
that Example B also breaks into two sentences the skinny on why
Sanders & Co. decided to also abandon the project and helpfully
numbers the three most important items, creating a list that helps
readers retain the contents. The bottom line: if you’re inserting a
transition after the verb, forget it. You’ve just done the equivalent
of offering a climber a hand-up after she’s already scaled the entire
bloody cliff.

Can you have too many transitions? Not really!

Typically, in American journalism courses, fledgling journalists are
cudgeled into beginning every sentence with a transition. (Full
disclosure: in one of my former incarnations, I performed a fair bit
of cudgeling myself when I taught journalism and advised journal-
ists on two student-run publications.) Journalists lean heavily on
transitions, and American newspapers rely on the bracing brev-
ity of single-sentence paragraphs for a good reason. Newspapers
have pegged the average American reader’s skill at a dismal fifth-
grade level. When you grasp that unpalatable fact, you begin to
74 CONTINUITY: PUTTING SENTENCES TOGETHER

see the sense of those single-sentence paragraphs, mostly contain-

ing sentences of twenty-odd words or fewer, all beginning with
transitions that connect them snugly to the preceding sentences.
However, transitions in journalism are not a sop to distracted read-
ers who mouth words as they read silently. Transitions also work
equally well in writing for audiences with longer attention spans
and a bit more education by mortaring sentences together, sup-
porting readers’ subconscious predictions of inferences across sen-
tences. With a single word as unobtrusive as also, slipped quietly
between your subject and verb, you seamlessly link two thoughts
that would otherwise seem isolated or require a bit of guesswork
to forge connections.
While no one is going to breathe down your neck, insisting
on a transition in every sentence, you should be using transitions
approximately every two to three sentences. And, unlike so many
other things in life, you can never use too many. Transitions also
offer your sentences an additional benefit: they vary the rhythm
of your writing. If you begin every sentence directly with a gram-
matical subject (or article and grammatical subject), the rhythm
of your sentences will seem to be monotonous: dum-dum-dum-
dum. But when you place transitions at the beginnings of your
sentences, you end up varying this rhythm, even when you use
transitions in two or three consecutive sentences.
Unlike music, a pleasing rhythm in writing involves variety, not
similarity, a concept we’ll explore in greater detail in Chapter 7. By
varying the structure of your sentences somewhat, as well as their
length, you’ll establish the pleasing rhythm that is the hallmark of
truly accomplished writers who are fully in control of their work.
And remember, readers always subconsciously register the rhythm
of their sentences, even when they’re reading swiftly and silently.

Writing myths you’ll wish you never learned:

Never begin a sentence with and or but

This edict, which teachers may have handed down to you in mid-
dle or high school, is simply wrong. Like the proclamations of the
grammar mavens hung up on split infinitives – a rule derived from
Latin to confer gravitas on the mongrel that is English – the ban on
 USE TRANSITIONS [CONTINUITY PRINCIPLE #2] 75

beginning sentences with conjunctions has two probable sources.

First, coordinating conjunctions like and, or, but generally connect
two major clauses. However, you won’t flummox readers encoun-
tering but at the beginning of a sentence that clearly inverts or
hedges on the suppositions of the preceding sentence for the sim-
ple reason that the full stop or period preceding but tells readers
the conjunction here isn’t connecting two major clauses. Instead,
the conjunction merely functions as an additive or contrasting tran-
sition. (See the Supplement, page 164, for the role of punctuation
as a means of avoiding garden-path stumbles by disambiguating
the function of words.) Second, the rule also stems from what we
might think of as the training wheels effect, where well-intentioned
teachers lay down principles as immutable as the Ten Command-
ments for seven- and eight-year-olds, then neglect to tell them that
these rules were merely supposed to curb wayward behavior, like
writing sentence fragments, rather than serve as guidelines for a
lifetime of practice.
In fact, you can begin any sentence with a conjunction. When
you begin sentences with and or but, you cement your sentences
firmly together. If, however, you feel a bit squeamish about begin-
ning a sentence with and, or your readers may well have been born
in the first half of the last century (and thus labor under the old ca-
nards about conjunctions beginning sentences), you might instead
want to use also or too and sandwich these transitions between
your subject and verb.
Rest assured that you can easily silence grammar mavens by
noting that the injunction against split infinitives, which would’ve
deprived Star Trek of its immortal opening phrase . . . to boldly
go . . . was tacked onto English to provide this Johnny-come-lately
of a language with at least one ancient rule derived expressly from
Latin. The problem, of course, is that no one on the planet could
split an infinitive in Latin or, indeed, in any Romance language.
The reason? In Latin and all Romance languages, infinitives are
single words, incapable of being split unless you’re aiming to cre-
ate a slang even more flummoxing than the Cockney rhyming
slang that counsels you to answer the dog because dog and bone
rhymes with phone.
76 CONTINUITY: PUTTING SENTENCES TOGETHER

Continuity Principle #3:

Sequence information in sentences in a
familiar–unfamiliar, familiar–unfamiliar pattern

The question of how to transmit what one speaker or writer knows

to another, from brain to brain, has first transfixed philosophers,
then linguists, and, finally, psychologists. Like the Artificial In-
telligence fad, this question – linguistic and psychological – has
driven research that would be a boon to writers everywhere, if
only they were willing to troll through archival issues of journals
with titles like Language and Consciousness and Memory & Cogni-
tion. Instead, most beleaguered writers just want to know how the
hell you tie together two ideas, coupling them so tightly that even
skeptical readers would have difficulty inserting a knife-blade of
resistance between them. Enter sequencing. More difficult to de-
scribe than to actually employ, sequencing is your best means of
binding sentences together when you’re describing material that
is highly technical, difficult to comprehend, or potentially conten-
tious to your readers.
To sequence information, you follow Continuity Principle #1
and place new and important information at the end of your sen-
tence. But you also introduce familiar information, continuous
from an earlier sentence, into the beginning of the sentence, eas-
ing your readers into the unfamiliar material. Sequencing plays
off three things driving the reading brain: assumptions about
relevance, inference processing, and the priming effects of words
in preceding sentences. First, in studies beginning in the 1970s,
researchers realized that readers process material already famil-
iar to them first, prior to tackling new information.16 In addition,
researchers also discovered that humans automatically see rel-
evance in statements in the world around them, in speech and
writing alike. We assume relevance between sentences, maximiz-
ing our comprehension with minimal cognitive effort. Sequencing
explicitly takes advantage of this tendency to see relevance by es-
tablishing an explicit link between new information at the outset
of each sentence and already familiar information introduced at
the close of the preceding sentence.17 Second, readers rely on rele-
vance in the inference stage of processing, focusing on the imme-
diately available content in the preceding sentence to make sense
 SEQUENCE INFORMATION [CONTINUITY PRINCIPLE #3] 77

of the sentence awaiting them.18 And, third, readers will also see
as intrinsically related sentences connected by similar words, tied
together by what researchers term referential continuity, where
one sentence refers implicitly or explicitly to the content of the
preceding sentence. In studies that deliberately scrambled or dis-
rupted referential continuity, readers demonstrated significantly
stronger recall of sentences that relied on referential continuity
than they did sentences with no apparent connection.19
Put simply, readers interpret material most easily when they’re
able to position what they’re about to read relative to what they’ve
just read. As a result, sequencing is as effective as transitions, if
not more so. Furthermore, when you use sequencing, you ease
your reader into information by creating an unobtrusive chain of
references that seem to be continuous, even if you’re introducing
new information at the end of every sentence.
Consider, for example, the following groups of sentences:

Example A:
Banks within a particular market tried to shore up their
particular market share, a practice that left them exposed
to a buyout. Other banking institutions were expanding
outside their current markets. The federal government
could force an institution to sell itself to a buyer that
the federal government had already chosen. The buyer
normally bought the bank being sold because the bank’s
sale price had fallen beneath its market value.
Example B:
Banks within a particular market tried to shore up their
respective market share, a practice that left banks exposed
to a buyout. This practice allowed for other banking
institutions to expand outside their current markets.
In some cases the federal government could force the
entity to sell itself to a buyer that the federal government
had already chosen. When given this option, the buyer
normally bought the bank being sold because the bank’s
sale price had fallen beneath its market value.

If you leave Example A feeling confused about exactly what the

hell happened, you’re not alone. In this version, every sentence is
78 CONTINUITY: PUTTING SENTENCES TOGETHER

isolated from the sentences before and after it by an absence of

transitions. But these sentences also lack continuous content and
any trace of referential continuity – shared words common to two
or more sentences.
Now look at Example B where the sequenced material is under-
lined. Note how using sequencing requires you to refer to the same
concepts, using some of the same terms to name them. You don’t
have to use the exact same words from sentence to sentence, al-
though readers will recognize immediately that you’re referring
to the same things you’ve mentioned in the preceding sentence.
However, to make this principle work, you need to introduce the
familiar words and concepts toward the beginning of any sentence,
immediately following their appearance at the end of the preceding
sentence.

Continuity Principle #4:

Try to keep grammatical subjects consistent
from sentence to sentence

You can also build rickety bridges between those little atolls of
meaning, less durable than the concrete causeways of sequenc-
ing that have your readers gliding from terra firma to terra firma
without ever realizing they’ve just crossed a scrap of open water.
In eye-tracking studies, readers typically lingered longer, not only
at sentence boundaries, but also at sentences with evident shifts
in topic from the preceding sentence.20 When writers carelessly
reference material they introduced as many as six or seven sen-
tences earlier, even the good readers strain to integrate the sen-
tence they’re struggling to comprehend with content that cropped
up much earlier in the same paragraph.21
Unlike transitions – the remedy for continuity that no writer can
overuse – too much sequencing can make your reader feel as though
you’ve dragged her through your document with a pair of pliers ap-
plied to her nostrils. Or he can feel as though you’ve lumbered him
with a sort of See Spot run. Run, Spot, run elementary simplicity that
comes across as a bit insulting for any reader north of second grade.
While sequencing has limited applications, common grammatical
subjects are instruments of continuity on which no writer can OD.
KEEP GRAMMATICAL SUBJECTS CONSISTENT [CONTINUITY PRINCIPLE #4] 79

Grammatical subjects unobtrusively ease your readers from

sentence to sentence by using a priming effect. Studied for dec-
ades, priming effects involve exposing readers to something as
unremarkable and thoroughly unmemorable as a list of unrelat-
ed words, pairs, phrases, or sentences.22 Readers shown a list of
words like
elephant
rubber ball
bus
brochure
have accurate recall of the list, even a day or days later, when they
see it again.23 The impact on recall seems due to priming as a form
of implicit learning that, remarkably, works as effectively with am-
nesiacs as with skilled readers.24 The repetition of key concepts
via the same word, used as a grammatical subject, has two effects.
First, the priming effect subconsciously nudges your readers to
identify the repeated grammatical subjects as important. Second,
this same priming effect also enables readers to recognize the top-
ic of the paragraph, aiding them considerably in inference building
and intuiting what researchers have called the macrostructure of
the paragraph.25
As off-putting as the term might sound, the concept of macro-
structure is useful for writers to bear in mind as they craft para-
graphs – as we’ll see in Chapter 5. Put simply, the macrostructure
is the organizing principle behind the paragraph, a central series
of points or events that rely on readers’ accessing a schema for
understanding what’s going on, that good old inference-building
stage of comprehension. If I’m reading a paragraph that contains
a screed on the fate of the apostrophe in the Internet era, I can
move easily from a sentence on its disappearance from possessive
plurals, as in Two Days Notice, to the unfortunate appearance of
an apostrophe wherever plurals crop up, as in Flea control for cat’s
and dog’s only. In other words, the macrostructure or schema of
the paragraph is about the abuse of apostrophes in modern Eng-
lish. When you use the word apostrophe itself as a grammatical
subject, you prime readers to expect the paragraph to bang on
about apostrophes.
80 CONTINUITY: PUTTING SENTENCES TOGETHER

Moreover, since the beginnings of sentences receive second-

ary stress, readers recall words from the beginnings of sentences
fairly clearly. In addition, grammatical subjects represent the core
of every sentence – or should, if you’re not relying on lame words
like this and it. Unfortunately, this linkage is seldom as sturdy as
sequencing. As a result, you should use some transitions between
sentences that you’ve connected only through same grammatical
subjects. Also, once again, as with sequencing, you can use simi-
lar terms to maintain continuity, rather than identical subjects in
every sentence.
Compare the following two examples:
Example A:
The services proposed have clear advantage over the
present services available, primarily due to the proposed
low cost and high quality initiatives. Captive technologies
will dissuade mimicry by new entrants through
introduction of Mr. Hutton’s “electronic notebook” training
tools. Essential skills, captive to the organization, are
adequate to the services offered with a high degree of
flexibility due to the owner being the chief employee. The
intended clients are both numerous and in need of the
services offered, as expressed in the disclosed research.
Example B:
ITS’ business plan offers clear advantages over its potential
competitors, primarily due to the company offering low cost
and high quality training. Moreover, the business will
discourage competitors from entering the field via Mr.
Hutton’s “electronic notebook” training tools. Furthermore,
ITS enjoys the advantage of its owner also acting as the
company’s chief employee, bringing valuable expertise to
clients. Finally, ITS targets a potentially large client base
who need its services, a claim the plan backs with market
research.

As you can see from looking at the underlined words in Example

B above, the grammatical subjects are different. Instead, in each
sentence the grammatical subjects only refer to a common set of
topics: the business, known as ITS, and its business plan. Notice,
KEEP GRAMMATICAL SUBJECTS CONSISTENT [CONTINUITY PRINCIPLE #4] 81

however, that transitions also help tie these sentences together.

Remember, common subjects alone will not tie sentences together
as tightly as sequencing or transitions will.

Writing myths you’ll wish you never learned:

Vary your word choice

Primary and secondary school teachers have ensured that this par-
ticular myth thrives, mostly because they wanted to justify giving
you all those horrible vocabulary spelling and definition tests every
week. In this instance, the training wheels effect also applies, as
teachers tend to encourage word variation for its own sake, which
should remind you of the peacock’s tail. Those long, drooping,
brilliantly iridescent feathers that handily attract the females also
make the peacock vulnerable to predators who can clamp onto his
tail and turn him into dinner. Get just a tad too creative with a the-
saurus and fail to double-check word choices against a dictionary,
and you can end up looking like an idiot. Or implying something
queasily sub rosa in a job application cover letter.
When you use multiple terms to refer to the same thing, your
readers might be so confused, they’ll fail to realize that you’re talk-
ing about the same thing with each change in word choice. Varying
company and corporation won’t derail any readers. But if you use
doctrine, edict, manifesto, declaration, statute, and charter to refer
to the same thing in a single document, you’ll almost certainly
confuse the pants off any reader.

EXPERT TIP: Use consistent wording

In addition to priming effects from using consistent grammatical
subjects, you also leverage your readers’ availability biases – their
tendency to recall things they encounter frequently. This attribute of
perception was first discovered by Amos Tversky and Daniel Kahne-
man, a pair of psychologists whose thinking has proven influential in
economics but applies equally usefully to writing.26 In fact, your read-
ers will spend less time focusing on words you repeat frequently – a
sign that their brains aren’t generating a head of steam to make sense
82 CONTINUITY: PUTTING SENTENCES TOGETHER

of what they’re reading – in contrast to words mentioned only once or

twice.27 These three beneficial effects – priming, availability, and fre-
quency – apply even when you use consistent word choices in nouns
that crop up in the middles or ends of sentences, lending continuity
to your paragraphs.

Continuity Principle #5:

Continuity is more important than clarity. If you can only
maintain a strong sequence or a consistent subject by
using passive construction, then use it

I’ll remind you of Ralph Waldo Emerson’s famous quotation –

usually mangled when quoted – “[C]onsistency is the hobgoblin
of little minds” – since you’re likely to be pissed off that, after
my banging on about the Importance of Clarity in Chapter 3, I’m
about to tell you to chuck it. While clarity has its place in making
reading less arduous than digging fence posts, you can make read-
ers work overtime if you omit continuity between your sentences.
In the early 1980s, researchers pounced on eye-tracking de-
vices to measure the movements of readers’ eyes and made sev-
eral startling discoveries. First, readers’ eyes paused significantly
on detail at the beginning of sentences, evidently straining to link
it to the content of the preceding sentence. Second, readers’ eyes
stopped longer on sentences that had no attachment to the preced-
ing sentences than they did on sentences that clearly referred to
the preceding sentence. And, third and most significantly, read-
ers focused on the common wording from sentence to sentence,
relying on this boost in making inferences about meaning, even
when the actual meaning of the sentences themselves depicted
highly implausible conditions.28 In other words, those sequences
and common grammatical subjects bolster your readers’ ability to
make sense of what you’re saying – even when you’re not mak-
ing a whole hell of a lot of sense. This research should console
procrastinators everywhere when they’re pulling an all-nighter as
they scurry to make the deadline on a proposal or research project.
“Aha,” you’re probably thinking. “Screw clarity and just ob-
serve continuity.” Not so fast, Spiderman (or superhero du jour).
 CONTINUITY TRUMPS CLARITY [CONTINUITY PRINCIPLE #5] 83

In about 90 percent of sentences, you can accommodate clarity

and continuity principles without compromising either. In some
instances, however, the two sets of principles clash. For example,
let’s say that you’re using consistent subjects or topics to keep
your readers on track. In the following examples, sequencing or
using consistent subjects or topics led to a sentence with pas-
sive construction. Eagle-eyed readers may already have spotted
the passive construction lurking in one of the sentences in the fol-
lowing examples:

Example A:
Not surprisingly, the paragraphs that follow the document
head make up the document body. The document body can
be broken into many parts, but every paragraph you write in
the document body must contain a paragraph head and body.
Example B:
When you use and, also, or too toward the beginning of
a sentence, you tell readers that the next sentence adds to
the information they’ve read in the preceding sentence.
In contrast, when you use but, however, conversely, on
the other hand, or yet as transitions, you announce to
readers that the content of the next sentence will contrast
with what they’ve just read. In other words, the upcoming
sentence will hedge on or even overturn the claim from
the preceding sentence. Fail to use transitions, and your
readers may wonder how one sentence relates to the other.

In Example A, both sentences are tied together through sequenc-

ing the document body, which appears in the stress of the first
sentence and in the grammatical subject of the second – a good
example of strong sequencing. This same sequencing, however,
means that the second sentence uses passive construction can be
broken. Nonetheless, in this instance the trade-off between con-
tinuity and clarity works well, since the sequencing binds these
two sentences together tightly.
Now, in Example B, note that the consistent wording here falls
at the beginning of sentences but is not part of the grammatical
subject in any sentence. However, because this snippet of writing
relies heavily on transitions, the sentences seem continuous.
84 CONTINUITY: PUTTING SENTENCES TOGETHER

As you might recall, readers struggle to forge links between sen-

tences to grasp what you’re writing about. As a result, when you
occasionally must use passive construction to create a stronger se-
quence or more unified wording in your sentences, you’re making
the right decision. Just remember to try tweaking your sentence,
attempting alternate phrasing or sequencing to see if you can main-
tain continuity and also use active construction at the same time.
In most cases, you’ll discover, both continuity and clarity function
together seamlessly, without interfering with one another.

EXPERT TIP: Tying continuity together

Remember, you don’t need to use all the continuity principles every
time you write a sentence. While you need to always put important in-
formation in the emphasis position of every sentence, you can choose
from the other principles to tie your sentences together tightly. If you
use consistent grammatical subjects, you might not need to use a tran-
sition in every sentence, and you can probably skip sequencing for that
sentence. Likewise, if you sequence tightly, you can skip transitions or
consistent subjects for a set of sentences.

Takeaways for continuity: what you need to remember to write

paragraphs that really hang together
• Continuity trumps clarity.
• Place important information in the ends of your sentences.
• Use one of three linking strategies to tie your sentences
together:
• transitions at the beginning of the sentences
• sequencing of information in an familiar–unfamiliar,
familiar–unfamiliar pattern
• using consistent grammatical subjects or, at least, consist-
ent wording to refer to common topics in some sentences.
CHAPTER 5

Organizing paragraphs
and documents
The third C: Coherence

In this chapter you will

• master how to make documents seem tightly organized
• learn the best way to structure paragraphs
• understand when and how to break paragraphs
• discover how to recognize when a paragraph is too short or
too long.

When you write a paragraph, do you

a. just start with the first item on your list of things to get
through?
b. throw in a return and create a new paragraph, whenever the
one you’re working on begins to threaten to give War and
Peace a run for its money?
c. observe the principles used by editors at newspapers like the
Times-Picayune and USA Today and begin a new paragraph
every sentence or two?
d. cop to being guilty of all of the above?

If you answered “yes” to any of these questions, consider your-

self busted. And in bloody good company, since legions of other
writers do exactly the same thing. However, avoid giving in to
despair. Unlike the principles behind clarity and continuity, where
nearly everything you learned was wrong-headed (if haplessly
well intentioned), you’ll discover you have a nodding acquaint-
ance with the principles behind coherence. Provided you managed
to stay awake during English in secondary school, you’ve already
had some exposure to the fundamental principles underlying
86 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

coherence. Your teachers were actually within shouting distance

of practices psychologists and neuroscientists agree promote rapid
comprehension and good recall of content.

Context is everything

In writing, as in life, where you say what and to whom is

vital to whether you observe some of the coherence principles.
For example, readers of email invariably expect messages to be
dynamic, brief, and easily skimmed.1 When writing an email,
keep your message and paragraphs alike short – unless you’re
writing a chatty update to an old friend who’s happy reading
the kind of missive Henry James might have sent. To stay on
the right track when writing an email or web page text, picture
your readers stumbling through a Mato Grosso of sentences and
know that, unlike doughty Portuguese explorers, your readers
are liable to quit reading at the sight of a paragraph that crawls
over three pages.
In an email as with most online text, readers tend to scan
paragraphs quickly, focusing on the first few lines of the first
paragraph and, if they find the content relevant, the next few
lines of the second paragraph. After that point, readers gener-
ally bail on that email you spent so much time lovingly crafting
as if your mother would read every word. While your mother
might, however, your online readers won’t, as revealed by
eye-tracking studies of several hundred users reading online
text.2 In contrast, in a proposal for the US National Institutes of
Health, a three-line paragraph would seem a bit flyweight, un-
less you were introducing your study’s Specific Aims, usually
broken down into sub-aims in detail that would be eye-glazing
for everyone except (1) the study’s authors, who are True Be-
lievers, (2) employees of the National Institutes of Health, or
(3) eagle-eyed Subject Matter Experts, who can spot the ques-
tionable use of a particular assay at a glance.

And so endeth the good news about coherence. Now, brace your-
self for what Orwell’s Party Members in Newspeak would call the
 THREE MODELS OF COMPREHENSION 87

double-un-good news. Coherence trumps not only clarity but also

continuity.
In fact, coherence is even more important to readers than con-
tinuity or clarity. If you’ve ever needed to reread a paragraph,
you’ve likely blamed yourself. “I try to get at least four hours in
bed,” you think, or “Why am I attempting to read a manuscript
on neurosurgery before I go under the knife, anyway?” Howev-
er, as a reader, you’re rarely at fault, especially not after you’ve
launched yourself into higher education or the worlds of business
and industry, all of which require us to read reasonably expertly.
Instead, the likeliest culprit is the structure of the paragraph itself.
If you’re reading a paragraph that dives straight into the topic
at hand in medias res, as a reader, you’re left in the position of
someone watching a film without knowing its genre. If you don’t
know whether you’re watching a romantic comedy or a slasher
flick, you can’t determine whether you should go warm all over at
the prospect of romance lurking in the basement or whether your
sphincter should tighten before the heroine reaches the bottom of
the stairs. Without knowing what we’re reading about, we’re left
scrambling to figure out the plain sense of what we’re reading.

Drunk, sober, and savvy cinema-goers:

three models of comprehension

We might think of readers as being like a pair of drunks who

wander into a cinema, purchasing tickets purely to escape the
Arctic blast of winter air outside. Our drunks weave their way
into one of the shoe-box cinemas inside a multiplex without
the foggiest notion of what’s playing onscreen. As they stare
at the flickering images – provided they’re not so soused they
simply doze off or would fall over during a sobriety test – they
use what happens during each scene to piece together what
they’re watching. If a meat axe appears onscreen, they’ll ex-
pect great lashings of gore to follow. In contrast, the watch-
ers around them have forked over cash to watch a thriller, so
they realize the appearance of the meat axe merely augurs a bit
of heart-pounding excitement as the heroine dashes through a
crowd to evade villains.
88 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

Without quite realizing it, our audience of drunk and sober

cinema-goers represent one of the models researchers have con-
structed to account for the way readers comprehend written texts.
The drunks rely on bottom-up processing, piecing together cues
from moment-by-moment occurrences, seeking the schema that
will impose sense on the bloody thing. In the same way, read-
ers accumulate tidbits from words and sentences to work out a
paragraph’s main point.3 As we’ve already seen in our scrutiny
of continuity, readers fall back on this bottom-up approach when
they fail to make sense of the sentence they’re reading.
In contrast, the sober souls sitting alongside our boozy pair know
they signed up for a thriller, giving them a ready-made schema to
impose order on the events they see unspooling onscreen. The
seduction scene between our hero and heroine isn’t a step toward
the film’s end-game of happy romantic union, despite the attraction
of opposites or improbable odds that are the hallmark of a romantic
comedy. Instead, that make-out scene is just a scrap of local color,
the modern-day equivalent of Shakespeare throwing a bit of banter
between grave-diggers in between scenes in a tragedy to give au-
diences a breather from so much strong emotion awash onstage.
Whether you’re watching Three Days of the Condor or The Bourne
Legacy, you’re not suddenly bewildered by this swerve into romantic
territory. You know the kissing and sex are just a quick detour before
the ducking and shooting recur. In this way, you’re functioning ex-
actly as most expert readers do when they’re reading anything half-
decently written. As researchers discovered, readers rely on knowl-
edge of schemas gleaned from tens of thousands of documents to
enable them to immediately grasp the meaning of paragraphs and
entire documents. This top-down strategy lets readers identify help-
ful cues like headings and topic sentences that, in turn, speed read-
ing and comprehension further.4
Now imagine yourself in the cinema, watching Psycho on the
night the film debuted in cinemas. You haven’t heard from your
mother that she hasn’t been able to step into a shower since the
first time she saw the film in 1960. In fact, based on the early
scenes, you believe the entire film is going to be about an adul-
terous couple embezzling wads of cash. And, when Janet Leigh
climbs into the shower, you no more believe anything is going
to happen to her than you grasp that Martin Balsam is going to
 THREE MODELS OF COMPREHENSION 89

inexplicably take a knife through the forehead (a kitchen knife pit-

ted against a skull is a no-brainer of a non-contest), just when he
seems poised to put an end to whatever the hell is going on in the
spooky house behind the Bates Motel. Why? Janet Leigh has top
billing. Headliners never get knocked off midway through a film.
And the trail of embezzled cash we thought we were following
disappears when the car carrying it sinks, the money unseen in
its trunk. To make sense of this film on a first viewing, you either
needed to be (a) born well after 1960, (b) the second person you
knew to see it and given the skinny on what to expect before the
opening titles rolled, or (c) able to understand the way director
Alfred Hitchcock toyed with his audiences by making the object
you think matters turn out to be so trivial, the MacGuffin that
vanishes or is forgotten mid-film. But, if you were among those
first audiences to see Psycho and not make for the exits after the
shower scene, you would’ve used the third strategy readers rely
on for comprehending texts: a combination of both bottom-up and
top-down processing. With this strategy, readers use local detail
to fine-tune their sense of the big picture, then leverage the big
picture’s schema to make comprehensible the local details.5
We can take away one striking detail from this dust-up among
researchers who advocate for each of the three different models
of comprehension. Regardless of whether they subscribe to the
bottom-up, top-down, or interactive processing – a combination
of top-down and bottom-up processing – researchers agree on the
centrality of having a sense of a robust schema to make sense of
what we read. Even if we can tease out the meaning of individual
sentences, if they fail to add up to a coherent paragraph, we’re
bound to either reread the paragraph or throw up our hands in
bewilderment. (Rather like the early audiences who made it to
Psycho during the film’s opening week, before the spoilers got
out.) We understand what we read only because we can make
sense of the Big Idea governing the local content.6 So, after listen-
ing to me harping on about the importance of clarity and continu-
ity, you now have to constantly remind yourself that coherence
is more important than either of the first two Cs. At least I have
some consolation for you in this third C. The coherence principles
you’re about to learn function independently of both clarity and
continuity. In other words, you’ll never have to choose between
90 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

observing clarity and continuity on one hand and sacrificing co-

herence on the other to preserve the first two Cs.
Remember, the primary reading mechanism – prediction – applies
to coherence in spades.7 When you focus on coherence, you’re
providing readers with the all-important context they need to help
them understand what they’re reading and what’s coming next.
And strong coherence also helps your readers to slot what they’ve
just read into what they already know – the familiar–new strategy.
In short, if your documents lack coherence, your bosses or editors
are likely to return them to you so covered in red that, even with
Track Changes on a digital page, you might wonder who’s had the
serious nosebleed.

How do you write for distracted readers?

Journalists write under what most of us would like to see as a merci-

fully unique set of conditions. They end up writing for people with
wandering attention, who are giving cursory once-overs to news
stories while also paying attention to a host of other things, includ-
ing traffic and maybe the location of the commuter bus they’re rid-
ing on, not to mention droning colleagues, the blast of music from
someone’s ringtone, Facebook updates, and Twitter feeds bristling
with hashtags. While your readers may grapple with the same dis-
tractions, when you sit down to write a proposal or a legal memo,
you probably still naively picture your intended audience studying
your writing attentively like a student memorizing material for a
test. Fat chance. Instead, try picturing your audience sitting elbow-
deep in a stack of papers – or, worse, a queue of virtual papers that
could fill a modest-sized library – in an office where the phone
rings incessantly, reading snippets of your magnum opus while
fielding requests for signatures on incomprehensible documents
and emails from a department head enquiring into their last six
expense reports. Unlike a swotting student, your typical readers
are unlikely to hang on your every turn of phrase as though they’re
reading Ulysses for their doctoral oral exams.
Journalists compete with all this noise from the outside world
by breaking paragraphs frequently to hold readers’ attention – an
approach that similarly can help most of us hold our audience’s
 HOW DO YOU WRITE FOR DISTRACTED READERS? 91

attention. But these relatively brief paragraphs also ensure that

readers can easily keep track of ideas and relationships between
ideas, events, and characters. These relationships are easiest to
track when paragraphs each contain only a single focus, topic, or
angle on a topic. Conversely, when you pack lots of content into a
lengthy paragraph that runs on for several pages, you’re practical-
ly guaranteeing that your readers will emerge from that paragraph
with only the vaguest notion of what you’ve been writing about.
And your readers will probably feel close to punch-drunk, to boot.

Why is brief usually better?

Cast your mind back to Chapter 4, where we focused on why long
sentences were so horribly difficult to read and recall. Remember?
Long sentences have few opportunities to flag important items for
readers, since a long sentence contains the same number of op-
portunities to emphasize information as a short sentence – one.
Furthermore, in long sentences, readers will only recall material in
the emphasis position at the end. And, if you’re lucky, your read-
ers will recall details from the beginnings of sentences, while their
grasp of the relationships between items is likely to be tenuous, at
best. If you find yourself unable to recall the continuity principles
about emphasis and using more short than long sentences, page
back through Chapter 4 and reread them. We’ll still be here.
These coherence principles also apply to your readers’ abilities
to recall content in paragraphs. Coherence principles don’t neces-
sarily require you to start fresh paragraphs every three sentences
or to ensure that you tackle only simple, easily developed ideas
in your paragraphs, things you can safely cover in a handful of
sentences. On the other hand, in breaking your paragraphs, you
should be attentive to even minute shifts in focus on a topic. For
example, while this paragraph still deals with the topic of para-
graph length, I’ve used a new paragraph to accommodate a shift
in focus from the preceding paragraph, which dealt with the rea-
sons why long paragraphs are difficult for readers to understand.
Both paragraphs still technically cover the same topic – paragraph
length – but each paragraph offers a different focus on paragraph
length. By using a series of relatively brief paragraphs to chart
shifts in focus, as well as shifts in topic, you will help your readers
organize the information they’re absorbing.
92 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

Your readers can perceive your paragraphs as resembling

either a stand of woods or a wilderness. Even if you’ve used con-
tinuity principles to mark out a series of clear paths for them,
your readers will generally be happier making their way through
a small wood instead of a vast, seemingly unbroken wilderness.
Shorter paragraphs – like small stands of trees – mean less cogni-
tive overhead for your readers. And frequent changes in focus will
also help maintain your readers’ interest – no small thing in the
age of 140-character microblogging and 3-second sound bites.

Coherence Principle #1:

Begin each paragraph with a set of comprehensive
overview sentences, a paragraph head

If you suffered through band in middle or high school, you probably

know what an overture is – a piece of music that contains snatches
of musical themes that are expressed more fully later in an opera
or symphony. Think of the paragraph head as your overture, the
sentence or series of sentences where you provide your audience
with an overview of what you’ll cover in detail in the paragraph. An
overture helps situate your readers, previewing for them the claims
or topic you’re about to address. This preview also provides them
with a detailed set of expectations, giving your readers the best
possible set of blueprints to guide them through even pernickety de-
tails, twists of logic, or barrages of statistics that may turn up in the
rest of the paragraph. Put simply: your paragraph head promises;
your paragraph body delivers – an analogy that functions a bit like
the way a successful proposition in a club might.

Writing myths you’ll wish you never learned:

Begin every paragraph with a topic sentence

I once had a graduate student who referred to topic sentences –

then required of high-school students on the New York State Regents
exam – as the “spawn of hell.” She loathed having to teach topic
sentences to her students. But she couldn’t avoid them because her
 USE PARAGRAPH HEADS [COHERENCE PRINCIPLE #1] 93

students had to write paragraphs using topic sentences to pass their

Regents exams and get out of high school. However, she hated them
for the same reasons we all have loathed them. Topic sentences are
nearly impossible to write.
Why? For starters, unless your paragraph is very brief or barely
has two ideas to rub together, you’ll need more than a single sen-
tence to provide a comprehensive overview. You’ll experience pure
torture when you try to shoehorn an introduction to a new topic,
as well as an overview of your central ideas or claims, into a single
sentence. This feat is also next to impossible. So avoid busting a
gut to cram everything into a single sentence, unless you’re dealing
with a short paragraph on a relatively uncomplicated topic.

The good news about writing a paragraph head rather than

a topic sentence: you can stretch your legs a bit. Think blogging
rather than tweeting. If you need more than a single sentence to
provide a comprehensive overview of your paragraph’s content,
write two or three. Just ensure you begin each paragraph with the
head, where readers expect to see an overview of the paragraph’s
central point.8
Paragraph heads focus readers’ attention and help them to
track and even recall content effectively for three reasons: (1) by
priming readers, (2) by giving them linguistic cues about content
that enable readers to generate inferences to make sense of what
they’re about to read, and (3) by providing explicit indications
about the relevant content to follow. Priming, as we saw in spades
in Chapter 4, nudges readers to recall content on a second encoun-
ter with it. And, as we saw in Chapter 2, readers engage in infer-
ence processing to make sense of content. When I. A. Richards de-
prived his undergraduates of the titles and authors of poems, they
foundered, waffled, and generally waxed uncomprehending about
what the hell the poem was about. Fancy reading the following
scrap of text without knowing its author or the poem’s title.
Thou still unravish’d bride of quietness,
Thou foster-child of silence and slow time,
Sylvan historian, who canst thus express
A flowery tale more sweetly than our rhyme:
94 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

What leaf-fring’d legend haunts about thy shape

Of deities or mortals, or of both,
In Tempe or the dales of Arcady?

Who, in God’s name, is this thou, which the seven lines of de-
scription that follow this segment make more, rather than less,
puzzling? Think back to reading’s three-step dance. We can make
sense of the words – sort of. But still unravish’d bride of quiet-
ness immediately throws down a gauntlet. Is the poet punning
with still and quietness or is still telling us that the bride is a virgo
intacta, still hanging onto her hymen? Since we know from the
arrangement of the stanzas that we’re reading poetry, we’re will-
ing to hang fire to some extent on the role each word plays syn-
tactically. But, as the problem with how to interpret still reveals,
recognizing what role a word plays in a sentence grammatically
and what the word means are as inextricably bound together as a
virgin is to her hymen.
Even readers who manage to stumble through to the last stanza
end up remaining mostly in the dark about the subject the poet
addresses, despite these last lines laying things out a bit more
concretely than the waves of metaphor that wash over us in the
opening stanza:
O Attic shape! Fair attitude! with brede
Of marble men and maidens overwrought,
With forest branches and the trodden weed;
Thou, silent form, dost tease us out of thought
As doth eternity: Cold Pastoral!

The capitalization of Attic helps a bit, if we know the adjective

refers to something that harkens back to the Athens populated
by the likes of Socrates, rather than the box-cluttered space at the
top of your house. Yet, even a gifted reader – or one of Richards’
undergraduates studying English literature at Cambridge, no less –
ends up drowning, not waving, if you were to thrash out the poem’s
meaning sans title or author. But tell us that we’re reading Keats’
“Ode on a Grecian Urn,” and we read the stanzas as descriptions
of a scene arrested for millennia on the side of a Greek vase with-
out feeling as though our brains had just completed an Ironman
competition.
 USE PARAGRAPH HEADS [COHERENCE PRINCIPLE #1] 95

While poetry represents challenging territory for interpret-

ation, you can nevertheless see why page editors of the likes of
Vanity Fair as well as the Washington Post have long held dear the
importance of a good headline – juicy or not. Consider the quan-
dary of your average reader of the New York Post, the newspaper
even New Yorkers love to scorn. The paper will, of course, con-
vey details of everything from the latest grisly crime committed
in the city’s five boroughs to a heart-warming feature on cops in
Chinatown receiving sensitivity training, to the triumphs of a pint-
sized chess champion. Readers need to determine which stories
they read closely, which, they skim, which, they will skip without
more than a glance. In these millisecond-long bouts of decision-
making, headlines equip them with vital cues about the poten-
tial value each column contains. Moreover, the pages themselves
strategically segregate stories: big stories on the front page, local
news tucked inside, features relegated to pages in the center of the
paper. Your digital edition of the typical newspaper appears, at a
glance, infinitely more democratic. Rather than relegating local
features to the unread ghetto of pages buried far inside, the digital
edition, instead, can put an abbreviated table of contents for the
entire paper on a single page. But even this democratized, flat-
ter playing field privileges only the stories editors judge likeliest
to gain a wide readership. Buried, mostly unread, beneath those
leading stories lie the stories that formerly languished mostly un-
read in the paper edition.
Furthermore, the vast quantity of news contained in national
newspapers sprawls across areas of interest, as well as nomenclat-
ures specific to politics, real estate, finance, and the arts. Here, too,
headlines provide invaluable guides through what would other-
wise seem a wilderness of sometimes enigmatic details. If the
average newsreader were to consume the entirety of the behemoth
that is the New York Times Sunday edition, a paper a British friend
once mistook for an old-fashioned phone book, he or she would
encounter everything from quantitative analyses of market fluc-
tuations on Wall Street to speculation on how a Roseate Spoonbill
turned up in Staten Island. A headline tells readers at a glance
(a) whether they even want to read the story beneath it and (b) what
to expect from the story. The headline performs some of the heavy
lifting of inference building. At the same time, headlines clearly
96 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

delineate the scope of the story to follow, particularly headlines

for what journalists refer to as hard news stories, pieces that con-
vey developments to the audience in terms of proximity, personal
relevance, and the importance of the characters involved.9 In fact,
news stories have three components that give readers a leg-up in
inference building: a headline, a lead, and what journalist call the
nut graph. The lead, or lede as the New York Times likes to spell
it, is the short, ostensibly compelling hook that snags readers’
interest. In contrast, the nut graph is a telegraphic, one- or two-
sentence representation of the story’s gist, the journalist’s equiva-
lent of a thesis sentence. Together, the headline, lead, and nut
graph summarize the content of the text and provide readers with
a sense of its scope.10

Can skimming be smart reading?

Despite everything your teachers told you, sometimes faster is

better. In one study, readers skimmed a text at 500–700 words
per minute, compared to the less brisk pace of close reading
at approximately 225 words per minute. Although this study
emphasized longer texts of 3000 words, making recall more
challenging, subjects who skimmed the sample answered gen-
eral questions as accurately as readers who read every word. In
addition, eye-tracking devices established that skimmers spent
longer periods – albeit milliseconds – looking at sentences and
lines containing important information, skipping some words
and phrases entirely. Skimmers used early lines in paragraphs
and texts to grasp the structure of what they were about to
read, enabling them to read more efficiently, including skipping
entire lines.11

Can you use more than one sentence in a paragraph head?

Depending on the complexity of your paragraph, you can use
one, two, or three sentences in your paragraph head. Just re-
member to keep your paragraph head to less than a third of your
paragraph’s length. In that study you just read, above, about
skimming readers, researchers discovered that skimmers (in the
 USE PARAGRAPH HEADS [COHERENCE PRINCIPLE #1] 97

reading, not hand-in-the-till sense) distributed their focus more

intensively on the beginnings of paragraphs. This fleeting bit of
scrutiny might stem from our conventional expectations that the
openings of paragraphs tell us to expect a paragraph on skyrock-
eting property values will focus on prices per square foot and
increased market demand, rather than, say, the grazing habits
of gnus.
In studies of readers in both primary and higher education,
readers expertly identified anomalous sentences quickly and ac-
curately when paragraphs began with one or more overview sen-
tences.12 In most paragraphs, your paragraph head will occupy
from one to three sentences. As long as these sentences appear
at the outset of the paragraph, readers will have no difficulty in
identifying more than one sentence as the paragraph overview.13
Generally, you’ll need a sentence to introduce your readers to your
topic or shift in focus. Then you might need a second sentence
to sketch out your claims. If your paragraph is complicated and
long, you might even need a third sentence to lay out your claims
and their limitations. In studies examining the effects of topic
sentences, researchers identified the opening sentence of paragraphs
as holding a privileged position for readers, indicated by the speed
with which they used the opening sentence to identify the para-
graph’s topic. The takeaway from the opening sentence as the
paragraph’s most memorable: the farther your paragraph head
stretches from the outset of the paragraph, the less likely your
readers’ recognition of the subsequent sentences as announcing
the paragraph’s topic.14 As a result, your head should occupy only
a third of the total length of your paragraph.
Studies, using eye-tracking devices, of readers’ focus on the
introductory snippets of web pages discovered similar eye fix-
ations on the opening sentences of web pages. However, with a
single overview sentence, readers performed far better than they
did with multiple overview sentences. The greater the number of
overview sentences – paragraph head sentences, for our purposes –
the more time readers’ eyes lingered over them, and the poorer
their recall.15
As you read the following paragraphs, one written by a gradu-
ate student, pay attention to how long you keep reading before
you latch onto its central point:
98 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

Example A:
For the large competitors such as PricewaterhouseCoopers,
Deloitte, and the others, their biggest strength is their name
and the resources available to them. These companies are
able to enter almost any market and to provide employees
with a particular expertise in the client’s subject matter.
The major weakness, however, exists in these companies’
overhead and the lack of specialization in a particular area.
With resources spread all over the country and world these
companies can find costly and time-consuming the effort of
bringing the right employees together. This location issue
and high overhead costs translate into high billing rates.
These high billing rates of the “Big Four” consulting firms
are heavily watched by the senior executives in our client’s
firm. As a result, we find it often difficult, even impossible,
to sign long-term contracts with our clients because we are
one of the “Big Four” and such an engagement would set off
red flags with the senior executives. Our smaller competitors
have realized these weaknesses and have designed business
models to get around many of the issues faced by the largest
competitors in the industry.
Example B:
The size and reputation of the “Big Four” accounting
agencies can actually work against them where clients
are concerned. With resources scattered nation- and
world-wide, Big Four agencies often expend substantial
costs merely bringing the right employees together.
Unfortunately, these location and overhead costs translate
into high billing rates. As a result of these costly rates,
many clients seek out smaller firms, since senior executives
would view contracting with PricewaterhouseCoopers,
Deloitte, and others as needlessly costly. On the other hand,
our smaller competitors have realized these weaknesses
and designed business models to take advantage of them.

Note Example A’s problem with coherence. By the time readers

have ventured as far as the third sentence, they’re through with
lifting vines and checking their compasses to determine the dir-
ection of the trail through the jungle ahead. Readers expect that
 USE PARAGRAPH HEADS [COHERENCE PRINCIPLE #1] 99

you’ve set them on a fairly straight path through the territory

ahead. Instead, in Example A, the writer probably uses the first
two lengthy sentences to segue from the preceding paragraph and
only introduces the actual focus of the paragraph in the third sen-
tence. By this point, even the most stalwart of readers will have
abandoned all searches for the paragraph’s topic and will stagger
forward, armed with a thoroughly lousy map of a paragraph that
dwells on the strengths of large US-based accounting consultan-
cies. Unfortunately, the paragraph actually focuses on the poten-
tial liabilities of these behemoth consultancies. Net result: your
reader trudges back through this paragraph, but likely only if he
or she is the writer’s workplace inferior. If the writer is a bottom-
feeder in her organization’s food-chain, she would likely have
ended up looking for another job after cranking out forty or fifty
paragraphs like these. In contrast, in Example B, the paragraph
head primes readers to batten onto the paradoxical proposition
that bigger is not invariably better – a paradox that might have
been slightly modified by the 2008 “Too Big to Fail” US financial
debacle. By the end of that single sentence, readers will identify
every sentence that follows it, on locations, on overhead, and on
the resulting competitiveness of smaller firms, as clearly support-
ing that paragraph head.
The upshot: if your paragraph head is longer than a third of
your paragraph’s length, your readers might be confused about
where your paragraph head ends and its body begins. Readers
tend to recall content at the beginnings of things relatively well. As
a result, front-loading your information ensures your readers will
actually recall your paragraph head even once they’re hip-deep in
statistics or examples. But when you make your head longer than
several sentences, you almost inevitably end up placing your most
important sentence – the one that lays out your important claims –
in the middle of the paragraph. And the middles of anything are
veritable Dead Zones in terms of your readers’ ability to recall con-
tent. These Dead Zones include sentences, paragraphs, and entire
documents, as you’ll recall from Chapter 4.
In most cases, you can preview for your readers the contents of
even fairly lengthy, complex paragraphs in only a sentence or two.
As with paragraph length, remember that shorter is better. If you
end up sketching out a paragraph head that seems even to you to
100 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

be horribly complex and hopelessly lengthy, try dividing the para-

graph into two, separating the contents according to slight shifts
in focus or topic. Your readers will unconsciously thank you for it.
Or at least they’ll emerge from reading your documents without
feeling as if they need a whacking good dose of Ritalin and a cou-
ple of hours of rereading to grasp what you were probably saying.

Spot the difference

Still skeptical about the utility of a paragraph head (or, if you’re
stubbornly Old School, the topic sentence)? Compare the ease
with which you read the following two paragraphs. Example A
comes from the first paragraph of an undergraduate’s research pa-
per on Irish emigration. Specifically, his manuscript focused on
the role played by Irish emigrants who departed for Europe, aim-
ing to enlist assistance in freeing Ireland from English rule – and
on their descendants’ political influence and widespread cultural
legacy throughout Europe. Example B represents part of a gradu-
ate student’s feasibility analysis of a business plan, written for an
audience of potential investors who are mulling over the prospects
of a small company that proposes to offer training to company
employees on common software applications.
Example A:
Throughout Irish history, countless generations of exiles
have cast off from the shores of Ireland in search of
sanctuary, adventure and alliances. After the signing of the
Treaty of Limerick in 1691, the “Flight of the Wild Geese”
marked the watershed moment of Irish migration to the
European Continent. Nineteen thousand Catholic Irishmen
and their families sailed to France in the hope of one day
returning to free Ireland from English tyranny. For the next
hundred years ships arriving on the west coast of Ireland
would depart with Irish recruits listed in ships’ manifests
as “Wild Geese” for the armies of Europe. Using this label
to escape detection by English authorities, these Wild Geese
would serve in armies across the Continent, including
Napoleon’s. Ultimately, these Irish soldiers and their
descendants would fan out across Europe and eventually
influence power, politics, and business throughout the
Continent well into the twentieth century.
 USE PARAGRAPH HEADS [COHERENCE PRINCIPLE #1] 101

Example B:
The “hands-on” approach with personal instruction
certainly appeals to the typical student. Although labeled
as innovative, the approach outlined by Mr. Hutton details
no new techniques for training that differentiate ITS
from competitors. With 15 students anticipated per class,
Mr. Hutton’s ability to provide individual attention to a
particular student will be constrained. Also, the business
plan fails to make clear the range of services ITS will
provide. Furthermore, small businesses may be unwilling to
invest in training for standard software applications, like
the programs in Microsoft Office, since most entry-level jobs
now require these skills of would-be employees.

In Example A, the underlined paragraph head announces the para-

graph’s main topic. The paragraph head, with its focus on Irish
emigration, swiftly zeroes in on a particular wave of emigrants,
listed as “Wild Geese” on ship manifests. The paragraph body
then delivers on only the promise of the head sentences, fleshing
out the roles, destinations, and influence of the Wild Geese. Note,
also, that the paragraph culminates in a sentence that serves as
the paragraph’s conclusion and also serves as the thesis sentence
for his entire manuscript – both features explored later in this
chapter. In this paragraph, the writer has expertly escorted readers
through material that supports a nuanced and relatively complex
point, rather than leaving readers the difficult task of determining
what on earth the paragraph is about and how its content contrib-
utes to the writer’s overall argument.
In contrast, however, Example B has no unifying statement in
the paragraph head. Instead, the opening sentence simply repre-
sents a place to begin the paragraph. As a result, the paragraph
covers five different points:
1. the business has an innovative, personal approach to deliver-
ing instruction on computer applications to students;
2. the approach is not as innovative as its instructors would like
investors to believe;
3. the instructor-owner will have only a limited ability to
provide hands-on, one-on-one training to classes of fifteen
students or more;
102 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

4. the business plan omits any mention of precisely what kind

of training ITS will cover;
5. many businesses require employees to have basic knowledge
of common applications, potentially nixing ITS’ attractiveness
to employers.

EXPERT TIP: Organized paragraphs

Fortunately, we have an easy solution to avoiding sounding like the
writer in Example B. You can address this difficulty of both marshaling
your main points and writing them down in an order that makes sense
to your readers by using four strategies.

1. Try sketching out your points for your paragraphs on a scrap of

paper before you begin writing each one.
2. Write a brief sentence or two that cover the contents of those
points.
3. Work on your paragraph heads in even informal modes of writing,
like email.
4. Focus on identifying paragraph heads whenever you read any form
of published writing.

These strategies will also help you write tightly constructed, well-
organized paragraphs, no matter how difficult your topic or complicat-
ed your points.

Creating a good paragraph head

If you write, think eternally of the mantra, “Prediction is the engine
that enables comprehension.” Just as we saw the power of predic-
tion in all matters concerning clarity and continuity, prediction is
likewise a heavy-hitter in coherence. The more specific the details
you give readers in your paragraph head, the faster their reading
and better their recall of the details to follow.16 So you should avoid
saying This business plan has particular promise, if the management
team addresses several issues concerning its target market. Instead,
specify the number of reasons why the business plan might not
be the equivalent of playing slot machines in Vegas: This business
plan has particular promise, if the management team addresses four
issues concerning its target market. Your readers can now track the
issues, via helpful transitions like First . . . Second . . . Third . . . Finally.
 SUPPORT EACH PARAGRAPH HEAD WITH A BODY 103

Caveat: avoid looking ahead to future paragraphs

A good paragraph head, like a soundly constructed paragraph,

focuses on a single topic and only on the content of the imme-
diate paragraph. Why? Your readers’ working memory is less
like a bottomless pit than it is a shoe-box – our working mem-
ories can only hold a limited number of propositions at one
time. Stuff more than a single pair of shoes into any shoe-box,
and the lid comes off, then the shoes tumble out, a reasonably
apt analogy for your readers’ brains when you introduce three
propositions crammed into a single paragraph.17 In addition,
remember that your paragraph head primes readers for the con-
tent to follow and also provides a structure for your readers to
hang their inferences on as they work out your paragraph’s
meaning. If you mention content that crops up in another
paragraph, readers may believe they’ve missed something in
the paragraph body.18 As a result, they’ll reread the paragraph
head and puzzle anew over the paragraph content. Research-
ers performed this particular trick in experiments on hapless
undergraduates, in which researchers deliberately mismatched
introductory sentences and the content that followed them. As
a result, students’ reading times increased significantly.19
In short, think of writing a paragraph head as akin to driv-
ing. You’re forever looking approximately one to three cars
ahead of you, not constructing models of what the traffic and
road will look like in a half-mile. At least not if you want to
avoid mowing down a few pedestrians and skateboarders in
your immediate vicinity.

Coherence Principle #2:

Support each paragraph’s head with a body

As in anatomy, the body supports the head – and also occupies a

lot more space than the head. Your paragraph body should be about
three times the length of its head and flesh out the concepts or claims
introduced in the head sentences. Your paragraph body should also
use examples, well-established facts, analogies, statistics, quotations,
104 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

and other forms of evidence to make credible the statements you’ve

trotted out in your paragraph head.
Again, remember, you’re not writing for the harried readers of
USA Today who prefer their paragraphs bite-sized and flattened to
a single concept. Even if you’re writing a blog post, your readers
need a bit of detail to believe the claim you’ve just made in the
paragraph head. The paragraph body not only substantiates the
claim rolled out in the head, the body also lends you a bit of cred-
ibility by providing a sense that you’ve done your homework and
can shore up your main contention with evidence.
For all the wealth of research on priming and primacy effects
from paragraph heads, research on the body itself is surprisingly
thin. In addition, researchers have traditionally focused more on
the ways arguments unfold within paragraphs, specific to differ-
ent kinds of arguments, rather than on the number of sentences
necessary to redress readers’ ignorance or skepticism about the
content in the paragraph head.20 Nevertheless, researchers have
determined that, as readers work their way through paragraphs,
their speed increases when the sentences fulfill expectations
established in the paragraph’s opening sentences.21

Writing myths you’ll wish you never learned:

Point-last paragraphs

If you never learned this particular approach to turning out para-

graphs, feel free to skip ahead. However, stop right here if you read
Joseph Williams’ mostly useful Style: Toward Clarity and Grace.
Williams was a pioneer in exploring the components of clear writ-
ing and in debunking myths about grammar and usage. But he also
provided his readers with some terrible advice on what he called
“point-last” structure in paragraph bodies or “discussions” to use
Williams’ parlance.22 In point-last paragraphs, writers put readers
through a Lewis-and-Clark sort of expedition into the meaning of
the paragraph, so that we discover the paragraph’s topic the way
the explorers stumbled across the Continental Divide, except your
readers lack the assistance of Sacagawea as a trusty guide. Unfor-
tunately, when you oblige readers to replicate your own process of
 SUPPORT EACH PARAGRAPH HEAD WITH A BODY 105

thinking about the topic, you cast them completely adrift, without
the cues that enable them to build sturdy inferences.23
Remember, most of us write to facilitate rapid and easy under-
standing of complex topics, not to earn a Nobel in literature, a
Pulitzer on feature writing, or an award for arty non-fiction. A
paragraph trots out a claim, then shores it up. You’re also not writ-
ing an episode of Murder, She Wrote, where, by telling your readers
up front what to expect, you completely kill both the suspense and
one’s reason for watching the bloody episode. Writing for enter-
tainment: suspense, good. Writing a murder mystery: suspense,
absolutely de rigueur. Writing a prospectus, a cover letter for a new
job, or a letter of resignation: suspense, positively bloody deadly.

Now, imagine the poor sods struggling through a paragraph in

the London Review of Books in October 2001, in a short commen-
tary by one of the myriad of commentators writing on the events
of September 11th:
Historical events, however, are not punctual, but extend
in a before and after of time which only gradually reveal
themselves. It has, to be sure, been pointed out that the
Americans created bin Laden during the Cold War . . . and that
this is therefore a textbook example of dialectical reversal. But
the seeds of the event are buried deeper than that. They are to
be found in the wholesale massacres of the Left systematically
encouraged and directed by the Americans in an even earlier
period. The physical extermination of the Iraqi and the
Indonesian Communist Parties . . . were crimes as abominable
as any contemporary genocide. It is, however, only now that
the results are working their way out into actuality, for the
resultant absence of any Left alternative means that popular
revolt and resistance in the Third World have nowhere to go
but into religious and “fundamentalist” forms.24

Floundering a bit, after reading that paragraph? Or, if you’re hon-

est, at what point did you bail on the paragraph entirely? Or, for
the more anally retentive among us, how many times did you
reread the sample? If you fall into the last category, you only stuck
106 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

with the paragraph out of diligence, masochistic tendencies, or

willingness to believe that the paragraph will eventually make
sense if you only keep on reading. Nevertheless, the likelihood is
close to nil that you got much sense from the paragraph, which
unfortunately is no easier to read if you got the whole shebang.
All four paragraphs of it.
What’s wrong with this paragraph, aside from every sentence
violating nearly all the clarity principles we’ve already explored?
First, the paragraph head contains a however that immediately
cues readers that the content to follow is about to refute the con-
tent of the preceding paragraph. But the preceding paragraph (not
excerpted here) mentions neither history, nor historical events.
Instead, the earlier paragraph focuses on media reports of Ameri-
can flag-flying in the days following the terrorist attacks in New
York and Washington, DC on September 11, 2001. In addition, the
likeliest culprit for the paragraph’s actual topic is buried in mid-
paragraph:
But the seeds of the event are buried deeper than that.
They are to be found in the wholesale massacres of the Left
systematically encouraged and directed by the Americans in
an even earlier period.

The paragraph has three problems tied directly to this paragraph

head. First, the topic of the paragraph, the one that should have
cued readers about the content at the outset, occurs several sen-
tences after readers traditionally expect the writer to have clearly
stated the paragraph’s topic. Second, the paragraph shifts its head
from the beginning to the dead zone that falls between primacy
and recency effects and thus tends to have the weakest recall by
readers.25 In fact, priming effects apply even to novel material,
rather than the already familiar material used in the design of
many experiments on memory and recall.26 If you’re an attorney,
you should thank the human brain for having a dead zone that
enables you to bury in the middle of a lengthy set of conditions
a particularly nasty disclosure guaranteed to freak your readers
out – like the fact that a passenger runs a slender risk of being
decapitated while on a world cruise for which they’ve forked out
over tens of thousands of dollars. But the rest of us should view
that paragraph’s dead zone as strictly the spot to shore up the
 SUPPORT EACH PARAGRAPH HEAD WITH A BODY 107

propositions in the paragraph head. Third and finally, the sheer

number of propositions crammed into the excerpted sentence can
make the reader feel punch-drunk, especially without the handy
continuity devices we explored in Chapter 4 that keep each sen-
tence tightly bound to both the preceding one and to the para-
graph’s overall theme. With each shift in topic, readers struggle
increasingly hard to keep track of the paragraph’s trajectory27 at
the same time the shifts in topic run bang up against the limits of
even skilled readers’ working memories.28

Writing myths you’ll wish you never learned:

Keep it short

Remember: you need at least one sentence in your paragraph head

to introduce your readers to your new topic and to trot out your
main claims. But, since your paragraph body needs to support its
head, your paragraph body should be at least two to three times
the length of the head. To shore up claims made broadly in your
paragraph head, you need to introduce examples, analogies, statis-
tics, quotations, and other forms of evidence. In studies of readers,
researchers identified readers’ expectations that paragraph bodies
provide generalizations or support for the paragraph head.29 At the
barest minimum, that requirement means that you need at least
a one-sentence head and two sentences in the body for even the
briefest paragraph. Anything shorter than two sentences – in any
form of non-fiction writing outside journalism – is more an ampu-
tated piece of a paragraph than a full-fledged version.

One final caveat on paragraph bodies

Never introduce a topic in your paragraph body that you haven’t
mentioned or covered in your paragraph head. The shift in topic –
even just a shift in focus on a topic – can confuse the hell out
of your readers, who will absorb information most easily if they
have a clear set of expectations for the paragraph. The paragraph’s
opening sentence or sentences provide readers with a schema for
understanding the paragraph. As a consequence, if you miscue
readers about content, their reading speed declines and their recall
108 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

of content gets fuzzy. One particular study about prediction and

reading makes me long for the 1960s, when employees chain-
smoked in their offices and drank Scotch midday. In that memo-
rable study, Bell Labs paid secretaries who knew nothing about
chess to read about the game and earn the princely sum of $7.50
if they correctly recalled the moves associated with each chess
piece. In 1969, that $7.50 would’ve probably bought the secretar-
ies – then predictably a veritable redoubt of estrogen – at least
a tank of gas. When the paragraphs used the opening sentences
to cue readers about the characteristics of each chess piece, the
secretaries displayed significantly better recall about bishops and
knights. In contrast, when the paragraphs contained no paragraph
head to prime readers about the content to follow, secretaries’
recall was poorer. And, when a single paragraph contained sen-
tences that jumped from details about pawns to knights to bishops,
secretaries performed still more poorly in recalling the characteris-
tics and moves for each chess piece. We can guess this last group
of secretaries failed to take home an extra $7.50 that week.30
If you have reread one of your own paragraphs and discover
that even you find its information difficult to absorb, carefully
check the contents of the paragraph body. You’ve probably intro-
duced a shift in topic or a change in focus that ranges outside the
scope of your paragraph head. If you can locate a place where
you can easily spot the shift in focus, break the paragraph at that
sentence. And, if necessary, craft a new paragraph head to clearly
anticipate the full scope of the new paragraph. Just remember not
to cut the paragraph off at the knees – or the paragraph head, a
bad practice with people, equally deadly for paragraphs.

Coherence Principle #3:

Documents need heads and bodies. Apply paragraph head
and body organization to your entire document

You should organize any document the way you would handle a
paragraph. Like paragraphs, documents need heads and bodies
to aid readers in inference building. Without a suggestion of the
main point(s) your document makes, readers spend more time
straining to figure out the macrostructure or overall argument
 APPLY HEAD AND BODY STRUCTURE TO DOCUMENTS 109

contained in the document. At a result, readers fail to grasp what

they’re reading, because their brains are busily shunting cognitive
resources to figuring out what they’re reading.31
For your entire document, document heads are introductions
to the overall content. These document heads may occupy one
or more paragraphs. Exactly as with paragraph heads, document
head paragraphs front-load your important information, inform-
ing readers what they can expect in the paragraphs to follow. In
addition, these front-loading paragraphs also provide your readers
with a blueprint that helps them in making predictions as they
read. And the better informed your readers’ predictions, the easier
the reading process and the better your readers’ grasp and recall
of your content in your text.
Unlike an abstract, a document head simply cues your audi-
ence of the overall scope of your proposal, article, or report. For
example, the document head for a research article tells readers
why the topic is important and worthy of research, prevailing wis-
dom on the topic, and how your research will redress gaps in prior
research. In contrast, a document head for a feasibility analysis
of a business plan introduces would-be investors to the business’
product or service, its target market, potential competitors, and
how the business will (or won’t) prove profitable. In a second or
third paragraph of a feasibility report, writers should explicitly tell
readers whether they recommend funding the business or whether
investors should steer clear. Remember, your would-be investor
is hardly expecting Murder, She Wrote and thinking your recom-
mendation will be a lovely surprise . . . at the end of the analysis.
Instead, investors want to know soon after discovering the basics
about the business whether you recommend funding it or not.
When readers lack cues about the central hypothesis or conclu-
sion of your document, they also lack the knowledge of which
details they should recall or even how to interpret the content
they’re about to read.
Like your paragraph head, your document head should occupy
several paragraphs for longer documents, perhaps only a single
paragraph for shorter texts under three pages. Longer documents
like proposals or research papers might have document head para-
graphs that occupy as many as three pages. However, unlike the
paragraph head, the length of your document head doesn’t have
110 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

to take up about a third of the document’s total length. Instead,

you should focus most intently on covering the full scope of the
content of your document in the document head paragraphs.
Document heads should have three components. First, use a
series of sentences introducing your readers to your topic, em-
ploying the same wording you’ll rely on in the document overall.
Second, your document head should contain specifics about what
your document will cover: the most central points you make in
the paragraphs to follow. And, third, contrary to everything you’ve
heard about not repeating yourself when you write, repetition ac-
tually benefits your reader. When we encounter familiar words,
particularly ones we’ve recently encountered in the same docu-
ment, we read more rapidly. In addition, during reading, we also
dedicate most of our working memory for recalling the concepts
we’ve just read.32

Avoid Big Bang Beginnings

Like many writers, you’re probably tempted to begin with the his-
tory of what you’re writing about, an extensive rationale for the
action you’re advocating, or a narrative of how you got where you
are. However, as far as your readers are concerned, you’re begin-
ning with the Big Bang, crawling through the Cambrian through
the Pleistocene and gradually working your way to the beginning
of human history, before finally getting to your topic. The entire
time they’re reading, your readers are either thinking “For God’s
sake, get on with it” or “What on earth am I reading?” And these
scenarios are the positive ones. In the worst-case scenario, your
readers simply bail, your document going almost entirely unread.
The takeaway: skip the backstory. Your readers seldom want to
hear a history, rationale, or narrative until they already know the
scope and purpose of your document.

Coherence Principle #4:

Place your thesis at the end of the head paragraph(s)

At last! You probably thought you’d never reach a familiar term,

something you could recognize from earlier writing instruction. In
this particular instance, your secondary school teachers who once
 PUT YOUR THESIS AT THE END OF THE HEAD PARAGRAPH 111

ranted about the importance of a thesis were absolutely on the

money. Your thesis is the single most important sentence in your
entire document.
Here, again, you don’t need to take either my word or that of
the martinet who gave gimlet-eyed scrutiny to your papers, bent
mostly on pinpointing your thesis (or lack thereof). A thesis exerts
a strong priming effect, just like paragraph heads. Moreover, any
piece of writing with an identifiable thesis also speeds reading
times and results in better recall than documents that contain
none. In one study, readers confronting a news story – which al-
ways contains a thesis via its nut graph – read more rapidly and
constructed better representations of the story’s overall structure
than did readers of a literary story, which inevitably omits any-
thing resembling a thesis.33 We can think of a thesis as a schemat-
ic cue that prevents us from glomming onto trivial details and also
ensures we don’t make wrong inferences about what we’re read-
ing.34 Just as our understanding of a taxi schema stops us from
asking the driver to tote our groceries indoors and unpack them,
the thesis tells your readers what they can reasonably expect to
encounter in a single text. For example, imagine you’re reading an
article that begins
On a Friday morning in April, I strapped on a headset,
leaned into a microphone, and experienced what had been
described to me as a type of time travel to the future of higher
education. I was on the ninth floor of a building in downtown
San Francisco, in a neighborhood whose streets are heavily
populated with winos and vagrants, and whose buildings
host hip new businesses, many of them tech start-ups. In a
small room, I was flanked by a publicist and a tech manager
from an educational venture called the Minerva Project,
whose founder and CEO, the 39-year-old entrepreneur Ben
Nelson, aims to replace (or, when he is feeling less aggressive,
“reform”) the modern liberal-arts college.35
One paragraph into an article with an inviting beginning (On a
Friday morning in April…), and we already know the article’s
central focus: a tech company focused on delivering education
entirely online. As we read on, we’re not looking for an AA meet-
ing in a church basement that will wean winos off the sauce or
112 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

a journalist’s foray into virtual reality via a headset and micro-

phone. Instead, the last sentence cues us to the entire article’s
focus, directing our attention and memories to the Minerva
Project’s aim to put higher education entirely online.
Note how the placement of the thesis in the final sentence of
the opening paragraph enables us to easily and quickly identify it,
even a thesis as colorful and descriptive as the one in this Atlan-
tic article. In studies spread across a variety of disciplines, from
linguistics to educational psychology, researchers have linked the
strategic placement of a topic overview sentence to rapid read-
ing, better recall, and a distinct sense of a document being well
organized. In one study, native speakers of English markedly pre-
ferred texts with an easily identifiable thesis statement early in
the content.36 Still other studies document the impact of a thesis
sentence in helping readers to identify and recall relevant material,
boosting reading speed.37 Most strikingly, in one study of nearly
1000 student essays, two judges scoring for overall discourse co-
herence had a mind-blowing 100 percent agreement in identifying
each essay’s thesis and impact on the essay’s coherence.38 If the
words mind-blowing and student essays don’t seem remotely re-
lated, consider the infamous 1961 study involving 300 papers and
fifty-three readers, drawn from six different fields. When the furi-
ous scribbling of scores ceased, 90 percent of the student papers
received seven different scores on a mere nine-point scale. One
shudders to think what scores the papers would have received
had the scoring scale run to even ten, let alone 100, points.39 With
its long and vexed history, the issue of scoring student essays
and what researchers call inter-rater reliability has few instances
of 100 percent agreement on anything, including probably the
color of the walls of the rooms in which the readers scrawled their
scores or the direction in which the sun rises. The agreement on
the location and centrality of a thesis to an essay’s coherence is, as
I said, either mind-blowing or a testament to the power of thesis
sentences to prime and frame reader expectations – or both.
Now, why place the thesis sentence at the end of the document
head paragraph or paragraphs? Admittedly, reader expectations on
what they’ll find and where they’ll find it play a role in the utility
of thesis sentences.40 However, the end of any paragraph also re-
ceives a bump in recall from the recency effect. The recency effect
 PUT YOUR THESIS AT THE END OF THE HEAD PARAGRAPH 113

extends over the ends of lists, sentences, and, most importantly

for our concerns here, paragraphs and documents.41 By putting
your thesis sentence at the end of a series of introductory para-
graphs, you ensure the thesis receives greater memorability than
the sentences around it, making it stand out and also act as your
readers’ guide to the thickets of details that lie ahead.

The skinny on thesis sentences

For those of you skimming this chapter for the meaty bits of
advice you can use on that proposal you’ve put off writing un-
til the night before it’s due, look no further. Your thesis is a
one-sentence summary of what your document covers. If your
document is a proposal, your thesis should summarize what
you aim to do and what benefits your readers might enjoy if
they green-light your proposal. If your document is a report or
an analysis, your thesis should state your chief findings or your
recommendations for action. If your document is an argument,
your thesis should state your anticipated conclusion. (Inciden-
tally, if you are uncertain about your conclusion when you start
writing your thesis, you’re in big troubs and should go back to
brainstorming and outlining your ideas before you proceed.)

Make your thesis your clearest sentence

Your thesis must be your clearest sentence. If your readers mis-

read or misunderstand your thesis sentence, you’ve set them
up for difficulties in understanding the contents of the para-
graphs to follow. The weight of your entire document rests on
your thesis, so take time writing and rewriting it.

EXPERT TIP: Use a preliminary thesis

Face it, your readers aren’t seven-year-olds stumbling through See
Ginger claw. Claw, Ginger, claw. sentences. You might be writing
about monoclonal antibodies or antitrust law for people who’ve never
114 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

been within shouting distance of a science laboratory or college of

law. But you’re still writing for people who’re bloody good at reading.
Consequently, you’re addressing readers who expect to learn some-
thing from what you’ve created and are as expert at extracting infor-
mation from your paragraphs as seasoned miners working a seam
underground. Your readers rely on signals they expect to see in the
text. The more signals you provide, the faster and more efficient their
reading.42
So what happens when you need to introduce something complex
and to also engage in all sorts of hand-waving about the importance
of your topic? In the sciences, law, and finance, writers need to elbow
their intended audience in the ribs: “Pay attention!” This nudge about
the growing number of people with diabetes worldwide or the compat-
ibility of doing good and making money can run to several paragraphs.
Meanwhile, your readers’ brains are busily scanning for the important
stuff, while they might consciously think “Where’s the bloody point?” If
we think back to the top-down and interactive models of reading, your
readers might well fail to register all your points about the importance
of what you’re about to tell them. To stave off this possibility, you need
to get to the point before you get to the point, in a manner of speaking.
In other words, we need a ballpark sense of where you’re going to take
us before you get to the thesis three paragraphs into the thing. And the
title of your article or proposal or manuscript ain’t going to cut it. For
starters, titles of things tend to be relatively brief and most likely strike
at the central ground between your overall focus and the specific point
you’re going to make.
Instead, in these situations, you should create a preliminary thesis, a
sentence that informs readers of the main topic of your document. The
preliminary thesis belongs in the last sentence of the first paragraph of
your document head paragraph, where it receives emphasis from the
recency effect, contributing to its memorability and the attention read-
ers will give it.43 When you use a preliminary thesis sentence, you’re
preventing your readers from scratching their heads, wondering what
your article or proposal is about as they try to take in the details you’re
giving. Your preliminary thesis provides readers with a temporary bit of
scaffolding they can use for handling information and storing it in work-
ing memory before they reach the thesis sentence.
 USING CONCLUSION SENTENCES [COHERENCE PRINCIPLE #5] 115

Coherence Principle #5:

End complex paragraphs with a conclusion sentence.
And end complex documents with a conclusion paragraph

Ideally, you want readers to perceive your writing as tightly organ-

ized – the equivalent of those Inca-built stone walls so well put
together that, even today, no one can slide a knife-blade between
the massive stones. To achieve this effect, you might want to intro-
duce conclusion sentences to some of your more complex para-
graphs. Your conclusion sentence serves as the foot to your para-
graph’s head and body. Conclusion sentences tell your readers
exactly what they should take away from the paragraph they’ve
just read.
In a paragraph conclusion sentence, recency can even trump
primacy or help readers sift through complex details to arrive at a
conclusion, as researchers discovered when they sought merely to
assess the influence of descriptions of a character’s behavior on
students’ perceptions of character. Readers who encountered an
assessment about a fictitious man, Jim, and his personality traits
after they read about his behavior, made more accurate predic-
tions about Jim’s behavior than readers who encountered an as-
sessment of Jim’s character first, then his specific behavior later.
Contrary to what we might think about the potency of primacy
effects, recency can trump primacy, particularly when a thesis or
conclusion is involved. Occasionally, even when researchers are
looking at social behavior, they’re telling us something valuable
about reading – with distinctly useful applications for writing.44
Think of conclusion or paragraph conclusion sentences as pro-
viding readers with an explicit, one-sentence summary of your
paragraph’s takeaways. In addition, because conclusion sentences
occupy the emphasis positions of paragraphs, your readers will re-
member them longer and better than they do other sentences from
the same paragraph. Moreover, conclusion sentences can help
your readers transition from one paragraph to another seamlessly.
Like conclusion sentences, document conclusion paragraphs
help readers by doing a bit of their work for them. Document con-
clusion paragraphs flag important findings and conclusions from
your document, singling out the central details you want your read-
ers to recall. In addition, by conforming to readers’ expectations
116 COHERENCE: ORGANIZING PARAGRAPHS AND DOCUMENTS

about document structure, these conclusion paragraphs help read-

ers group the important information you’ve provided in a single
or series of coherent chunks they can easily recall.45 And, because
the conclusion paragraphs are the very last things anyone reads in
your text, your readers will remember them better than any other
paragraphs in your document, courtesy of the recency effect on
memorability.

Writing myths you’ll wish you’d never learned:

Repetition is always a bad thing

I can hear you thinking “But isn’t this head–body–conclusion

structure committing that fatal sin of repetition my teachers al-
ways warned me about?” Actually, repeating yourself has its mo-
ments, in spite of everything you’ve heard. When students con-
fronted complex explanations or presentations in one experiment,
researchers discovered students demonstrated more accurate recall
of information when they encountered it more than once.46 This
recall was especially strong when readers encountered data repre-
sented graphically that reinforced information presented in a text,
potentially due to our brains having separate systems for perceiv-
ing and comprehending visual and verbal information.47

EXPERT TIP: Headings and subheadings are your readers’ life rafts
Remember, your readers aren’t poring over your document for pleas-
ure. Nor are they reading every word in the way they might study for
a test. Instead, they’re looking to extract information as efficiently as
possible.48 Expert readers, as we’ve seen earlier in this chapter, rely on
markers that enable them to understand both the gist of the text and
the information they seek. Imagine your readers stumbling around in
your document as haplessly as I. A. Richards’ undergraduates floun-
dered when the authors’ names and titles were ripped from their poetry.
Then help them out with headings and subheadings that cue them to
content they’re about to read. Headings and subheadings prime and
fine-tune our expectations about what lies ahead, resulting in better
comprehension.49
 TAKEAWAYS FOR COHERENCE 117

Takeaways for coherence: what you need to remember to make

your paragraphs and documents tightly organized
• Check that your paragraph heads provide a comprehensive
overview of the paragraph body.
• Make sure your paragraph heads occupy only a third of the
length of the paragraph.
• Ensure your paragraph bodies only contain information referred
to in the paragraph head.
• Use document head paragraphs to introduce readers to your
topic, its significance, and relevance for your readers.
• Place your thesis sentence in the last sentence of the last
paragraph of your document head.
• If your paper has multiple head paragraphs, insert a preliminary
thesis at the end of the first paragraph.
• End complex paragraphs with a conclusion sentence that sum-
marizes the paragraph’s highlights or importance.
• End complex documents with a conclusion paragraph that
stresses your paper’s conclusions, as well as their significance
or relevance to your readers.
CHAPTER 6

Maximizing efficiency
The fourth C: Concision

In this chapter you will

• learn how to recognize and eliminate redundant pairs
• discover how to spot and weed out unnecessary narration,
hedges, and amplifiers
• understand why you should avoid using negatives
• know how to recognize linguistic fossils and throat-clearing.

PRESIDENT: Well, you had quite a day, today, didn’t you? You
got, uh, Watergate, uh on the way, huh?
HALDEMAN: How did it all end up?
DEAN: Uh, I think we can say “Well” at this point. The, uh,
the press is playing it just as we expect.
HALDEMAN: Whitewash?
DEAN: No, not yet; the, the story right now –
PRESIDENT: It’s a big story.
DEAN: Yeah.
PRESIDENT: (Unintelligible)
HALDEMAN: Five indicted,
DEAN: Plus,
HALDEMAN: Just so they have the fact that one of –
DEAN: plus two White House aides.
HALDEMAN: Plus, plus the White House former guy and all
that. That’s good. That, that takes the edge off whitewash,
really – which – which was the thing Mitchell kept saying
that …
PRESIDENT: Yeah.1

In 1974, the Chicago Tribune published a forty-four-page excerpt of

taped Oval Office conversations, unedited and never intended for
 CONCISION: MAXIMIZING EFFICIENCY 119

public consumption. The meeting, held September 15, 1972, followed

on the heels of the press airing a link between the Nixon Administra-
tion and five men arrested during a break-in at Democratic Party
headquarters, which they clearly intended to bug.2 At the time, the
American public found Nixon’s vocabulary shockingly rich in
the four-letter words some media outlets still touchingly bleep out
as if protecting our tender ears from corruption. But this snippet
of business-as-usual conversation in the White House might also
seem remarkable for the lack of real information traded between
an American president, his Chief of Staff, and the White House
Counsel.
However, if you think Nixon and his two aides have banality
cornered, you might want to turn a microphone on your own
conversation and assess how much information and how much
noise get exchanged. Answer: you might sound more like Nixon
than like Lincoln on the stump or Churchill addressing the House
of Commons in the thick of the Battle of Britain. Verbal conversa-
tions tend to rely heavily on shared understandings and on the
presence of someone who can grunt in acknowledgment, nod,
wince, smile, or ask us what we’re talking about. Speech can
be vague, elliptical, and maddeningly redundant. No matter how
badly you might yearn to fast-forward through your colleague’s
soliloquy on his plague of in-laws – that five-minute number on
endless repeat where you can just hear the semicolons slotting
into place – you can’t. Still, even the party bore’s endless nat-
tering has some tenuous relationship with meaning. When we
converse, we assume the words we exchange are informative,
relevant, succinct, and also avoid conveying more information
than we need to hear.3
In contrast, writing lies there on the page, extending your
voice beyond your physical reach, beyond the confines of space,
even beyond the reach of time. We can read Plato no matter how
long he’s been dead. And, if we fail to understand exactly what’s
got him so exercised in Phaedrus, we can reread his words until
we’re satisfied we’ve made sense of them. We can skim ahead
or prospect. More important, we can look back or retrospect,
especially when we discover we were wrong in our predic-
tions about where a paragraph was headed.4 Writing, as Plato
had Socrates complain in Phaedrus, just keeps saying the same
120 CONCISION: MAXIMIZING EFFICIENCY

thing, no matter how frequently you return to it.5 Modern-day

jeremiads about how our intelligence or memories are all going
straight to hell, courtesy of our ability to look up what’s-his-face
on imdb.com, should consider the dual ironies of Plato’s own
miniature jeremiad against the written word. Plato simulated a
conversation in Phaedrus, as he did in his other dialogues, using
the written word to fix his own convictions about the evils of the
written word. As the French would say with a Gallic shrug, La
plus ça change…
Unlike the first three Cs, the fourth C, concision, involves leav-
ing aside neuroscience and psychology in favor of examining the
origins of our linguistic hangers-on, those unnecessary words
Strunk and White so helpfully tell us to omit without specifying
exactly what sorts of words we should be looking to cut. We’re
taking a detour through linguistics, where hard-headed data from
labs meet the study of why people say the things they do, with a
side helping of the history and development of English.
If you skipped “An extremely short history of English” on pages
31–32, you might want to read that now. Why? The history of English
has implications for practically every sin against efficiency com-
mitted in English, the topic that brings us to the heart of concision.

Think you already avoid “unnecessary words”?

Think again

Imagine you’re an employee of county government, somewhere in

the United States in a county I daren’t name. Now imagine a two-
page, single-spaced memo lands on your desk, one of two dozen
memos you have to read every morning. Moreover, the memo
begins:
During our staff meeting of November 11, I had an opportunity
to discuss my personal and professional feelings and thoughts
regarding recent events that surround our government. More
specifically, the unpleasant experiences we have had as of
late with continuing revelations of mismanagement, political
patronage and, worse yet, corruption in significant parts
of our organization. In the past few weeks, subsequent to
 THINK YOU ALREADY AVOID “UNNECESSARY WORDS”? 121

extensive media coverage on the pavement/repair contracts

at the Water and Sewer Department, I have been approached
by our Mayor and Commissioners, members of the media
and residents of this community expressing their outrage and
dismay at County government. All, without exception, told me
(as I expressed to you during this meeting) that they are fed
up with the ongoing reports on corruption. Frankly speaking, I
too am fed up not only with these revelations but more so with
our “business as usual” attitude, our perceived inability to
act and the lack of professional and personal concern shown
throughout the years for these issues.

Chances are, if you didn’t reread that first paragraph outright,

you’ll start skimming the rest of the memo at warp speed. Un-
fortunately, the rest of the memo reads like the first paragraph,
since the memo was written by a writer who should be called the
Double Man.
The Double Man seldom says anything in a single word if he
can use two – or even, on occasion, three. He also specializes in
throat-clearing, sentences that sound as if they’re communicat-
ing something but instead simply showcase the debris knocking
around the writer’s head while he figures out what on earth he’s
going to say. The Double Man is the equivalent of the party bore
who bends your ear for hours with the juicy details of his last four
meals at McDonald’s, the contents of this year’s 1040 tax return,
and the kind of really great beer he once drank at a restaurant
whose name he now can’t quite recollect. Like the party bore, the
Double Man seldom gets to his point, or, if he reaches it, you can’t
spot it amid all the debris. The Double Man can also sometimes
display a weakness for Big Bang Beginnings. And the Double Man
will fit beautifully into most law firms and some of the civil ser-
vice, where he will spend a career writing sentences so complex,
lengthy, and riddled with jargon that you’ll have to hire a lawyer
yourself to ensure the contract you’re about to sign won’t end up
bankrupting you.

What’s double about the Double Man’s writing?

Enough about the Double Man. Let’s look at what earned him this
title:
122 CONCISION: MAXIMIZING EFFICIENCY

During our staff meeting of November 11, I had an opportunity

to discuss my personal and professional feelings and thoughts
regarding recent events that surround our government.

Every underlined word in the sentence above is unnecessary. For

instance, the audience for this memo only recently attended the
meeting, so the date is unnecessary. Presumably, since the people
receiving the memo were also at the meeting, they also can skip
the reprise of the last meeting, especially since the Double Man
seems like the kind of guy who would have held forth for a hefty
chunk of the meeting, making a reminder of what he said also
superfluous.
Furthermore, while the Double Man had the opportunity to do
something, readers only care about what he did, not whether he
had an opportunity to do something. Think of telling your read-
ers about having an opportunity as the equivalent of the old saw
about a tree falling in the forest with no one to hear it. If you did
something fairly routine – as with the Double Man holding forth
at a meeting – then the opportunity is the equivalent of the tree
falling with no one to hear the crash of its contact with the forest
floor. Ditto, if you had the opportunity and failed to do anything
with it. You should report an opportunity only if you were denied
one and pulled off a coup against all odds, capitalized on an extra-
ordinary opportunity, or are in the throes of some scandal and are
avidly denying accusations. So a writer should scratch I had the
opportunity to discuss.
Finally, that surround our government is also superfluous, as
well as a bit Big Bangish, since Water and Sewer employees are un-
likely to care about what their fellow civil servants are doing over
in, say, Tax Collection or Waste Management. If the details involve
their department, the Double Man should tell them so. If the details
don’t, the Double Man should refrain from sending the Water and
Sewer employees a memo about the other divisions, period.
In every instance, the Double Man is sinning against written
English by producing a memo that might bring to mind a two-
pound textbook on legal writing. In addition, he’s violating the
basic rules for conversation identified by one of the godfathers of
the study of speech and communication, Paul Grice. Like the pro
that he is, the Double Man trips right over Grice’s first two rules
 THINK YOU ALREADY AVOID “UNNECESSARY WORDS”? 123

on the quantity of communication required to grease the wheels

that spin during our conversations:

1. Make your contribution as informative as required to keep the

conversation going.
2. Avoiding packing your contribution with more information
than is necessary.6

Remember, everyone reading the memo was present at the staff

meeting the Double Man recounts in the exhaustive detail com-
mon to doting parents’ descriptions of their kid’s Little League
game. Everything in those opening lines is a review of an experi-
ence for an audience who were at the same meeting. If the Double
Man were speaking to you in the flesh, you could at least cut him
off for violating Rule #2 and usher him to his point – whatever the
hell his point is. But we’re reading and miles away from the Double
Man. Worse, we’re reading a memo with more than a passing
resemblance to Bleak House without the good prose. Furthermore,
the memo lacks paragraph heads or anything with remote kinship
to a thesis. In other words, the Double Man has produced a memo
that frustrates our every attempt to skim it, at the same time he’s
produced a masterpiece illustrating every sin against concision.
The Double Man is the Hieronymus Bosch of business, painting
‘‘The Seven Deadly Sins’’ that handily illustrates every deadly sin
against concision in action, as his memo proceeds to its second
through fourth paragraphs of a seven-paragraph, two-page memo:

While it is evidently clear that events at this department are

not the makings of present management as some would like
us to believe, it is also evident that those who throughout the
years had oversight responsibilities for these areas failed to
act. They are now shedding their responsibility by indulging
in hollow rhetoric or pointing the finger in an opposite
direction. Let’s not fall in this trap. Our job as professional
public administrators is to stay above this fray and stay the
course by continuing to work through these difficult times in a
professional manner.
That is what our community and our elected officials
expect of us. That is what we must expect of ourselves
and what I expect of you. We must say “enough.” As
124 CONCISION: MAXIMIZING EFFICIENCY

we discussed, we are all on notice. I expect each of you

(and your department management teams) to have a
“zero tolerance” attitude towards any factual revelation
of mismanagement, malfeasances, misfeasances and
public corruption. I will hold departmental management
accountable for these instances. In return, you must
hold your managers and employees accountable as well.
Furthermore, you have my strongest support when taking
appropriate actions.
I want to take this opportunity and clarify a major
misconception some of us have. We must be cognizant that
when it anywhere [sic] in County management it is not
their problem, but ours. All of us are touched and stained
by these unacceptable actions. As I stated, most of the
cases I know may not directly be our doing, but we are
here now; and it is up to us to make it right, and make
it right immediately. At the same time we must not act
hastily and capriciously for political expediency; we must
be thorough, fair and deliberate in our actions irrespective
of the unwarranted criticism to be expected from those with
different agendas.

Are you referring to a feeling or a thought?

Now look at personal and professional and feelings and thoughts.
Again, the Double Man’s audience really cares only about the pro-
fessional, not the personal. In any case, you can bet that the other
civil servants in the Water and Sewer Department are not going to
lounge around the local bar, wagering about whether the Double
Man’s feelings are personal or professional. Better yet, the Dou-
ble Man should just axe the adjectives personal and professional
altogether, since his audience already knows the context is pro-
fessional. Similarly, his readers are unlikely to care whether the
Double Man is airing a feeling or a thought. He should just stick
with one, probably thoughts, as most employees in business re-
ally tend not to care about feelings – or at least not about feelings
other than their own. Talk about violating Grice’s second prin-
ciple about quantity in conversation. Moreover, offenses against
 THINK YOU ALREADY AVOID “UNNECESSARY WORDS”? 125

quantity in conversation are merely an obnoxious waste of time.

In contrast, excessive details, especially narrated detail already
familiar to readers, fly in the face of the raison d’être for writ-
ing. Writing ensures we can abandon the redundancy inherent in
speech,7 unless we’re emphasizing a point we particularly need
readers to remember, the good-repetition phenomenon that makes
a paragraph conclusion helpful, rather than a waste of time.
The Double Man reminds us why researchers assessing student
and professional writing inevitably include concision as a category
in their evaluation criteria, irrespective of whether they’re assess-
ing legal briefs or engineering student projects. Most of us sense
vaguely that we’re thrashing around in swampy prose with mass-
es of passively constructed sentences. Or that we can’t pick out
the important ideas in paragraphs that run to several pages, stud-
ded with sentences peppered with semicolons and bristling with
clauses. In contrast to clarity, continuity, and coherence, we can
spot a lack of concision easily and put a name to that particular
affliction on the page before us. Most reviewers and researchers
easily identify redundancy, the reason why the experts from so
many fields agree on whether a writer can write concisely – or is
doing an uncanny impression of the Double Man.8
In any case, the Double Man’s memo is the perfect speci-
men of anti-concision. Writing concisely entails saying what
you need to say in the fewest words possible. The Double Man,
in contrast, seems to have written the equivalent of A Tale of
Two Cities into his memo. His employees get every nugget of
what’s happened since November 11, except for the one all-
important piece of information they’re looking for: what they’re
supposed to do after receiving the memo. In fact, if you were
to analyze the Double Man’s full memo – which I’ve mercifully
avoided reproducing here in its full glory – you’d discover that
he never really gets to his point in the memo, let alone calls
for any action from his audience. But, nevertheless, we should
be grateful to the Double Man for providing us with a sterling
example of all the ways in which the English language can
conspire to pad out your sentences, waste your readers’ time,
and generally make your message as confusing and inefficient
as possible.
126 CONCISION: MAXIMIZING EFFICIENCY

Concision is also precision

Concision is what Strunk and White are after when they instruct
you to “Omit needless words.” Concision is what your high-school
English teachers meant when they scrawled “wordy” in your mar-
gins. But, to eliminate these evils, you first must learn how to rec-
ognize the ways in which English nearly automatically lends itself
to repetition – and how to guard against this purposeless kind of
redundancy in your own writing.

English: the doubled language

English is a mongrel language, a hybrid of several languages that
developed after the Normans, formerly Norsemen, invaded what
later became England. As a result, after 1066, residents of Eng-
land spoke three languages: French, Latin, and English. French
became the language of sophistication, a heritage that even today
distinguishes English speakers who use words and phrases bor-
rowed from French like façade, charade, and tête-à-tête. Latin, on
the other hand, rapidly became the language of government, law,
and religion. Even today, most legal terminology comes to us from
outside English’s Anglo-Saxon roots, evident in words like felony,
perjury, plaintiff, and attorney.9 English, however, survived be-
cause the Normans almost immediately began intermarrying with
English speakers. As a result, speakers tended to pair words, using
a term from Anglo-Saxon alongside a term from either French or
Latin, to ensure they made themselves understood – or sounded
appropriately learned – in this mixed linguistic environment. After
the Norman Conquest, you could call the head of state a king,
using the Old English standby, or be a bit flashier and plump for
sovereign and royal, both shamelessly pilfered from French, or
regal, which debuted the same year as royal but arrived courtesy
of Latin.10
At first glance, this embarrassment of linguistic riches seems
like a boon, one of the few spoils of the Norman Conquest left
around to comfort the illiterate, conquered inhabitants of what
later became England. But, instead, the westward drift of French
and Latin across the English Channel merely continued what
had been a long-held tradition in Old English: seizing whatever
interesting linguistic tidbits washed ashore and using them
 AVOID REDUNDANT PAIRS [CONCISION PRINCIPLE #1] 127

alongside existing words. Hundreds of years before the Normans

busied themselves turfing their illiterate English-speaking sub-
jects from their lands, English speakers used Old Norse words
alongside Germanic and Old English words. The Latin-scribbling
Normans used written titles to lay claim to English lands, where-
as the illiterate English speakers used objects, oaths, and seals –
all infinitely more difficult to falsify than a single deed. However,
300 years later, the English still used both systems – objects and
oaths alongside written deeds. Apparently, the English were as
reluctant to surrender to a single system as they were to a single
language.11 As a consequence, today you can rear (English) or
raise (Old Norse) a child,12 or ask (Germanic), question (French),
or interrogate (Latin)13 the bartender over that monstrous bar tab
you ran up while buying rounds for people you actually didn’t
know.
Today, these redundant pairs dog English, long after the reason
for their pairing vanished, largely because English became a writ-
ten language soon after speakers began using redundant pairs.
And if writing freezes language, then the invention of Gutenberg’s
printing press ossified English, halting any dramatic developments
in structure, usage, or grammar.14 Ultimately, we can blame the
written word and history for English’s abundance of redundan-
cies. But look on the bright side. Once you know how to spot
English’s infamous redundant pairs, you’re well on your way to
avoiding ever becoming the Double Man.

Concision Principle #1:

Avoid redundant pairs

Redundant pairs clutter up sentences, providing stumbling blocks

for your readers or wearing out their patience as they plod through
unnecessary pairs like first and foremost or basic and fundamen-
tal. Since each word in the redundant pair is similar or even identi-
cal to the other, you should only use one word. Any time you spot
a redundant pair, drag your pen or mouse over the and, plus the
offending first (or second) word, and delete two of the three. In
retaining one of your pair of nouns or adjectives or verbs, go for
the familiar and axe the arcane. Some of the words in these pairs
128 CONCISION: MAXIMIZING EFFICIENCY

are linguistic fossils, a word with a meaning that survives only

inside a larger expression, as in hale within hale and hearty.15
Try using hale next time as an adjective to describe someone and
watch the befuddled response. Just be careful to try this fossil out
only on your friends.

Snares to avoid: redundant pairs

Not all redundant pairs feature a word from French and a word
from Old English. Some, as the list below reveals, borrow one
word from French and the other from French, or a noun from Latin
and a noun from French. Or one of the pair from Norse or Old
High German or whatever language was available:
aid (French) and abet (French)
any (English) and all (Old Danish)
basic (French) and fundamental (Latin)
breaking (English) and entering (French)
cease (French) and desist (French)
each (English) and every (English)
first (Old High German) and foremost (English)
final (English) and conclusive (Latin)
fit (English) and proper (French)
full (English) and complete (Latin)
holy (English) and sacred (French)
heirs (French) and assigns (French)
new (English) and novel (French)
null (French) and void (French)
peace (English) and quiet (Latin)
so on (English) and so forth (English)
true (English) and accurate (Latin)
various (Latin) and sundry (English)

Writing myths you'll wish you never learned:

When in doubt, copy other writers

At some point, most of us find ourselves under tight deadlines or

on uncertain footing with the subject we’re writing on – the mo-
ment when we decide to just imitate what other writers have done.
 AVOID REDUNDANT MODIFIERS [CONCISION PRINCIPLE #2] 129

After all, their work survived and made it into print or some kind
of record, right? Actually, if only writers had ignored this timeless
and wrong-headed bit of advice, writing everywhere would be far
easier to read.
Even the non-legal minds among you will recognize the unusual-
ly high percentage of legal terms that seem eternally shackled
to one another, despite each word meaning nearly or exactly the
same thing as the other word. Education has a long and distin-
guished history of once-useful but now crappy advice, beginning
in the Renaissance with Erasmus, who remained convinced that
imitating other writers should continue being taught as a staple of
rhetoric. But the first rhetoricians were gifted orators in an illiterate
society with no means of circulating brilliant speeches, aside from
memorization and imitation. Thousands of years later, Erasmus
was either caving to the Renaissance tradition of ennobling all ele-
ments of the classical education or to the relative scarcity of books
during the early Renaissance period.
The same principle trickled down into the nineteenth century,
where clerks hunched over their desks learned how to write cor-
respondence by copying already-existing letters, warts and all.16
This practice of parroting and preserving in amber every dread-
ful verbal tic of our predecessors today survives most durably in
legal writing, where legal drafters were once paid by the word,
making wordiness a boon to the bottom line. Today, instead of
slicing through the thickets of verbal tics and redundancy pro-
duced in earlier statutes and codes, legal drafters mostly bolt
extra bits onto boilerplate text as new scenarios and technologies
occur.17

Concision Principle #2:

Avoid redundant modifiers

Redundant modifiers represent a slightly different case from re-

dundant pairs because one term in the duo only implies the other
rather than repeats it. However, since one term implies the other,
one of the words is unnecessary. In one study, researchers pro-
vided an explanation for how we almost immediately grasp the
130 CONCISION: MAXIMIZING EFFICIENCY

meaning of a paired modifier and noun or verb. Even when we’re

merely hearing, rather than reading words, we assign meanings
to the blur of syllables, so we need to fix the words’ grammatical
function as well as the plausibility of the novel pair. But using that
same mechanism for comprehending potential millions of novel
combinations immediately, we also expect the second word to
convey new information, providing a constraint that illuminates
the use of the other word.18 In redundant modifiers, however, the
added word merely repeats the core word, both violating our ex-
pectations and eliminating the extra word’s usefulness.
As a result, you should omit one of the pair, whenever you spot
a set of redundant modifiers. Grammar mavens will notice, if they
scan the list below, that redundant modifiers involve adverbs and
verbs or adjectives and nouns, or, occasionally, a verb and preposi-
tion. While redundant pairs take different grammatical forms, the
pairs are nevertheless a sign that you’re writing at less than maxi-
mum efficiency.

Snares to avoid: redundant modifiers

basic (French) fundamentals (Latin)
completely (Latin) finish (French)
consensus (Latin) of opinion (English, French)
continue (French) on (Old English)
each (English) individual (Latin)
end (English) result (Latin)
evidently (Latin) clear (French)
final (French) outcome (English)
free (Old High German) gift (English)
future (French) plans (French)
important (French) essentials (Latin)
more (English) specifically (Latin)
past (French) memories (Latin)
personal (Latin) beliefs (English)
revolve (French) around (English)
split (Middle Dutch) apart (French)
Not surprisingly, the Double Man proves as fond of redundant
modifiers as he is of redundant pairs, chucking a perfect pair like
evidently clear into his second paragraph. If things are evident,
 AVOID REDUNDANT MODIFIERS [CONCISION PRINCIPLE #2] 131

they are also, not terribly shockingly, clear. Likewise, most of us

are unfamiliar with things like unimportant essentials or semi-
final outcomes or impersonal beliefs or incompletely finished
tasks for the simple reason that these things, if they exist, are a
bit like Lewis Carroll’s mome raths that do a bit of outgrabing
in the poem “Jabberwocky.” We can guess that they might exist,
but don’t ask us to describe them. Words like evident also
imply that something is clear, just as to finish something means
to complete it. Likewise, we expect essentials to be important,
fundamentals to be basic, and outcomes to be final. On those
rare instances, however, where the outcome is preliminary, ra-
ther than final, we expect a modifier like preliminary to inform
us of the fact. Similarly, while plans can exist in the past, most
plans apply to the future, so readers expect that plans describes
a blueprint for a project or actions that have not yet taken place.
If, on the other hand, the writer is referring to a blueprint for
past actions that didn’t quite pan out, readers expect this sort of
plan to be modified by the word past, satisfying our notion that
the added word is actually informative. Likewise, each refers to a
single individual, while most gifts tend to be free – unless some-
one in retail or travel is offering one to us, in which case free gift
is usually an oxymoron, or anything but free.

Yet another holdover from Ye Olde Englande

Intrigued by the mystery of precisely why English has such a
superfluity of extra words knocking around? Consider the early
English tendency to refuse to surrender anything that washed
ashore: traditions, Latin, French, stray words from invaders and
trading partners alike. If we examine the etymology of the most
common redundant modifiers, we discover the same pattern we
found in redundant pairs. One word stems from English or French,
the other from French or Latin, or whatever other language hap-
pened to land on the western side of the Channel.
In a few instances, pairs like future and plans could have end-
ed up paired for hundreds of years because English was still tot-
tering from its collision with two distinctive types of French – that
spoken by the Parisians and the dialect used by the Normans who
marched over England after the Conquest, their French tinged
(tainted, the Parisians would’ve said, tartly) with shards of their
132 CONCISION: MAXIMIZING EFFICIENCY

Norse forebears. This persistence of French doublets accounts

today for the ever-so-British gaol (Norman) and the more common
jail (Parisian). But the borrowing from two rather different sorts
of French also gave us warden (Norman) and guardian (Paris-
ian), as well as warrant (Norman) and guarantee (Parisian).19 We
might wonder what we can learn about English speakers by their
excessive preoccupation with all matters legal … right down to the
niceties of the penal system.

Concision Principle #3:

Avoid negatives

When my brother once exhibited some squeamishness about his

qualifications for a job in Hollywood, a friend stuck his face close
to my brother’s and advised, “Don’t tell nobody nothin’.” I puz-
zled over that bit of advice for at least a few hours, struggling to
determine whether that meant he should tell everybody every-
thing – or nobody something. Check out some of the US Internal
Revenue Service publications online or practically any legal docu-
ment, and you’ll also discover convoluted thickets of negatives,
the sort that prompt you to wonder if you should apply the old
rules you learned in elementary school math and make a positive
of two negatives.

Don’t use no negatives

Actually, negatives are just plain difficult for readers to process,
period, a discovery researchers made when they tested the corti-
cal activity and slowed-down speed of readers struggling through
both passively constructed and negative sentences.20 Any nega-
tive in a sentence implies what is to readers by telling them
what isn’t. Put simply, your readers have to perform the mental
equivalent of the old game Twister to figure out what on earth
any sentence containing negatives means. So when a boss writes
a memo stating, Anyone found violating company policy may be
subject to a range of disciplinary actions, up to and not excluding
termination, his employees are nearly standing on their heads
in the mailroom, trying to determine whether he means includ-
ing termination or not including termination. The difference for
 AVOID NARRATING OR HEDGING [CONCISION PRINCIPLE #4] 133

some of us is the distance between a paycheck and the dole

queue. That old rule about double negatives, hanging over from
math, also tends to mess with your head. Does not entirely un-
pleasant mean unpleasant, mildly unpleasant, or so dreadful I
can’t tell you about it without resorting to four-letter words begin-
ning with f?
Simply put, avoid negatives whenever you can. Instead, state
things positively. Substitute young for not old enough, and differ-
ent for not alike, or lacks for does not have. Once you sensitize
yourself to negatives, you’ll start seeing how easily you can avoid
them. Moreover, writers avoid negatives fairly naturally when
they start making their grammatical subjects and verbs actively do
stuff. Seriously. Try it in your next email.

Swaps for concision: replacing negatives

not the same – different
not many – few
did not – failed to
does not have – lacks
did not stay – left
did not accept – rejected
did not remember – forgot
did not consider – ignored
not necessary – unnecessary
not possible – impossible
not certain – uncertain

Concision Principle #4:

Avoid narrating or hedging

In other words, just bloody get to your point, or, as Joe Friday in
Dragnet would’ve said, “Just the facts, ma’am.” I love the ma’am
bit, suggesting that women always need to be steered away from
imminent hysteria or great swaths of irrelevance and kept on the
straight and narrow path to Fact-dom.
Joseph Williams had a name for about 90 percent of the content
in the Double Man’s memo – metadiscourse. By metadiscourse,
134 CONCISION: MAXIMIZING EFFICIENCY

Williams meant writing about the act of writing, as in the You’re-

probably-wondering-why-I’ve-asked-you-all-here moment in the
library in so many cozy mysteries that involve murders in vic-
arages. Nevertheless, Williams could be downright beady about
what counted as metadiscourse, including helpful transitions
like In conclusion, which signals to our working memories that
they can take a breather after the next sentence.21 Williams broke
metadiscourse into three broad categories: hedges and emphat-
ics, sequencers and topicalizers, and attributors and narrators. In
Williams’ examples, metadiscourse besmirches any prose it lays
its greasy little mitts on – and even mostly useful transitions re-
ceive an indictment. In contrast, Ken Hyland of the University of
London broadened Williams’ skimpy taxonomy into categories
that suggest entire families and species of metadiscourse, some
of which serve useful purposes when writers are mindful of why
they use them. Hyland and his colleagues identified two basic
types of metadiscourse, interactive and interactional. Then Hy-
land broke each type into categories, giving us eight specific types
of metadiscourse, sorted according to the purpose each serves on
the page.

Can metadiscourse ever serve a purpose?

Rather than contenting himself with exorcizing metadiscourse
wherever he discovered it bedeviling a perfectly innocent piece of
prose, Hyland instead set about investigating why writers used it.
Some types of metadiscourse were interactive, anticipating read-
ers’ responses to the words on the page, including transitions,
frame markers to indicate shifts in topic, and evidentials or exter-
nal sources, all of which you’ve already met in these chapters in
spades. (For examples of each type of metadiscourse, see “A short-
cut to understanding metadiscourse,” below.) In contrast, inter-
actional metadiscourse preserved Williams’ hedges but replaced
emphatics with boosters, then added engagement markers that ex-
plicitly anticipated readers’ responses, attitude markers that draw
the author’s emotions and opinions directly onto the page, and
self-mentions, which explicitly put the writer in the material as
either a narrator or commentator.22 For example, You’ll note that
this paragraph contains an engagement marker contains an inter-
active bit of metadiscourse. In addition, Concision Principle #3’s
 AVOID NARRATING OR HEDGING [CONCISION PRINCIPLE #4] 135

explanation contained a scrap of self-reference: I puzzled over that

bit of advice for at least a few hours …

A short-cut to understanding metadiscourse

Type Category Examples

Interactive Transitions but, however, in addition, first, last, for
example
Interactive Frame this study proposes to, we intend to, in
Markers conclusion
Interactive Evidentials according to a White House source, Douglas
et al. state
Interactional Hedges virtually, probably, in all likelihood, most
likely
Interactional Boosters indeed, clearly, evidently, of course
Interactional Engagement This attitude contrasts strikingly with our
Markers expectations for the way readers’ brains
process metadiscourse; This methodol-
ogy relies on retrospective data analysis,
lending itself to biased interpretations of
data
Interactional Attitude We were surprised to discover; I can
Markers only imagine your distress at this
development
Interactional Self-mentions I find this researcher's explanation to be
clever but also slick and insufficiently
explanatory for why our brains respond to
art; I can recall sharing our community's
outrage during that media event; I am
convinced I'm the perfect candidate for this
research fellowship.

In other words, certain types of metadiscourse have their

uses and are even obligatory in certain scenarios. Most works of
non-fiction, including articles in the Daily Telegraph or Chicago
136 CONCISION: MAXIMIZING EFFICIENCY

Sun-Times contain evidentials in the form of attributions for

quotes and statistics, without which the article’s credibility heads
straight down the bog. Similarly, advertisers in the United States
lean heavily on weasel words like virtually, may, and might to
keep them from being fined by the Federal Trade Commission for
faulty advertising.23 Moreover, researchers routinely rely on hedges
to ensure their peer reviewers don’t rip them into bite-sized
bloody chunks and fry them up as a starter. As Hyland and other
researchers have discovered, hedges including likely, probably,
usually, and that hybrid, the hedging self-mention, we speculate
or we suggest or we expect, all crop up depressingly frequently
in academic writing. Hedges and self-mentions figure heavily in
graduate student dissertations and even in well-regarded medical
journals including the BMJ and Lancet, normally redoubts of good
writing. Unsurprisingly, qualitative research in the social sciences,
seen as squishy and relatively unrigorous, features significantly
more hedges than the biological sciences, perceived as quantita-
tive and robustly data-driven, which has fewer.24 Hedges let us
assert something without losing sleep over whether we’re going
to be sued into penury, particularly in America, where many el-
ementary schoolers master the immortal phrase, “My daddy will
sue you.”

But stay clear of most metadiscourse

Stop before you begin your next email with The reason I’m email-
ing you or worse, in cold-call emails, My name is Gustavo Skura-
towicz – a surreal bit of overkill, given the hefty signature block
at the bottom of the email and the emailer’s own name, helpfully
staring at us from the first line. In addition to some form of meta-
discourse turning up on the shit-list of most guides to writing,
metadiscourse also gets short shrift for, among other things,
authors relying on it to self-promote their own work.25
Nevertheless, researchers have studied the usefulness of meta-
discourse for one of the primary reasons researchers scrutinize
the usefulness of any phenomenon … because they’re astonished
to discover something so universally reviled actually has its oc-
casional uses. In some studies, a writer’s use of metadiscourse
enabled researchers to determine whether blind reviews of manu-
scripts enabled subject-matter experts to determine a writer’s
 AVOID NARRATING OR HEDGING [CONCISION PRINCIPLE #4] 137

levels of expertise (graduate students or professionals), familiarity

with discipline-specific conventions, and even gender. The review-
ers responded strongly to the anonymous author’s use of frame,
attitude, and engagement markers in creating a sense of the author
as a male graduate student, writing for a seminar. In reality, the
author was a male graduate student whose manuscript – the one
the reviewers both rejected – had recently been accepted for pub-
lication by a major journal in his field.26
Think of metadiscourse as the linguistic equivalent of bug
spray, marvelous in small quantities at steering away from our
whereabouts all the pests that bite or sting or spread germs, but
toxic in large quantities. From the studies on metadiscourse, we
can extrapolate five basic principles:
1. Use frame markers only in the form of transitions or when
you can incorporate the shift in topic centrally in the
sentence.
Metadiscourse Frame Marker: This study aims to establish the
potential for Ehlers–Danlos Syndrome Hypermobility subtype
to prevent adverse cardiovascular events.
Textual Frame Marker [no explicit metadiscourse]: Ehlers–
Danlos Hypermobility subtype can potentially prevent adverse
cardiovascular events.

2. Unless you’re a journalist, incorporate evidentials into the

central sentence structure.
Metadiscourse Evidential: In a 2012 study, Douglas et al.
found that minocycline had anti-inflammatory effects …
Textual Evidential [no explicit metadiscourse]: Douglas et al.
(2012) noted minocycline’s anti-inflammatory effects …

3. Use engagement and attitude markers, as well as self-

mentions, only to establish a conversational tone or direct
rapport with an audience.
Attitude Marker: Unfortunately, our calculations for
profitability in Q4 failed to take into account declining
demand for new construction.
138 CONCISION: MAXIMIZING EFFICIENCY

Engagement Marker: We appreciate your input on our new

employee engagement survey.
Self-mention: I apologize for the delay in your shipment of
pelleted goat feed.

4. Avoid all other metadiscourse like the bloody plague,

especially emphatics and boosters.

5. Avoid considering transitions as forms of metadiscourse –

as Joseph Williams did in his Style series – as transitions
boost readers’ speed and ease of comprehension in reading.
Williams included transitions like first … second … finally in
his list of forms of metadiscourse, a misleading classification,
as transitions perform uniformly useful functions in
sentences, whereas metadiscourse plays a meaningful role in
only the handful of exceptions in item 3, above.

Concision Principle #5:

Avoid throat-clearing

Throat-clearing, like its name suggests, rarely conveys any mean-

ing whatsoever. Like pesky burrs, these verbal tics have stuck
to our everyday prose, with the chief offenders including for all
intents and purposes and in order [infinitive]. While no audience
cares particularly if a speaker clears his or her throat periodically
during a lengthy speech, even an audience of Mother Theresas
would probably exhibit visible signs of intense suffering through a
talk where the speaker ahemed, ummmed, and aahhhed in every
sentence – the fate of readers encountering persistent throat-
clearing.
To all intents and purposes debuts, predictably, in legal lan-
guage, that first and last refuge of all things redundant. In 1546,
Henry VIII’s Act 37 includes to all intents, constructions, and pur-
poses, which, by 1709, has morphed into the more abbreviated to
all intents and purposes.27 While not quite a linguistic fossil, the
expression no longer means anything wherever it appears – even
in legal documents. Try covering it up, rereading any sentence in
which the expression appears, and seeing if the meaning changes
one whit. You needn’t fret, even you lawyers. To all intents and
 AVOID THROAT-CLEARING [CONCISION PRINCIPLE #5] 139

purposes fails to tells readers what intents or purposes are in-

volved, even in legal writing. If you want to specify intents or
purposes, go ahead and be specific about them. If you’re not,
forget the expression exists.
And, finally, the most ubiquitous and utterly meaningless ex-
ample of throat-clearing is in order to, which never contributes
so much as a smidgen of meaning to any sentence. If you simply
delete in order and leave the infinitive to ______, the meaning
of your sentence will never change. But, given how the expres-
sion in order to crops up in sentences everywhere and how little
it contributes to any sentence, we might wonder just what bog
hatched this particular bit of throat-clearing. In the early decades
of the twentieth century, a crusty old school master, H. W. Fowler,
published what became the Bible of English usage, Modern Eng-
lish Usage. Fowler’s edicts in that little tome saddled writers for
decades with an imperative to begin restrictive adjective clauses
with that and the fluffier, non-restrictive clauses with which. If
you’ve ever suffered under this rule during a Microsoft Word spell-
check – invented because Fowler thought the unregulated use of
that and which seemed shamefully sloppy – you can curse Fowler.
On the other hand, the school master also had little patience for
that bugbear, in order to. According to Fowler, the expression once
represented a means of flagging an analogy.28
In order to _______ could, however, also just as easily have
come to us courtesy of the Norman Conquest and the French en
train de [faire], which translates as to be in the process of [doing
something].29 Tellingly, the Oxford English Dictionary records in
order to first popping up in the Douay Bible in 1609. The Douay
(or Douai) Bible represents the first translation of the Bible into
English, the product of untold toil by the University of Douai in
the service of the Catholic Church.30 Douai is in France, and the
Bible’s translators both lived in France and worked during an era
when English was still shamelessly borrowing words from Latin
and French, as if conscious of its mongrel origins. We probably
use in order to _______ because French scholars who spoke Latin
relied on en train de [faire] over 400 years ago. News: we can
stop now.
Whatever its origins, in order is white noise in any sentence.
Run the same acid test on the meaning of in order to _______ by
covering in order wherever you find it and reading the sentence
140 CONCISION: MAXIMIZING EFFICIENCY

without it. You’ll discover in order contributes absolutely bloody

zero to any sentence’s meaning. To weed the expression out com-
pletely, run a Find and Replace search on your entire document for
in order and systematically delete it. You’ll need to use tools like
Find and Replace because in order to ______ is so ubiquitous that
spotting it is rather like seeing air. You know air is around you. You
just need help in seeing it.

EXPERT TIP: Prefer short and simple to lengthy and complex

For clarity, choose short and simple words – the ones preferred by
George Orwell in “Politics and the English Language” over those
invasive, Frenchified species. Short, simple words give sentences
concreteness and also invite readers to see cause and effect. For con-
cision, short and simple words have an extra benefit. We read sim-
ple, easy-to-pronounce words more rapidly and recall them far more
clearly than we do more complex terms, a good reason for preferring
use to utilization.
Once he discovered that shorter words led to sharper recall, Alan
Baddeley and colleagues aimed to isolate the reason behind the im-
proved recall. Hypothesizing that readers needed to subvocalize
longer words (that thing you do as you move your lips when you first
encounter, say, defenestration), Baddeley prevented readers from re-
hearsing any of the words. Rehearsing longer words, Baddeley and his
collaborators supposed, together with the words’ greater complexity,
overloaded some part of the brain’s speech systems, which could ac-
count for the fall-off in recall of these more complex words compared
to simpler ones. In addition, Baddeley’s study also controlled for the
complexity and potential unfamiliarity of words on readers’ ability to
recall them. To do so, he and his colleagues generated two sets of
words with the same numbers of syllables, letters, and phonemes. But,
in their study, one list had words that readers would have needed more
time to articulate, had they spoken them aloud. Even after discounting
for these alternative effects, the study revealed something startlingly
simple. Words that needed to be spoken slowly were harder to remem-
ber.31 Orwell was, in a way, right after all. Most words with their roots in
Old English tend to be simple and more memorable than their Latin or
Frenchified counterparts.
 TAKEAWAYS FOR CONCISION 141

Takeaways for concision: what you need to remember to make your

writing efficient
• Eliminate redundant modifiers and pairs.
• Prefer positives to negatives whenever possible.
• Avoid hedging, amplifiers.
• Use frame markers and evidentials in the main parts of your
sentences, not as extra phrases.
• Only use self-mentions or attitude and engagement markers to
create a conversational tone or rapport with readers.
• Weed out throat-clearing.
• Prefer short, familiar words to longer, less common words.
CHAPTER 7

Making music with words

The fifth C: Cadence

In this chapter you will

• learn how our brains recognize the rhythm of sentences even
when we read silently
• know how to handle items in a list or series
• discover how to create a sophisticated cadence to your
sentences by varying their length and structure.

In our clumsy and unscientific way of assessing writing that works –

or spectacularly fails to – we resort to using vague terms to diag-
nose issues that neuroscience, psychology, and linguistics can
help us pinpoint. When, for example, readers fuzzily describe a
paragraph as just not flowing, as we saw in Chapter 4, they’re
identifying discontinuities between sentences. But they’re also
straining to put a finger on something harder to pin down, the
thing that most books on writing are content to breezily label
style. Unfortunately, most of us aren’t aiming for the next Nobel in
literature (or even favorable critical reviews savoring our prose),
so glowing explanations of how writers avoid clichés, use bra-
cingly fresh analogies, or powerful rhetorical devices like antitheses
are noticeably less helpful in getting our proposals or memoranda
out the door than great lashings of caffeine. At 3 am, you don’t
give a shit whether you sound like Richard Dawkins at his most
sparkling and original. You just want to avoid sounding like an
illiterate, sleep-deprived, nine-year-old chucking together a report
hours before your deadline.
Follow the four Cs and you’ll end up sounding as though
you’ve at least had the benefit of an education, in addition to en-
suring your writing project will receive generous amounts of time,
thorough attention, and thoughtful consideration. Notice if I’d
 VARY SENTENCE STRUCTURE [CADENCE PRINCIPLE #1] 143

said … in addition to giving your writing project generous amounts

of thorough attention, thoughtful consideration about the mating
habits of wombats, and time, you’ll tilt toward the illiterate, sleep-
deprived and time-pressed nine-year-old end of the spectrum,
rather than away from it. Why? Our notions of flow and style alike
have one thing in common that few of us (aside from rhetoricians)
consider – cadence.
Open any book on writing or even scrutinize studies of reading
or writing, and you’ll be hard-pressed to find any references to ca-
dence, outside of three areas: studies of rhetorical devices, certain
categories of stroke, and of how beginning readers in their native
or second language tackle unfamiliar words. Cadence is currently
the province of rhetoricians, neurologists, and linguists obsessed
with phonology, all of whom spend their lives immersed in issues
that concern cadence or the rhythms of our speech. Notice, for
starters, that the key word here is speech. In contrast, in writing,
cadence mostly resembles death – the thing we all know exists but
seldom witness and lack a vocabulary to discuss without resorting
to weird approximations and euphemisms. Just as we use passed
or gone to describe death, as though your dead grandfather merely
ducked into the john and will be back momentarily, we use words
like choppy or awkward to clumsily identify problems tied strictly
to cadence.
And yet cadence is central to our sense that something is
particularly well written, beyond clarity, continuity, coherence,
and concision. Even if you observe nearly all the principles tied
to these first four Cs, if you write in nothing but short, simple
sentences, you can still come off like that mostly illiterate nine-
year-old.

Cadence Principle #1:

Vary the structure of your sentences

Sentences sound monotonous when you begin every one with

either your grammatical subject or words like the or this. Begin
some sentences with a transition or a brief phrase, or, occa-
sionally, an introductory clause. Compare the following two
examples:
144 CADENCE: MAKING MUSIC WITH WORDS

Example A:
Your sentences seem to lack syntactic complexity. Your
reader can feel jack-hammered by your sentence structure.
Your reader might guess that you stopped reading even
cereal boxes long ago. Your sentences all observe clarity
and continuity principles. And yet something feels wrong.
Something is wrong. Your sentences are all the same type –
simple. Your sentences all begin the same way. Your
readers perceive your writing as the equivalent of “Twinkle,
Twinkle, Little Star” rather than something by
Rachmaninoff. You’re not out to win the Van Cliburn
International Piano Competition. You just aim to sound
authoritative, not amateurish or idiotic.

Example B:
When you ignore cadence, your sentences will seem to lack
syntactic complexity. Your readers can feel jack-hammered
by your sentence structure, and they might guess that you
stopped reading even cereal boxes long ago. Note that all these
sentences observe clarity and continuity principles –
yet something feels wrong. In fact, something is wrong.
Your sentences are all the same type – simple. (If you’re
flummoxed by the difference between simple, compound, and
compound-complex sentences, see the Supplement, page 164).
For starters, your sentences all begin the same way, with not
only the same words but also the same sentence structure:
Subject–verb. Subject–verb. Subject–verb. Moreover, with your
over-reliance on one kind of sentence structure and length,
your readers can perceive your writing as the equivalent of
“Twinkle, Twinkle, Little Star” rather than something by
Rachmaninoff. While you’re not out to win the Van Cliburn
International Piano Competition, you do, however, aim to
sound authoritative – not amateurish or idiotic.

Aside from the bit of parenthetical advice on the definition of

simple sentence and the introductory adverb clause in the first
sentence, When you ignore cadence, all other changes between
Examples A and B are trivial in terms of meaning. Example B
 VARY SENTENCE STRUCTURE [CADENCE PRINCIPLE #1] 145

only offers several transitions, a few bits of jazzy punctuation, and

changes in sentence structure, complexity, and length. However,
try reading the two examples aloud, preferably to someone who
hasn’t read this book.
If you feel a bit like you’ve been sledge-hammered by the rhythm
of Example A, you’re not alone. Nearly every sentence in the first
example begins with the grammatical subject and – worse – with
the same word, You, varied only by the delightfully unpredictable
insertion of Your and the shocking novelty of Something beginning
one sentence. But no matter how you size these sentences up,
they have the same, relentless structure, as though pounded out
on a die: Subject–verb. Subject–verb. Subject–verb. Subject–verb. In
addition, every sentence in Example A is a simple sentence, con-
fined to just a major clause. Both the rhythm and similarity of the
sentence type and beginning have more than a little of the elemen-
tary school My Dog Spike-sound to them. Generally, this sort of
monotony is the hallmark of a careless or inexperienced writer –
or of someone pounding away at the keyboard with a scant three
hours remaining to a thesis’ drop-dead due date.
In contrast, Example B begins with a complex sentence (a
minor coupled with a major clause), followed by a compound-
complex sentence (two major clauses joined by a coordinating
conjunction with a minor clause embedded in the second major
clause). The passage rolls on with variations of complex, com-
pound-complex, and even a simple sentence in the mix. In addi-
tion, the opening clauses of the compound-complex sentences,
as well as an ample use of transitions, ease readers into the sen-
tences from both cognitive and rhythmic perspectives. Moreover,
this variation in sentence structure is one of the features that dis-
tinguishes polished writing by a writer in control of his game from
writing by someone who sounds as if he might belong back in the
My Dog Spike leagues.

EXPERT TIP: Just use transitions, period

Observing Continuity Principle #2 will also even out the cadence of your
sentences, eliminating the risk of your writing sounding monotonous,
choppy or, worst of all, as though a sleep-deprived and not particularly
well-read nine-year-old produced it.
146 CADENCE: MAKING MUSIC WITH WORDS

Cadence is always with us

While the authors of books on writing wax endlessly lyrical about

matters of style, which seems bound up nebulously with fresh
phrasings, adroit metaphors, and sentences that gracefully unfurl,
no one tells us why in god’s name we would hear something we
read silently. Both scholarly and popular titles on style remain si-
lent on how and why we hear cadence in signs on a page, all read
without speaking. And the silence is widespread for a bloody good
reason. To explain cadence scientifically is extraordinarily com-
plicated and involves delving into the history of the written word
and the specific areas of the brain that work to produce and com-
prehend spoken and written language alike. In other words, you
need more than an appreciation of Shakespeare’s use of iambic
pentameter or the artful way F. Scott Fitzgerald’s The Great Gatsby
genuflects linguistically to Joseph Conrad’s Heart of Darkness. You
need to understand at least a scrap of neuroanatomy and possess
a willingness to engage with neurologists’ debates about the parts
of the brain that may be involved every time we drag our eyes
across a line of text.
For all our silent reading and zapping emails in the throes of
some effortful standing on the elliptical at the gym, we hear what
we read – one reason why c u l8or remains comprehensible rather
than reading like a rune. As early as 1512, Renaissance scholar
Erasmus, writing in Latin, advised writers to focus on variation
in their sentences, counseling writers to take a group of their sen-
tences and phrase them in as many variations as possible.1 By
1926, H. W. Fowler was nothing short of rabid on the subject of
rhythm. Modern English Usage is straightforward and bracingly
opinionated on issues like the use of the word queer (“It has be-
come dangerous to apply this apparently innocent adjective to a
person …”). But when Fowler gets to the topic of rhythm, his entry
reads the way the Nile floods its banks, spreading a bounty of
edicts and exhortations over pages in columns of mouse type that
bring to mind the King James Bible:
[L]ive speech, said or written, is rhythmic, and rhythmless
speech is at the best dead … So it is that the best prose
writer’s guide to rhythm is not his own experiments … but
an instinct … cultivate[d] on one condition only – that [he]
 CADENCE IS ALWAYS WITH US 147

will make a practice of reading aloud … [R]eading aloud

need not be taken so literally; there is an art of tacit reading
aloud … reading with the eye and not the mouth, that is, but
being as fully aware of the unuttered sound as of the sense.2

And that’s the Morse Code version of the entry. Fowler’s strident
insistence on rhythm seems peculiar to Americans, who seldom
read aloud after their earliest years in school. His insistence on the
virtues of reading aloud might seem potentially a bit much even
to Brits made to suffer untold humiliations at the hands of their
teachers for stumbling over bits of Gerard Manley Hopkins as they
read aloud at school. However, his advice has surprising scientific
weight behind it, much of it undreamt of by Fowler, for all his
volubility on the subject.
We “hear” the words on the page for three reasons. First, we
rely on sounding out words or phonological processing during
the lexical or word-identifying stage of reading. This phonological
phase sticks with us beyond our earliest stumbles in learning to
read, surfacing whenever we encounter new or unfamiliar words
or tackle a new language. Second, during the long history of the
written word, reading was far from silent. Reading was public, vo-
cal, and voluble. And, third, our brains use overlapping auditory
and visual processing in the act of reading, particularly during our
early attempts at learning to read. Our brains “hear” prose when
we encounter too many short, simple sentences or strings of
lengthy, complex sentences – the outliers that violate our sense of
cadence. To learn just how profoundly our brains “hear” language,
we only need to consider the brains of readers of sign language
and braille, as well as the brains of stroke patients with neural
deficits that impair one of three primary areas implicated in using
and reading language.

Before reading was silent

When he read, his eyes scanned the page and his heart
sought out the meaning, but his voice was silent and his
tongue was still. Anyone could approach him freely and
guests were not commonly announced so that often, when
148 CADENCE: MAKING MUSIC WITH WORDS

we came to visit him, we found him reading like this in

silence, for he never read aloud.3

The man who would later become St. Augustine thought the
sight of the future St. Ambrose reading silently so remarkable
that the thirty-year-old St. Augustine recorded this soundless
reading in his Confessions. In AD 383, silent reading was so
rare that St. Ambrose represented St. Augustine’s first encoun-
ter with it. Moreover, in the West, silent reading would not
become commonplace until the tenth century.4
However, reading aloud lived on far beyond that century
in universities where, until considerably after Gutenberg over-
hauled an old wine press to print Bibles, books remained scarce.
Students’ encounters with books came from that now-familiar
staple of education, the lecture. In fact, the very word comes
to us courtesy of Latin, lectio, where the word actually means
reading. Teachers passed on the wisdom of others by reading
aloud to students from books still expensively created by hand
in medieval scriptoria. In other classrooms, students attended
dictation sessions to make up for the paucity of texts.5 The act
of reading aloud for centuries was impelled not by any impulse
toward being sociable – after all, when you slept, screwed, and
chewed within earshot of your family, sociability wasn’t exactly
thin on the ground – but by economics. Books were pricey and
scarce, lumbering students with the need to learn the wisdom
of theologians and philosophers through a lecturer’s droning
reading aloud.6

Why do we “hear” written language? Three neural

explanations
Our ancestors depended on reading aloud for their education and
for learning about the world outside their village. Or, in the case
of my illiterate great-grandfather, outside the Glasgow docks. But
as early readers, we also replicate this reliance on the relationship
between hearing speech and reading. In addition, we rely on what
we’ve heard spoken conversationally around us to begin decipher-
ing what we see on the page. Of all our adaptations as a species,
 CADENCE IS ALWAYS WITH US 149

the widespread use of reading and writing is perhaps our most

breathtaking – and not for our having produced King Lear or Mid-
dlemarch or Moby-Dick. After all, the same system of alphabetic
script also gives rise to The National Enquirer, hardly material
you’d want to hurl into the galaxy aboard a capsule as representa-
tive of humankind’s greatest achievements. Instead, reading and
writing require a complex ballet within our brains. That first step
in the three-step dance of reading, the lexical or word-identifying
phase, requires our brain’s visual areas to break down words into
letters and graphemes. Graphemes are the smallest components of
words that include roots of words, prefixes, and suffixes, all useful
in figuring out what the hell defenestration means the first time
we come across it.
Think back to the last time you came across an unfamiliar
word. Can’t remember? Try this one, then: glomerulonephritis, a
word I can never come across in the classroom without my lips
moving to sound it out. Every time I do it, I’m doing that ontogeny-
recapitulates-phylogeny shuffle because I’m reliving the birth of
reading in phonological processing or translating the graphemes,
bits of words, into sounds or phonemes. We first do this phono-
logical, lips-moving, muttering translation of marks on the pages
into bits of sound with lowly words like place during our earliest
attempts at reading. And, if you’re an intellectual carpetbagger,
you can manage to stay mired in this phase into your dotage by
coming across words like glomerulonephritis. For place or even
the perversely spelled poignant, hearing the word aloud is suf-
ficient for us to grasp what it means because we’ve heard others
use the same word in speech. For glomerulonephritis, however,
that translation of graphemes into phonemes helps us break the
word down into its meaning, as if we’re breaking a suspect under
interrogation. Stare at the guy sweating under the bright light,
and, eventually, something’s gonna give. Here -itis tells us we’re
dealing with an inflammatory disease, -nephr- tells us kidneys are
involved, and the really tricky bit, glomerulo-, with the aid of a
dictionary, resolves into a problem with the minuscule, looping
blood vessels within the kidney.
Every time you struggle to learn to read in a new language or
encounter a novel word, you engage a route to reading you other-
wise abandoned once you stopped reading while moving your
150 CADENCE: MAKING MUSIC WITH WORDS

lips. You use the brain’s visual areas to break marks on a page
into letters, then units of words, and, finally, words. But you do
so expertly and silently. However, to understand how these let-
ters and bits of words work together, your brain must also rely
on areas of the brain reserved for both speech and hearing.7 This
concerted coordination of visual, auditory, and speaking areas of
the brain in the act of reading is what makes widespread reading
and writing one of the more marvelous inventions of our spe-
cies. Despite our being non-nephrologists or even non-scientists,
we can decipher and make sense of even a sentence dense with
unfamiliar terminology:
Glomerulonephritis is an inflammation of the kidney
characterized by proteinuria, impaired filtration, and the
accumulation of extracellular matrix within the damaged
glomeruli.

When we emerge at the end with more than a glimmer of com-

prehension – at least, the non-biomedical types among us – we
engage in an activity that repurposes parts of our brains never de-
signed to read the Daily Mail, let alone Nature or an explanation of
how credit default swaps work (or mostly don’t). Unfortunately, to
explain how reading and writing harness and, depending on your
view, repurpose or even rewire our brains, we need to examine
three hypotheses of how the reading brain perceives speech that’s
frozen and silent on the page.

Explanation 1: Speech, auditory, and visual systems

work together in reading or this is your brain saying
glomerulonephritis
In 1993, researchers used positron emission tomography – known
to the rest of us as PET scans involving significant amounts of
radioactive material – to measure cerebral blood flow during si-
lent reading. Even though the experimental subjects never spoke
a word, PET scans revealed increased blood flow to areas previ-
ously believed to be dedicated solely to the physical or motor side
of speaking: the supplementary motor area (known as the SMA)
and cerebellum. Researchers expected to see increased blood
flow, indicating specific areas of the brain were at work, in the
 CADENCE IS ALWAYS WITH US 151

left lingual gyrus, a part of the brain that, when damaged by a

stroke, prevents right-handed patients from being able to recognize
words, let alone read. This area researchers had suspected must be
associated with some visual processing of letters. However, they
were floored to find blood flow also increased to Broca’s area, a
part of the brain hitherto believed to be responsible only for un-
derstanding and forming spoken words.8 Even during silent read-
ing by expert readers of familiar words and phrases in a noiseless
environment, their brains apparently relied on the sounds of words.
Together, the involvement of Broca’s area, the left lingual gyrus,
SMA, and cerebellum all signify an overlap with our auditory and
visual systems, as well as the motor and comprehension systems
for speech. Even seasoned readers who consume the New York
Times with breakfast apparently rely on a loop for articulating and
hearing speech during silent reading. This mechanism preserves
the order of words, phrases, and clauses, at the same time that the
speech–auditory–visual loop helps us detect errors and wend our
way through particularly challenging bits of prose – like an explan-
ation of how credit default swaps work.

An extraordinarily brief tour of the reading/speaking brain

Talk about reading and speaking with any neurologist, and

you’ll mainly hear about two areas of the brain: Broca’s and
Wernicke’s areas. Broca’s area, once thought to deal mainly
with our ability to speak, takes its name from Pierre Paul Broca,
who identified an area in the brain’s frontal lobe as responsible
for robbing two patients of their speech after injury to the pos-
terior inferior frontal gyrus. While some people with damage
to Broca’s area can speak, they speak in telegraphic shorthand,
packed with nouns but mostly devoid of function words like
of and from and, rather more important for communication,
also lacking in syntax. On the other hand, Wernicke guessed
at a link between the area of the temporal lobes named for
him and another type of aphasia, this one involving the abil-
ity to comprehend speech or to produce speech that conveys
meaning. Someone with Broca’s aphasia battling a severe sinus
infection might say, helpfully, “Nose,” as her entire explanation
152 CADENCE: MAKING MUSIC WITH WORDS

to her doctor. In contrast, a patient with Wernicke’s aphasia

would produce strings of functional words and lovely syntax,
all sound and fury signifying nothing that her doctor or any of
us can make sense of. “Well, there’s smelling, a lack sensate,
only a mucus and tile then tonations and inability to trinagu-
late the cogulatory.”9 If you think I’m exaggerating, try reading
transcripts of what actual Wernicke’s patients say, and, by com-
parison, my attempt will seem as comprehensible as an entry in
the Oxford English Dictionary.

Frontal lobe
Motor cortex
Broca’s area
Arcuate fasciculus
Parietal lobe

Angular gyrus

Temporal lobe
Wernicke’s area
Occipital lobe

Explanation 2: Your visual, speech, and auditory centers

are hard-wired together
Viewed in terms of evolutionary time, humans have been reading
for a mere sliver of our species’ history. And we also spend so
little of our days reading – some of us, at least – that our brains
are unlikely to have formed a neural circuit dedicated solely to
making sense of, say, instructions for operating a Taser. Instead,
reading seems to have become possible from connections among
 CADENCE IS ALWAYS WITH US 153

the areas of the brain enabling us to translate visual marks into

speech, areas enabling us to speak, and areas that convert torrents
of sound into sentences with syntax. Broca’s area (think: rhythm,
affect, and syntax) in the frontal lobe and Wernicke’s area (think:
words and meaning) in the temporal lobes are linked by a band
of fibers called the arcuate fasciculus. Moreover, the other part
of the brain central to language, the angular gyrus, is helpfully
positioned at the bottom of the parietal lobe at a point where the
occipital (visual functions) and temporal (auditory functions)
lobes intersect. In fact, our brains recruit more blood to the angu-
lar gyrus in both the temporal and occipital lobes when we read
relatively unfamiliar words like exsanguinate or epistemology. But,
surprisingly, introduce a pseudo-word, resembling the kind pro-
duced by someone with Wernicke’s aphasia – say, cogulatory –
and the brain sends more blood to both Wernicke’s and Broca’s
areas. Because we fall back on trying to sound out unfamiliar
words to make sense of them, we recruit speech, hearing, and
visual areas. Except, in this hypothesis, borne out by PET scans
of brains of subjects reading single words and lone pseudo-words,
this circuit exists because these areas are interconnected by prox-
imity and the arcuate fasciculus.10 You can read this sentence, ac-
cording to this view, because the visual and speech processing
areas of the brain have both a junction box in the form of the
angular gyrus and hard-wiring via the arcuate fasciculus.
The angular gyrus uses visual associations that enable us to
work out the meanings of words by separating the in- and un- and
de- beginnings of a word from its root and -ation ending, acting
as a visual dictionary that matches the words on a page with our
knowledge of what they mean. In right-handed readers, we rely
on the left angular gyrus when we read familiar words. But we
also use Broca’s and Wernicke’s areas when we encounter words
like those in Lewis Carroll’s “Jabberwocky” that includes pseudo-
words which native speakers of English recognize as unlikely to
be actual words: brillig and toves because we see letters clustered
together in formations unlikely to exist in real English words. The
-ig in brillig fails to crop up in words we know, just as the tov in
toves fails to conjure up any word root in English. In one study
of skilled and dyslexic readers, pseudo-words cause us to fall
back on our earlier sounding-out strategies in reading because the
154 CADENCE: MAKING MUSIC WITH WORDS

angular gyrus fails to identify the word on the page with our visual
dictionary of words and their meanings.11

Explanation 3: Neuroplasticity – reading and writing wired

your visual, speech, and auditory centers together
We’ve all heard two old saws, or, at least, the undeniable geeks
among us have. First, neurons that fire together, wire together.
And, second, that, when we lose one sense, our other senses
mysteriously become sharper. We now know that neuroplasti-
city exists in the language areas of the brain, courtesy of some
striking studies of parts of our brains becoming repurposed for
other functions. If you were born without sight, when you read
braille, your visual reading areas light up – even though you’re
relying solely on your sense of touch to read.12 Similarly, if you’re
congenitally deaf, the supratemporal gyri, normally reserved for
processing sound, instead get hijacked for understanding sign lan-
guage. But, if you lose your hearing after learning how to talk,
sign language engages only the visual processing areas of your
brain. Our brains may well have recruited areas for speech and
hearing to enable us to understand written language, following the
evidence of our brains’ neuroplasticity in the experiences of braille
and sign users who rely solely on touch or sight in understanding
communication.13

No matter how you look at it, reading is a

multi-input process

If you survived these for-geeks-only explanations, you might have

spotted that none of the three explanations rules out the validity
of the other two. In other words, we can read because we can
sometimes fall back on the old phonological–lexical loop in read-
ing unfamiliar words, just as we did in elementary school when we
learned how to read – especially if you’re reading new terminology
that you’ll have to use later in discussions with experts. And we
can read because the parts of our brain that now turn “Whoops! I
did it again!” into sound and sense are interconnected. And we can
also read due to our brains’ ability to adapt to the demands placed
on it, even within a single lifetime or relatively brief span of years.
 VARY THE LENGTHS OF SENTENCES [CADENCE PRINCIPLE #2] 155

If you’re still wondering what the hell this detour into neuro-
anatomy has to do with cadence, the answer is everything. Even
when you’re one of those showy sods who can consume an en-
tire issue of The American Scholar during the über-predictable
flight-safety information preceding take-off, you’re still hear-
ing the sound of words on the page. If you’re reading the New
York Review of Books, the non-uniform lengths of sentences –
those bracing simple sentences, bracketed by compound and
compound-complex sentences that go on a fair old bit – will tele-
graph to your reading brain a pleasing cadence, one caused by
a lack of monotony in sentence length. If you’re reading a legal
statute or doctoral dissertation in the humanities, the pages
whisper to your auditory and speech processing centers of the
droning monotony of apparently endless sentences, heaving with
semicolons, colons, commas, and parentheses. Or, if you’re read-
ing anything by Hemingway or Twitter feed, you “hear” staccato,
almost entirely simple sentences of similar lengths, as in this ex-
ample from one of Hemingway’s The Nick Adams Stories:
“Oh, well. He’s pretty small,” his father said.
“That’s no reason to bring him into the woods with us.”
“I know he’s an awful coward,” his father said, “but we’re all
yellow at that age.”
“I can’t stand him,” George said. “He’s such an awful liar.”
“Oh, well, forget it. You’ll get plenty of fishing anyway.”14

Cadence Principle #2:

Vary the lengths of your sentences

Even if you use transitions, you still might not be safely out of
the “What I Did on My Summer Vacation” stakes if you fail to
vary the lengths of your sentences. Sentences of roughly the same
length can put your readers to sleep or make your sentences seem
choppy. Exceptions here include novelists or professionals strain-
ing to imitate Hemingway and perfectly happy to face the prospect
of the dole queue after a pitch fails to charm clients the way The
Sun Also Rises has generations of American writers. For instance,
compare the following examples:
156 CADENCE: MAKING MUSIC WITH WORDS

Example A:
People enjoy corporate-owned coffee joints. Even people
who don’t enjoy drinking coffee enjoy corporate coffee
joints. They love this magical land where the word “tall”
equals small. People love using words in Italian and not
knowing what they mean. They like how every store has
the same atmosphere. There’s a certain comfort to be found
in familiarity.

Example B:
People enjoy corporate-owned coffee joints. Even people
who don’t enjoy drinking coffee enjoy corporate coffee
joints. They love this magical land where the word “tall”
equals small, where they get to use words in Italian and
not know what they mean. They like how every store has
the same atmosphere and find a certain comfort in
familiarity.

If you’re not reading attentively, these two paragraphs might

seem like the old “Spot the Difference” games you played when
you were a kid, intently scanning two drawings that looked, at
first glance, nearly the same. But, if you’re reading reasonably
carefully, you’ll notice that the third sentence in Example B is
nearly two lines long – about twice the length of every other sen-
tence in this little paragraph – and the final sentence combines
what were two short sentences in Example A. In contrast, Exam-
ple A’s sentences are all nearly the same length – a little under a
line. The result: in Example B, the student writer’s short, snappy
sentences seem as if she crafted them for deliberate effect, their
rhythm balanced by the longer and more complex third sentence.
In Example A, however, readers could get the impression that the
writer lacks the ability to write a fluid, reasonably complex sen-
tence. The choppy sentences seem a tad on the simplistic side –
and their rhythm is monotonous. One reason why Hemingway
took the literary world by storm was his journalistic prose, a dra-
matic and telegraphic break from the you-need-eight-breaths-to-
read-this-sucker sentences penned by the likes of Henry James,
who was still alive when Hemingway was a reporter for The
Kansas City Star.
 VARY THE LENGTHS OF SENTENCES [CADENCE PRINCIPLE #2] 157

Cadence matters even in fiction – perhaps most of

all in fiction

We can amuse ourselves, after stalwartly bearing up under so

much neuroscience, with a brief comparison of the thing Hem-
ingway might have been rebelling against in that excerpt from
The Nick Adams Stories, quoted above. To avoid inflicting un-
due pain and suffering on writers who remember Henry James
only as the guy who wrote that spooky story that inspired a few
films called The Turn of the Screw, we’ll consider one of James’
more palatable works. James wrote The Aspern Papers during
an era when he wasn’t busily embedding clauses within clauses
within phrases and joining apparently endless sentences with
colons and semicolons. In other words, he hadn’t yet got to the
sort of writing you produce after you complete a PhD or DPhil:
Their name had been mixed up ages before with one of
the greatest names of the century, and they lived now
in Venice in obscurity, on very small means, unvisited,
unapproachable, in a dilapidated old palace on an out-
of-the-way canal: this was the substance of my friend’s
impression of them.15

But, predictably, Hemingway summarizes the entirety of The

Aspern Papers in considerably fewer words and sentences than
James needs to preface the introduction of one of the novel’s
main characters:
Did you ever read Henry James? He was a great writer who
came to Venice and looked out the window and smoked his
cigar and thought.

Tellingly, The Kansas City Star’s guide for reporters gave advice
to its journalists that applies to everything Hemingway wrote
after his stint there as a cub reporter: “Use short sentences. Use
short first paragraphs. Use vigorous English. Be positive, not nega-
tive.”16 Note to anyone writing anything today: use this advice
if you want to seem like an imbecile or a seven-year-old not ter-
ribly keen on this dreadful thing called writing. Even on Twitter.
158 CADENCE: MAKING MUSIC WITH WORDS

To vary the lengths of your sentences, first assess the typical

sort of sentence you tend to write. Most of us lean toward either
relatively short sentences or sentences that stretch for lines, spill-
ing into cascades of clauses and little puddling phrases. If you
tend to write on the short side, try joining two shortish sentences
together. Rather than simply using and to join your sentences, try
tying sentences together via phrases or short, modifying clauses.
The undergraduate who wrote Example B could have simply tied
the sentences together by saying, They love this magical land
where the word “tall” equals small, and they love using words in
Italian and not knowing what they mean. Instead, she chose to
unite the two ideas more subtly, tacking on another where clause
to parallel the first one: They love this magical land where the
word “tall” equals small, where they get to use words in Italian and
not know what they mean. Her writing signals to her readers that
they’re in the hands of an accomplished writer who has a way
with stringing sentences together.

Cadence in the brain is also real

Our ability to make sense of a stream of syllables is remarkable, as

you discover quickly when you toddle into the beginning stages
of learning a second language. You get along swimmingly in
forming sentences like Où est le petit coin? which can give you
the slangy sort of cachet you achieve as a non-native speaker of
French the first time you ask for the john, rather than the toilet or
that euphemistic American alternative, the rest room. Then you
watch a Parisian police procedural, where every third word is ver-
lan slang or listen to a French pop song and want to request the
nearest petit coin – the slangy term for that little corner where the
medieval peasants relieved themselves – so you can stick your
head in the bog to drown your despair at ever making sense of
French.
But, hell, you don’t even need to venture into a second lan-
guage to encounter the cognitive strains of comprehending speech
gone terribly wrong. Witness the mishearing of “Hark, the herald
angels sing,” which a childhood friend insisted was “Dark the
 CADENCE IN THE BRAIN IS ALSO REAL 159

hair, old angels sing,” an interpretation that makes rather more

sense than the original lyrics when you think about it. Or the
coda in Michael Jackson’s “Wanna Be Startin’ Somethin’,” which
includes the line “mamako, mamasa, maka makossa.” Most of
us probably equated those sounds with the Squeeze lyric in “If I
Didn’t Love You”: “Cocoa mugs sit side by side.” A friend sang
that line as though Squeeze had written scat with only vague
aspirations toward sensible English, which is how most listen-
ers interpreted the coda to “Wanna Be Starting’ Somethin’.” In
fact, Jackson’s “mamako, mamasa, maka makossa” was Duala,
not English at all, and handily borrowed from a 1970s Cameroon-
ian dance hit. Predictably, because the borrowing involved an
American with deep pockets, a lawsuit eventually ensued. But
until 2009, few of us knew we were singing scraps of Duala.17
With their blending of musical and speech cadences in lyrics,
songs remain the testing ground where even native speakers can
realize the limits of their abilities to disambiguate one word from
another.
Music also gives us two further clues about how deeply
entangled cadence is in our ability to understand speech, both
spoken and written. Damage to Broca’s and Wernicke’s areas have
two striking impacts on our perception and production of ca-
dence. Someone who has Broca’s aphasia speaks, if at all, with
an absence of functional words or meaningful syntax.18 Yet, startl-
ingly, people with Broca’s aphasia can sing with all the words
correctly articulated if the stroke impacts the left hemisphere,
where most of us (the right-handed among us, at any rate) pro-
cess language after we learn to read.19 Reverse the location of
the stroke and impact the right hemisphere, and the aphasia
leaves patients able to speak normally but unable to sing, chant,
or recite even daily prayers familiar from childhood onward.20
Furthermore, and perhaps more crucially for our perception of
cadence and rhythm, Broca’s and Wernicke’s areas seem cen-
trally involved in our ability to perceive tone and harmony in
music. Patients with either Broca’s or Wernicke’s aphasia, de-
pending on the hemisphere affected, have marked inabilities
to perceive tones and harmonies in music easily perceived by
others.21
160 CADENCE: MAKING MUSIC WITH WORDS

The upshot: your reader will hear your writing. Always

Just practice the three cadence principles here, the fifth C, as

religiously as you will the other four Cs. The same mechanisms
that enable us to read, even silently and speedily, transmit the
sounds from the page and into our heads.

Cadence Principle #3:

In a list, series of phrases, or entire sentence, place
the item with the least number of words and syntactic
complexity attached to it first, with the longest item, last

Ever feel like some sentences screech to a halt, like a beginning

driver stomping on the brakes in Drivers’ Education? Or can you
recall some lists that seemed to stumble to a conclusion, rather
than winding down gracefully? Both sentences and lists with these
characteristics suffer from poor cadence. In this instance, how-
ever, the problem is more widespread than a jarring rhythm in a
sentence we expect to roll out smoothly. When you place the most
complex items at the beginning of a list, you knock your reader’s
sense of cadence off-kilter. Compare, for example:
Example A:
We ended the day with a recall of the week’s events:
the days spent digging ditches, swatting at plane-sized
mosquitoes, squinting into driving rain, and work.

Example B:
We ended the day with a recall of the week’s events: the
work, the days spent digging ditches, squinting into driving
rain, and swatting at plane-sized mosquitoes.

Example A’s off-kilter list strikes us as awkward for two reasons –

and these two reasons leave aside the entire issue of the list’s dis-
ruptive, jarring cadence. First, the rhetoricians in ancient Greece
were onto something when they insisted on using the principle of
klimax or what Romans termed, less provocatively to our sens-
ibility, scala. In a well-turned scala, items proceed in order of
 HANDLING ITEMS IN A SERIES [CADENCE PRINCIPLE #3] 161

importance (least-to-most), size (smallest-to-largest), or syntactic

complexity (simplest-to-most complex).22 More important, when
you shoehorn items into a series in any old order, you inflict sig-
nificant cognitive burdens on your reader’s brain. Remember, way
back with the first C, clarity, how readers needed to hold items
in working memory until they reached the verb and object? The
more detail you embed early in a list, the harder your readers’
brains must work to keep track of which word is playing which
grammatical and syntactic roles.23

EXPERT TIP: If you don’t regularly read something well written, start. Now
Studies have demonstrated that reading exerts a stronger influence
on writing than writing does on reading.24 Look carefully at the writing
of your colleagues and superiors, and you can tell who’s been read-
ing PowerPoint decks on YouTube videos and who’s been reading
The International Journal of Occupational and Environmental Health.
The lost soul who reads PowerPoint decks on YouTube for intellectual
enrichment will avoid writing anything aside from emails and, when
confronted with a writing task, will produce something that sounds
wretchedly like the “My Dog Spike” essays school children labor over.
In contrast, you can also easily spot the reader whose diet consists
solely of academic journals – and nearly all of them read as woodenly
as The International Journal of Occupational and Environmental Health
and many, far less comprehensibly. He or she will produce those ap-
parently endless sentences, as well as paragraphs that straddle three
pages, all rife with passive construction and cluttered with jargon.
In writing, as in speaking, we tend to adopt the style, vocabulary,
and rhythms, however non-existent they may seem, of the things we
read.25 On the other hand, readers tend to perceive our prose as
more sophisticated when sentences contain rhythmic cadence and
syntactic complexity, one hallmark of good writing on which most re-
searchers agree.26 If the last book you read was a textbook, subscribe
to one of the two notably well-written magazines in English, including
The Economist, Atlantic Monthly, American Scholar, The New Yorker,
The Times Literary Supplement, London Review of Books, or New
York Review of Books or dailies like the Guardian, New York Times,
and Wall Street Journal. Get in the habit of at least reading the inter-
esting bits over breakfast. And make a point of reading something well
162 CADENCE: MAKING MUSIC WITH WORDS

written before you write anything, where the language style matching
James Pennebaker observed in conversations applies equally strong-
ly with the word choices, sentence length and complexity, and overall
cadence of our writing.27

We can even see the impact of mangled syntactic complexity

in a sentence innocent of lists or anything faintly resembling a
klimax. The original writer of Example A might have intended
some snappy ending to a sentence after artfully creating tension
via syntax, a sort of whodunit encapsulated in a sentence – even if
the topic is genetic engineering described at its most prosaic level
and not in a cloning-Dolly-the-Sheep sort of way.
Example A:
Of the many areas of science important to our future, few
are more promising than a new way of manipulating the
elemental structural units of life itself, which are the genes
and chromosomes that tell our cells how to reproduce to
become the parts that constitute our bodies, or genetic
engineering.
Example B:
Of the many areas of science important to our future, few
fields are more promising than genetic engineering, a new
way of manipulating the elemental structural units of life,
including the genes and chromosomes that both instruct
our cells on reproduction and ultimately create our entire
bodies.

Example A stumbles to its knees around the same time the read-
er’s brain does, figuratively speaking. After we reach few, we fe-
verishly root around for the noun to anchor the tsunami of ap-
positive, gerundial phrases and adjective clauses that pile up, then
teeter, precariously, atop the minute noun, genetic engineering,
that, unlike Atlas, fails to hold up the miniature world stacked
above it. When you place the least complicated item first, genetic
engineering, your reader’s brain no longer strains to hold the sen-
tence open and keep syntactic complexity in place. In Example B,
 TAKEAWAYS FOR CADENCE 163

we can easily assign syntactic roles to the still-significant drifts

of clauses and phrases after genetic engineering because we’ve
already made sense of both the sentence’s basic meaning and its
sentence structure. We also have better recall of the modifiers that
introduce us to the wonders of genetic engineering. And we’re un-
likely to need a second reading of Example B, in contrast to Exam-
ple A. If the writer’s lucky, a forgiving reader actually dives back
into the pile to make sense of Example A. If she’s unlucky, the
reader decides the writing is unworthy of a close reading, skims
the rest, or simply chucks the entire thing.

Takeaways for cadence: what you need to remember to make your

writing rhythmic
• Vary your sentence structure.
• Use transitions at the beginnings of sentences to create varia-
tion in your sentence structure.
• Vary the lengths of your sentences.
• When introducing lists, put the simplest item first and the most
complex item, last.

Remember, for all of us, rereading is a luxury and effort we spare

only when stakes are high. We all face our queues of email waiting
for replies and exponential increases in communication – spurred
by how freely and quickly we can send out emails, entire books,
proposals, pitches, memos, texts, and the ubiquitous social media
updates. To hold our attention, you have to write well. The scar-
city of both our time and attention demands nothing less.
SUPPLEMENT

Everything you ever wanted to

know about grammar, punctuation,
and usage – and never learned

This short guide serves as a single reference for all of you who
were never taught grammar, punctuation rules, or usage by
(a) ruler-wielding nuns, (b) Old School English teachers born prior
to 1945, (c) parents who worked as high-school English teachers,
or (d) the necessity of mastering at least the grammar of another
language. Written after a lengthy search failed to turn up a single
guide that was brief, readable, and not rife with errors, I posted
this comprehensive guide online over fourteen years ago to serve
merely as a reference for my students tormenting themselves over
grammar and punctuation. Nearly a decade later, I received a query
from IBM’s head of in-house publications, requesting permission
to use “Where do you put the ___? Punctuation made painless,”
a rump version of the guide I posted online, That guide, she ex-
plained, was superior to every other source they’d investigated as
a guide for writers company-wide. Of course, IBM was only too
glad when I idiotically made the entire guide available to them –
gratis. In the same spirit, I’m now idiotically making the same
easy-to-access reference available to you, to complement the writ-
ing skills you’ve just mastered in the preceding pages.
Note that I’ve tried to use plain, commonsense terms instead
of the labels nuns and Old School types prefer. During the small
hours, when you’re fretting over the final draft of a proposal
or manuscript, you’ll be grateful for my using subject pronoun,
rather than nominative case. Likewise, when you need only know
whether a pronoun can act like a subject or an object and which
one does what, I avoid delving into the differences between per-
sonal, demonstrative, and relative pronouns. Ultimately, you need
to know how to use something correctly, rather than what the
 GRAMMAR MADE (RELATIVELY) PAINLESS 165

experts call it. Unless you’re aiming to compete on a quiz show.

Or put off someone you’ve met on a first date.

Grammar made (relatively) painless

Subjects
The subject is the main person, place, or thing in a clause or
sentence, always paired with a verb. You must know how to
locate the main subject of any sentence, as the dynamism and
clarity of your sentences rely on your using an actor or a tangible
object as your subjects in most sentences. To pinpoint the main
subject of any sentence, find the major clause in it. You can rec-
ognize major clauses, as these clauses can stand on their own,
without reading like either a sentence fragment or an unfinished
sentence.

Example:
Before he became the original James Bond, Sean Connery
had worked as both a bricklayer and a truck driver.

Note how the second clause can stand completely on its own, if
you sever it from the rest of the sentence:

Sean Connery had worked as both a bricklayer and a truck

driver.

If you try the same tactic with the first clause, however, the result
reads like a sentence fragment, a piece of a sentence that’s clearly
missing something:

Before he became the original James Bond.

Now find the main actor or thing in the sentence’s major clause –
the noun or pronoun that refers to the main action or verb in the
sentence.
Sean Connery is the main noun in the sentence, since that
proper noun refers to the main verb. He in the first clause
doesn’t count, as the first clause is a minor clause. Inciden-
tally, Sean Connery is a proper noun, since we’re looking at
166 SUPPLEMENT

the specific name of a person, place, company, product, or object.

Proper nouns are always capitalized. In contrast, common
nouns, which refer to general categories of objects, places, and
things, are never capitalized.

COMMON PROBLEMS WITH SUBJECTS

Subject–verb agreement refers to how the subjects and verbs of
sentences must be consistent in both gender and number. Gen-
der refers to masculine or feminine; number refers to singular or
plural. You can try saying, Before she became the original James
Bond, Sean Connery had worked as both a bricklayer and a truck
driver. However, you’re giving Sean Connery a sex change, albeit
purely a grammatical one. In the same way, you can try the Amer-
ican delicacy about gendered pronouns by using the genderless
plural their, but you’re violating subject–verb agreement. You can
say As every employee knows, they must squash their cigarettes at
least fifty feet from the building. But employee is singular, while
they is always plural. You can exercise delicacy or American po-
litical correctness in avoiding assigning a universal he to nouns
that refer to people by simply using the plural: As all employees
know, they must squash their cigarettes at least fifty feet from the
building.
1. When you have two nouns or pronouns as your subject linked
by and, the verb should be plural, since the subject is multiple:
The buoyant economy and astronomical growth of the
Internet have contributed to the attractiveness of investing in
high tech startups.

2. When you have two nouns or pronouns linked by or or nor

(usually preceded by either or neither), the verb is singular:
Neither Carlos nor Adele was able to accurately forecast the
market trend.

3. Be wary of collective nouns, which include the names of com-

panies and institutions, as well as groups. These usually refer
to groups acting as a single unit:
General Motors is holding its annual meeting in Dearborn.
 GRAMMAR MADE (RELATIVELY) PAINLESS 167

Note: For a nation of believers in rugged individualism, Americans

have a peculiar habit of using the singular to refer to companies and
organizations. In contrast, British speakers of English tend to refer
to companies in the plural, presumably because they are made up
of scores of people, not because Brits are a fractious lot who have
difficulty working in groups:

Cunard are pleased to offer their passengers the first-ever cruising

awards incentive programme.

4. Be wary when using their to escape identifying a person or

position with a gender when you’ve been using the singular.
While we commonly use this dodge in spoken English, in
standard written English pronouns and nouns must be consist-
ent in number and gender:

Wrong!
The employee had left their files in an utter disarray.

Correct versions:
The employees had left their files in an utter disarray.
The employee had left his files in an utter disarray.
The employee had left her files in an utter disarray.
The employee had left his/her files in an utter disarray.

5. Beware of seizing on the number or gender of the noun clos-

est to the verb and allowing that to determine the number and
gender of the verb:

The task, with its complex details and hours of mind-

numbing repetitions, seem endless. [The writer has relied on
repetitions to determine the form of the verb, which should
be seems.]

Pronouns
Pronouns replace nouns, relieving us of the need to repeat the
same noun in a sentence or string of sentences.
168 SUPPLEMENT

COMMON PROBLEMS WITH PRONOUNS

1. Pronouns can make unclear what they refer to, especially if the
pronoun refers to (a) a noun in an earlier sentence or, worse, (b)
the content of an earlier sentence. Your readers will have to hunt
through an entire preceding sentence or sentences, or, worse, may
assume you’re referring to a different noun entirely from the one
you intended. As a result, ensure your pronouns refer clearly to
only a single noun, preferably one reasonably close to the pronoun.
Wrong!
Guilt, bitterness, and cruelty can be emotionally destructive
to you and your family. You must get rid of them.
Correct versions:
Guilt, bitterness, and cruelty can be emotionally destructive
to you and your family. You must get rid of these emotions.
Guilt, bitterness, and cruelty can be emotionally
destructive to you and your family. You must get rid of
these destructive feelings.

Unless, of course, you actually mean to get rid of your family.

2. Even if you use a string of pronouns, you can make their refer-
ences clear by tying each pronoun to a single noun:
The office manager bought cheap, knock-off netbooks for
her assistants, but the netbooks fell apart quickly because
they were not suited to heavy use. [Readers otherwise
might think the assistants are unsuited to heavy use – not
a particularly pretty image.]

3. Place pronouns as close to their referents as you can, since the

more distance between the pronoun and the noun it refers to,
the greater the likelihood your reader will get muddled:

Wrong!
The statement that the supervisor made and that she
issued it as a formal policy inflamed the city council, who
knew it would result in widespread anger.
 GRAMMAR MADE (RELATIVELY) PAINLESS 169

Correct version:
The supervisor made a statement and the mayor issued a
formal policy. This policy, however, inflamed the city
council, who knew the new policy would result in
widespread anger.

4. Ensure that the pronouns it, this, that, and which refer to only
one referent, since multiple pronouns can make sentences
especially difficult to understand:
Confusing!
According to some sources, the Federal Reserve ought to
have interceded before Lehman Brothers crashed and
burned, and the markets responded with panic. That
contributed to the Dow Index losing more than 1,000
points in a single day.
Clear version:
According to some sources, the Federal Reserve ought to
have interceded before Lehman Brothers crashed and
burned, and the markets responded with panic. That
bankruptcy filing and subsequent market panic
contributed to the Dow Index losing more than 1,000
points in a single day.

5. Who versus that or which: Who always refers to people or

groups of people (even companies); that or which refers to
animals and inanimate objects, as well as some groups.

6. Some grammarians and people of a certain age, as the French

like to say, insist that should be used only for what we’ll call
defining adjective clauses, while which should be used solely
for non-defining adjective clauses. In a defining clause, the
clause identifies the object being modified absolutely, nailing
its identity down to a single thing:
The book that I had laid on the table has been stolen.

In a non-defining clause, the clause only provides more informa-

tion about the object being modified:
170 SUPPLEMENT

The book, which has grease spots on its dust cover, is one I
consult frequently.

While the Oxford English Dictionary – the bible of modern Eng-

lish usage and grammar – no longer observes the rule governing
when you use that and when you use which, some fogeys out
there will swear by it. However, one rule still does legitimately
refer to the difference between the two: never place commas
around a defining clause; always place commas around a non-
non-defining clause. You need the defining clause to identify a
specific noun, hence the lack of commas around it. In contrast,
you should bracket the non-defining clause with commas, as the
clause merely tells readers more about the noun – rather than fix-
ing the identity of the noun.

7. Pronouns fall into four cases that reflect whether they flag a
subject, object, possession, or self-reference.
a. Use subject pronouns when your pronoun is the subject of a
sentence or a pronoun to which the subject refers. Subject
pronouns include I, you, he, she, it, we, they, who, whoever,
this, that, which, these, and those.
If you’ve actually read Chapter 3, you already know that using
this, that, these, and those as grammatical subjects will kill the
clarity of your sentences. However, you can use this, that, these,
and those to modify nouns, which, for the record, turns them into
adjectives. Confused? Just keep reading.

I know of no other person in the company who is as

smarmy as he. [Here, he refers to an implied is commonly
left unstated after an as clause.]
Yes, this is she. [She is a predicate pronoun, a pronoun
that modifies the subject this.]
It is I. [Sounds bizarre or the sort of thing reserved only for
the pompous? Try saying “It’s [your name].” Problem solved.]
It’s me. [In this sentence, you’re identifying a hero or
culprit or the person in charge, not your identity as, say,
the person on the end of a phone call or at the door.]
 GRAMMAR MADE (RELATIVELY) PAINLESS 171

b. Use object pronouns to refer to a pronoun or noun that re-

ceives action – or to pronouns that receive action or which
flag relations in space and time in prepositional phrases.
(Confused? See p. 178 for a discussion of prepositional
phrases.) You can always tell a word receives action if you
can insert to or for before it without changing the meaning.
Object pronouns include: me, you, him, her, it, us, them,
whom, whomever:
Whom can you send to help us? [Whom receives the
action from the verb send.]
The website gave my sister and me some interesting
ideas.
c. Use possession pronouns to indicate ownership or, at least,
someone attached to something. Possession pronouns in-
clude its, your, their, whose, hers, his, mine, my, yours, ours,
whose, whosever.
Note that it’s, you’re, they’re, and who’s are all contractions,
that is, combinations of words where a missing letter is repre-
sented by an apostrophe.

Note: Possessive forms of pronouns are dicey only when they

modify entire phrases, when most writers tend to mistakenly rely
on the objective form:

Wrong!
We were miffed at him running out on us like that.
Correct version:
We were miffed at his running out on us like that.
[His modifies the entire phrase running out on us
like that, which acts as a single part of this
sentence.]
Also correct version:
We were miffed at him, running out on us like that.
[Him is the object of the preposition at.]
172 SUPPLEMENT

d. Use self-reference pronouns to intensify or to pinpoint an

earlier pronoun:

I myself stopped believing my mother could blow on traffic

lights to make them turn green after our car ran straight
through a red light. [Intensify.]

I can see myself turning up for the first day of school in the
seventies – sporting clothes Shirley Temple might have
worn during the Depression. [Pinpoint.]

Verbs
Verbs are shape-shifters, conveying information through changes
in their form.
Verbs can convey:
• Time – when action takes place: past, present, future.
• Person – who or what acts or experiences the action.
• Number – how many subjects act or receive action.
• Mood – the attitude expressed toward the action.
• Voice – whether the subject acts or is acted upon: active or
passive voice.
In addition, verbs can be either active or passive: Non-action verbs
express states of being: is, appears, has, exists, seems, represents,
has been. Action verbs express concrete action.

COMMON PROBLEMS WITH VERBS

1. Lie versus lay
Lie means to repose. Lay means to put down:
To lie: [Not in the sense of telling porkies.]
Present tense: I lie down all afternoon.
Past tense: I lay down all afternoon.
Present perfect: I have lain down all afternoon.
Future: tense: I will lie down this afternoon.
To lay:
Present tense: I lay my iPhone on the desk.
Past tense: I laid my iPhone on the desk.
Present perfect: I have laid my iPhone on the desk.
Future: tense: I will lay my iPhone on the desk.
 GRAMMAR MADE (RELATIVELY) PAINLESS 173

2. Moods
Ah, moods: verbs have three of them – unfortunately not as
straightforward as good, bad, and indifferent. We most commonly
use what we’ll call the default mood to express action, since Eng-
lish’s most common forms of verbs indicate statements and ques-
tions of fact.
We use the imperative mood to express commands and direct
requests. In the imperative mood, the subject is almost always
omitted and is always you. That is, the verb form always agrees
with you:
[You] Shut your mouth.
[You] Please, just stop talking.
Speakers of French, Italian, and Spanish rely on the subjunctive
mood more frequently than do speakers of English. The subjunc-
tive expresses conditions, speculations, recommendations, wish-
es, and indirect requests – in other words, something that, while
stated, hasn’t yet come to pass. When you use the subjunctive
mood, always use be for the future tense and were for the present,
with no fiddly guessing about whether to use singular or plural
forms of the verb:
Whether it be now or later, we must eventually face the truth.
If he were going to remain my manager, I might be inclined to stay.

3. Passive versus active construction

Passive construction is wordier and more difficult for readers to
follow than active construction. If you’ve blithely skipped Chap-
ter 3, you might want to skip this brief section, and, instead, read
the chapter’s description of the mostly negative effects of using
passive construction.
Verbs fall into passive construction when the action is per-
formed upon the subject, rather than by the subject:
A mistake has been made.

You should prefer the passive over the active whenever you’re
wriggling out of assigning blame:
McMurty was passed over for promotion for the nineteenth time.
174 SUPPLEMENT

You should also prefer passive over active only when emphasizing
someone’s status as victim or focusing more on the effects of an
action than on its actors:
My ideas were thoroughly shredded by everyone at that bloody
conference.

Or when you cannot identify who performed the action:

The fateful phone call was made at 6 am, but we’re still
investigating who made it.

Note: Cannot is one word, not two. Just one of those lovely
irregularities of English.

Adjectives and adverbs

Since both adjectives and adverbs are modifiers – or words that
describe other words – telling one from another can be tricky.
Adjectives, however, describe a noun or pronoun, while ad-
verbs describe a verb, adjective, or another adverb.
In most cases, you can’t use them interchangeably. Good is an
adjective that should modify a noun or pronoun. Well is an adverb
that should modify a verb or adjective. One reason why You done
good sounds awful is that, while you can do good work, you can
only do work well. In addition, think about the standard American
greeting which actually is anything but an enquiry after your cur-
rent state of mind or health: How’s it going? or How’re you doing?
If you’re bursting to be grammatically correct, the answer is [I’m]
good if your life is going swimmingly. However, if you’re describ-
ing the state of your health, the answer is [I’m] well.
Whatever you do, just don’t tell an American you had a root
canal or that your cat has feline leukemia. By the time you finish
your sentence, the American, who seemed so solicitous with that
greeting, has disappeared down the hall.

COMMON PROBLEMS WITH ADJECTIVES AND ADVERBS

1. Use the comparative form (the -er or more form) to explicitly

compare two things – which must be in the same sentence:
 GRAMMAR MADE (RELATIVELY) PAINLESS 175

Donald Trump is more successful than Don Ameche or Don

Ho. [adverb]
Your memory is better than mine. [adjective]

2. Use the superlative form to compare three or more things; the

items to which the noun or pronoun is being compared don’t
need to appear in the same sentence:
That meeting was the worst one in our company’s
history. [adjective]
He finished the exam quickest of all. [adverb]

3. Don’t be tempted into overkill and use both -er and more or
-est and most at the same time, or you’ll end up with the most
heaviest sentence of all.

4. While you can get away with using less and fewer interchange-
ably in many instances, if you have a bona fide grammarian
as a superior or you’re in the military, chances are someone
will point out to you that less refers to amounts that you can’t
count (as with volumes) while fewer refers to items that you
can count.
I’m drinking less milk these days than I did when I was
still growing up.

This milk has fewer calories than what I used to drink.

Common disasters: or, what the hell’s a split

infinitive, anyway?
BEFORE WE BEGIN . . . WHAT’S THE DIFFERENCE BETWEEN
A CLAUSE AND A PHRASE?
A clause always contains a subject and a verb. If the string of words
excludes a verb, you’re looking at a phrase, not a clause. In add-
ition, clauses either play major or minor roles in a sentence. If the
clause plays a major role, the clause can stand alone, severed from
the other clause, without seeming like a rump of a sentence. Major
clauses (dubbed independent clauses by grammarians) may have other
clauses alongside them in a sentence or may occupy the entire
176 SUPPLEMENT

sentence. In contrast, minor (also called dependent or subordinate

by grammar grouches) clauses accompany major clauses. In add-
ition, minor clauses prove useful in pointing out cause and effect,
indicating the timing of an event, or modifying information about
the main clause in the sentence. Furthermore, clauses may also
modify individual nouns or verbs. Clauses include noun clauses,
adjective clauses, and adverb clauses – all named for their function
within the sentence.
Phrases modify nouns, verbs, and, occasionally, adverbs;
they may also convey relationships in time or space. Phrases
include prepositional phrases, gerund, participle, and infinitive
phrases – all named for the type of grammatical object that be-
gins them. (For more on these ominous-sounding things, see
below.)
1. Dangling modifiers: we say that a modifier dangles when
its subject isn’t included in the sentence, or when its
subject isn’t the subject of the sentence. Most dangling
modifiers are phrases, not individual words. To correct a
dangling modifier, rewrite the modifier as a minor clause or
rewrite the main clause so that the subject is modified by
the phrase:

Wrong!
Do not sit in the chair without being fully assembled.

Correct versions:
Do not sit in the chair without its being fully assembled.
Do not sit in the chair unless it is fully assembled.

Wrong!
Strolling down Fifth Avenue, the Empire State Building
came slowly into view.

Correct versions:
While we were strolling down Fifth Avenue, the Empire
State Building came slowly into view.
Strolling down Fifth Avenue, I watched the Empire State
Building come slowly into view.
 GRAMMAR MADE (RELATIVELY) PAINLESS 177

2. Misplaced modifiers are words, clauses, or phrases that crop up

far from the thing their writer intended them to modify. The re-
sulting sentence not only fails to convey your intended mean-
ing; it usually also implies something unintentionally funny.
At this point, the old Strunk and White dictum to “put related
things together” actually comes in handy:

Wrong! [from the London Review of Books, no less]:

The rare exchange of letters between Vita Sackville-West
and Virginia Woolf was found, of all places, in the
morning room of Sissinghurst by a Dutch graduate student
curled up in a drawer.
Correct version:
A Dutch graduate student discovered the rare exchange of
letters between Vita Sackville-West and Virginia Woolf
curled up in a drawer in, of all places, the morning room
of Sissinghurst.
Wrong!
Two cars were reported stolen by the Gainesville police
yesterday. [This version is accurate if the police run
a stolen-car ring, which would merit a screaming
headline, not a “just the facts, ma’am,” report in the
local paper.]
Correct version: Yesterday, the Gainesville police
reported two cars stolen. [Tip: If you try to rely on
people or tangible objects as your subjects, you’ll find
it nearly impossible to write dangling or misplaced
modifiers.]

3. Split infinitives result when you place an adverb or phrase be-

tween the to portion of the infinitive and the verb. According
to the trusty Oxford English Dictionary, split infinitives are now
officially kosher, provided you don’t work for (a) anyone born
prior to 1957 or (b) anyone British or (c) professors of English
who haven’t checked the changing rules in the past fifteen
years. As Irish playwright George Bernard Shaw once put it in
a letter to the London Times:
178 SUPPLEMENT

There is a busybody on your staff who devotes a lot of time to

chasing split infinitives … I call for the immediate dismissal of
this pedant. It is of no consequence whether he decides to go
quickly or to quickly go or quickly to go. The important thing
is that he should go at once.

4. The same ruling applies to prepositions at the ends of sen-

tences. The old rule states: “A preposition is something you
should never end a sentence with.” My advice: if the writer
stating the rule can’t follow the rule, neither should you. The
rule mainly evolved to prevent the awkward rhythm created
by a preposition dangling at the very end of a sentence, and
prepositions certainly can waste the all-important stress po-
sition of sentences, but, ultimately, the correctness judgment
call is yours to make.

A phrase is a phrase: or, what the hell’s a preposition/

participle/gerund/appositive/infinitive?
Prepositions indicate relationships in space or time. Prepositions
link nouns or pronouns to other words in a sentence and include
words like: to, from, in, out, on top of, behind, before, during,
under, beneath, after, since. Prepositional phrases consist of prepo-
sitions and nouns or pronouns. Common prepositional phrases
modify or provide additional information about nouns, pronouns,
verbs, or adverbs:
On Tuesday, the market plummeted to its lowest point in eight
years.
The incentive package included four days vacation at my
choice of Caribbean resort.

WHY YOU SHOULD CARE ABOUT PREPOSITIONAL PHRASES:

1. By identifying prepositions and prepositional phrases correctly,

you minimize the possibility of confusing your subject with
the pronoun or noun in the prepositional phrase – which can
result in your using the wrong form of a verb.
 GRAMMAR MADE (RELATIVELY) PAINLESS 179

2. You may use a comma after a prepositional phrase that

begins a sentence. If the phrase is seven words or longer,
however, you should follow it with a comma to avoid your
reader sifting through the string for likely subject–verb
combinations.
3. You should NOT use a comma before a prepositional phrase
that ends a sentence.
4. You rarely need to use commas to set off prepositions or prepos-
itional phrases from the rest of the sentence.
Participles are forms of a verb that end in either -ed or -ing and act
like adjectives. Participles can stand alone in a sentence, function-
ing strictly as an adjective, or they can begin participle phrases,
which simply modify a noun or pronoun.
Munching slowly, the employees worked their way through the
retirement buffet.

WHY YOU SHOULD CARE ABOUT PARTICIPLE PHRASES:

1. Participle phrases cannot be preceded by possessive pro-

nouns.

2. Participle phrases should generally be set off with commas,

wherever they appear, to avoid your reader stumbling down
the garden path by seizing on the participle as your sentence’s
or clause’s verb. One comma should separate the participle
phrase from the main clause if the phrase begins or ends the
sentence. Use two commas to bracket the phrase if it occurs
anywhere else in the sentence.
Gerunds are not noxious forms of tropical diseases, despite the
image this word conjures up. Instead, gerunds are forms of a
verb that end in -ing and act like nouns. Gerunds are true gram-
matical shape-shifters. A gerund can act alone as the subject of
the sentence, begin a phrase that acts as the grammatical subject,
act as an indirect or direct object, or even serve as the object
of a preposition. Gerunds can also begin gerund phrases, which
also act as either subjects, indirect or direct objects, or objects of
prepositions.
180 SUPPLEMENT

WHY YOU SHOULD BOTHER KNOWING ANYTHING ABOUT

GERUND PHRASES:

1. Gerund phrases must be preceded by possessive pronouns:

I was annoyed by the employees’ munching slowly as they
worked their way through the retirement buffet.

2. Because gerund phrases are integral parts of the sentences they

occur in, you should never set them off with commas:
Partying hearty requires great endurance. [Gerund phrase
acting as a subject.]
We enjoyed partying hearty. [Gerund phrase acting as a direct
object.]
We objected to everyone’s partying so heartily. [Gerund phrase
acting as the object of a preposition.]

Appositives are nouns, pronouns, or a short phrase that renames

another noun or pronoun; they most commonly rename the subject
of the sentence. Appositive phrases modify nouns or pronouns and
should appear as close as possible to the items they modify. They
are also great stylistic devices because they can help you to create
more graceful sentences, eliminate repetition, create a rhythm in
your writing, and make your sentences more interesting.

WHY YOU SHOULD GIVE A DAMN ABOUT APPOSITIVES (ASIDE FROM

THE NUMEROUS ADVANTAGES LISTED ABOVE):

1. Appositives and appositive phrases are always set off by com-

mas, usually bracketed off from the rest of the sentence by two
commas:
Bob’s car, an utter wreck, died a grisly death by the side of
I-75.
Do you know my friend, Bill?
David Prowse, the guy in the Darth Vader suit in the Star
Wars movies, did not find out that his lines were going to
be dubbed over by James Earl Jones until he attended the
premiere screening of the film.
 GRAMMAR MADE (RELATIVELY) PAINLESS 181

Phineas T. Barnum, the great American showman, was near

death in 1891 when the New York World asked if he’d like to
have his obituary published while he could still read it.

Infinitives are the equivalent to baseball’s utility infielders: they

can act as nouns, adjectives, or adverbs, either by themselves or
as part of a longer phrase. Infinitives are formed by the word to
and the form of the verb that accompanies to: to be, to succeed, to
yield, to promote.

WHY YOU SHOULD WORRY ABOUT INFINITIVES:

You should punctuate infinitives, either by themselves or as part
of a phrase, according to their function within the sentence.
1. If they function as a noun, they’re usually essential to the sen-
tence and should NOT be set off from it by commas.
To succeed takes courage, foresight, and a hefty dose of luck.
[Infinitive acting as noun and subject of the sentence.]
Alone in her cubicle, she wanted only to survive. [Infinitive
acting as noun and direct object of sentence.]

2. If the infinitive or infinitive phrase functions as an adverb but

not as an essential part of a sentence, you usually set it off
from the sentence with a comma or pair of commas.
Afraid to move, he froze in terror. [Infinitive modifying
afraid, acting as adverb.]

3. Most commonly, infinitives and infinitive phrases acting as

adjectives are also acting as essentials within the sentence,
usually as appositives (see above definition) or as predicate
adjectives, adjectives that rename the subject of the sentence,
always following a non-action verb.
The firm’s hope was to grow the business enough to
warrant an IPO within two years. [Infinitive phrase acting
as a predicate adjective.]
His goal, to break into Fort Knox, was, of course, never
achieved. [Infinitive phrase acting as an appositive.]
182 SUPPLEMENT

A rule that will guide you in punctuating both infinitive phrases

and just about anything else in a sentence: Never place a single
comma between the subject and verb. Always use either two –
to bracket off a phrase or word – or none.

Adverb, adjective, and noun clauses

Clauses come in three forms – adverb, adjective, and noun – and
differ from phrases in that clauses always have a subject and verb,
while phrases lack both, having only an object. Major clauses can
stand on their own; minor clauses, not surprisingly, cannot (see
first page of this Supplement, p. 164, above).
Adverb clauses tend to convey relationships in time and
space, results or causation, choices, conditions, contrast, and lo-
cations, although they can also sometimes modify or hedge the
statement in the main clause. Adverb clauses begin with words
like: when, since, whenever, because, although, until, as soon as,
while, once, as, for, unless, provided that, even if, rather than, in
order that.

WHY YOU SHOULD PAY ATTENTION TO ADVERB CLAUSES:

1. They enable you to put sentences together smoothly, giving the

illusion of cause and effect, even when none exists.
2. They enable you to demonstrate concisely complex relation-
ships between items or situations by relying on the sentence
structure to establish part of the relationships.
3. They allow you to vary the rhythm of your sentences radically
and easily, as an adverb clause beginning a sentence provides
the most stark contrast possible with the usual subject–verb–
object of standard written English sentences.
4. Adverb clauses are governed by especially quirky rules con-
cerning punctuation:
a. Always set an adverb clause off from the main clause when
it begins a sentence:
Whenever Microsoft introduces a product with a zero at
the end of the version number, I avoid it completely.
 GRAMMAR MADE (RELATIVELY) PAINLESS 183

b. While you don’t always need to punctuate adverb clauses

in the middles of sentences, placing commas around the
clause can help distinguish the clause from your all-impor-
tant subject and verb. And, since Americans tend to prefer
more punctuation over less punctuation – enabling them
to better control the way readers interpret their writing –
when in doubt, plug in a comma. Just remember to use
either two or none, never just one:
We looked at the product, and, since it was water-
damaged, marked it down steeply.
c. You should insert commas in adverb clauses at the ends of
sentences only when they begin with the words as, for, or
since when they are used to mean because. You should NOT
punctuate adverb clauses that begin with because or with
for or since used merely to indicate time or location. This
rule probably originated in the need to distinguish caus-
ation from time/spatial relationships:
We marked the product down, since it was water-
damaged. [Since here is synonymous with because.]
We deep-discounted the product because it was water-
damaged. [When because begins an adverb clause,
you never use a comma.]
I enjoyed Hearts of Darkness, A Filmmaker’s
Apocalypse more than I enjoyed Apocalypse Now. [All
other kinds of adverb clauses are also not set off with
a comma when they end the sentence.]

Rob had watched the Francis Ford Coppola film

more than 200 times since its first release. [Since
here indicates a relationship in time, not a causal
relationship.]

Adjective clauses exclusively modify a noun or pronoun within a

sentence, sometimes merely providing supplementary information
about it, occasionally identifying the noun absolutely.
184 SUPPLEMENT

WHY YOU SHOULD PAY ATTENTION TO ADJECTIVE CLAUSES:

1. While the distinction is no longer observed in the Oxford Eng-

lish Dictionary, some grammarians still insist that should be
used only for defining adjective clauses, while which should
be used solely for non-defining adjective clauses.
a. In a defining clause, the clause identifies the object being
modified absolutely, nailing its identity down to a single
thing:
The book that I had laid on the table has been stolen.
b. In a non-defining clause, the clause only provides more
information about the object being modified:
The book, which has grease spots on its dust cover, is one I
consult frequently.

2. The only rule that still does legitimately refer to the difference
between the two is: never place commas around a defin-
ing clause; always place commas around a non-defining
clause.
Noun clauses are minor clauses that act like nouns. We seldom
set noun clauses off with a comma or set of commas, since
they can function as essential parts of sentences – subjects, dir-
ect and indirect objects, objects of prepositions, or predicate
nouns (which restate the subject and always follow an non-
action verb).
Noun clause as subject:
How the accountant managed to dredge up those profit and
loss figures is a mystery.
Noun clause as direct object:
No one understands why a reputation is something people
usually bother about only when it’s in shreds.
Noun clause as appositive:
Adolescent males, the members of the population who most
want to control their lives and, arguably, feel least in
control, represent 80 percent of the market for US video
games.
 PUNCTUATION MADE PAINLESS 185

WHY YOU SHOULD CARE ABOUT NOUN CLAUSES FOR

PUNCTUATION’S SAKE:
Noun clauses can also function as appositives, requiring
two commas to bracket them off from the subjects and
verbs of sentences (see above).

Where do you put the ___? Punctuation made painless

In addition to the punctuation principles we’ve covered above,

you should be aware of some weird or particular usages for the
common elements of punctuation: the period, question mark,
comma, semicolon, colon, and dashes.

Periods, of course, are the workhorses of the punctuation stable.

1. Always use them after a complete sentence, but also after

2. a command or a request disguised as a command:
Can you please return this report to me as soon as possible.

3. Use periods after most abbreviations: Dr. Ms. Jr.

4. But not after individual letters in an acronym: NATO.
5. Use a period after an initial: John F. Kennedy.
6. Always place a period inside a quotation mark that ends a
sentence:

The sign read: “A pest is a friend in need.”

Question marks tend to be underused, as writers forget that ques-

tions require question marks, just as a rising inflection in a spoken
sentence indicates a query or uncertainty.
1 Always use a question mark to end a question.
2. Place a question mark inside of closing quotation marks if the
question mark belongs in the original quotation:
In a dream I heard someone asking, “Isn’t atheism a
non-prophet organization?”
3. Place the question mark outside the closing quotation marks if
the question mark is not part of the quotation:
186 SUPPLEMENT

Didn’t your mother say, “I knew I shouldn’t have sold my

Coca-Cola shares in 1958”?

Commas control how your readers interpret your writing, indicat-

ing when they should pause and which clusters of words should be
read separately from the major clause. When in doubt, you’re usu-
ally better to over-punctuate than to under-punctuate as long as you
avoid separating the subject and verb with a single comma – the
big comma error – or use a comma to separate two major clauses.
1. Use a comma to show contrast between phrases:
The neighbors returned home at all hours, often
falling-down drunk.

2. Use a comma to stand in for a word that has been omitted:

To err is human, to forgive, divine. [Since your reader can
easily infer the second is, you can leave it out but must
indicate its absence with the comma.]

3. Use two commas to bracket off appositives and appositive

phrases (see above).

4. Use a comma to set off direct address, or words that express to

whom a remark is addressed:
Gary, are you with us on this one?

5. Use a comma to set off parenthetical expressions, or phrases

and expressions that interrupt a sentence:
The issue, as you well know, is a potentially explosive one.

6. Use a comma to separate items in a series:

We purchased an entirely new suite of office equipment,
including a digital video camera, two digital still cameras,
and a digital editing suite.

7. Use a comma to set off a direct quotation:

“Lawyers are the larval forms of politicians,” he said.
[Quote occupies the beginning of a sentence.]
 PUNCTUATION MADE PAINLESS 187

He said, “Lawyers are the larval forms of politicians.”

[Quote occupies the end of a sentence.]
“Lawyers,” he said, “are the larval forms of politicians.”
[Split one-sentence quote, with the attribution (he said) in
the middle.]
“Lawyers are the larval forms of politicians,” he said. “Why
else do you think I suffered through three years at NYU’s
College of Law?” [Split two-sentence quote.]

8. Use a comma to separate two major clauses joined by and, or,

but, for, so, and yet:
Our management strategies failed to work, and we ran out
of money.

9. Use a comma after or commas around coordinate conjunctions

like however, consequently, conversely, nevertheless, moreover,
furthermore, therefore:
The wide-body plane lumbered down the runway, seeming
far too heavy to ever take flight, however, its thrust lifted
the heavy aircraft rapidly away from the ground.

10. DO NOT use a comma to separate two clauses when they share
the same subject. If you use a comma between two verbs refer-
ring to the same grammatical subject, you sever one verb from
the subject it refers to:
We stopped thinking about an IPO and started worrying
about staying solvent.

Semicolons are halfway between a comma and period in terms of

their emphasis. While semicolons don’t cement the ends of sen-
tences the way periods do, they indicate totally major clauses, not
subordination. They also, however, have a variety of uses.
1. Use a semicolon between closely related major clauses:
Bigamy is one spouse too many; monogamy is the same idea.

2. Semicolons can also flag contrasting relationships between

major clauses:
188 SUPPLEMENT

The man appeared to be begging us for help; he was

threatening us with outspread hands.

3. Use a semicolon to separate items in a series when they con-

tain internal punctuation:
When Cunard decided to take its fleet upmarket, the
company’s ships consisted of several transatlantic class
mid-sized ships, which had mostly seen their better days; a
few mid-sized, flashy ships that were flat-bottomed and
therefore restricted to the Caribbean; and the venerable
Queen Elizabeth 2, the most powerful name brand at sea.

4. Place semicolons outside quotation marks:

The team acknowledged they were “psyched about the
profit sharing agreement”; they still, however, held out for
more vacation time.

Colons receive more abuse than any other form of punctuation,

usually by academics who seem to hate ending a sentence with
a period and insist on seeing everything as endlessly connected.

1. Use a colon before a list:

In the Thoroughbred industry, you can make money in several
ways: through breeding and selling as weanlings or yearlings;
through breeding and selling as two-year-olds; or through
buying weanlings or yearlings and selling as two-year-olds.

2. Use a colon before a long quotation:

Abraham Lincoln said: “Fourscore and seven years ago, our
fathers brought forth upon this continent a new nation,
conceived in liberty and dedicated to the proposition that
all men are created equal.”

3. Use a colon before a part of a sentence that restates the subject

of the sentence:
I know only one certain way to make a small fortune from
horse racing: start with an extremely large one.
 PUNCTUATION MADE PAINLESS 189

4. Use a colon after the salutation of a business letter:

Dear Chancellor:
Dashes enable you to break some of the monotony that can ac-
company one too many sentences bristling with loads of commas.
Dashes can replace any form of punctuation, save for the period.
If you’re in doubt about what piece of punctuation to use, use the
trusty dash. Just remember that dashes are obtrusive bits of punc-
tuation, excellent for emphasis, but sufficiently “loud” that you
should restrict yourself to using them only once per paragraph.
1 Dashes are distinct from hyphens, which link words together
or which link nouns together to indicate they’re being used as
an adjective: two-year-olds, age-related illness.
2. Dashes represent changes of thought, show emphasis, and
stand in for both commas and semicolons:
An archeologist – of course, I’m not referring to
Dr. Montebello – is someone whose career lies in ruins.
Ahead lay our choice – to do or die.
We made up our minds – tomorrow we were going to begin
rowing between Australia and New Zealand.
ENDNOTES

1 So much advice, so much lousy writing

1 Richard Marius, A Writer’s Companion. 4th edn (New York:
McGraw-Hill, 1999).
2 William Strunk, Jr. and E. B. White, The Elements of Style (Boston:
Allyn & Bacon, 1979).
3 Sheridan Baker, The Practical Stylist (New York: Longman, 1997).
4 Knoblauch and Brannon (1984): 163.
5 The sentence in the California Penal Code, Chapter 1.5, Section
631a, reads: Any person who, by means of any machine, instru-
ment, or contrivance, or in any other manner, intentionally taps, or
makes any unauthorized connection, whether physically, electri-
cally, acoustically, inductively, or otherwise, with any telegraph or
telephone wire, line, cable, or instrument, including the wire, line,
cable, or instrument of any internal telephonic communication sys-
tem, or who willfully and without the consent of all parties to the
communication, or in any unauthorized manner, reads, or attempts
to read, or to learn the contents or meaning of any message, report,
or communication while the same is in transit or passing over any
wire, line, or cable, or is being sent from, or received at any place
within this state; or who uses, or attempts to use, in any manner,
or for any purpose, or to communicate in any way, any information
so obtained, or who aids, agrees with, employs, or conspires with
any person or persons to unlawfully do, or permit, or cause to be
done any of the acts or things mentioned above in this section, is
punishable by a fine not exceeding two thousand five hundred dol-
lars ($2,500), or by imprisonment in the county jail not exceeding
one year, or by imprisonment in the state prison, or by both a fine
and imprisonment in the county jail or in the state prison. If the
person has previously been convicted of a violation of this section
or Section 632, 632.5, 632.6, 632.7, or 636, he or she is punish-
able by a fine not exceeding ten thousand dollars ($10,000), or by
 ENDNOTES 191

imprisonment in the county jail not exceeding one year, or by im-

prisonment in the state prison, or by both a fine and imprisonment
in the county jail or in the state prison.
6 Scott Adams, qtd. in Lepore (2014): 73.

2 The new science of writing

1 Moore (1965): 114–117.
2 These daunting equations form the basis of the three bibles of
readability formulas:
1. Flesch (1974).
2. Flesch (1979).
3. Gunning and Kallan (1994).
3 David Denby, “A Famous Man: The Collected Works of James
Agee,” The New Yorker, January 9, 2006: 85.
4 Lynne Truss, Eats, Shoots & Leaves: The Zero Tolerance Approach
to Punctuation (London: Profile Books, 2003): 71.
5 Dehaene (2009): 15–16.
6 Perfetti (1999).
7 Zwaan et al. (1995); Zwaan (1996); Zwaan and Radvansky (1998).
8 Fleischman (1990).
9 Bruner (1986): 16–25.
10 Leslie and Keeble (1987).
11 Rumelhart (1975); Anderson et al. (1978).
12 The Egyptians relied on a schema to represent important people
as large (and nonentities as small). In contrast, during the 1817
exhibition at the Royal Academy of Arts, an entirely different schema
prompted one critic to describe painter John Constable’s now-classic
painting Wivenhoe Park, Essex, as “that nasty green thing.” See
Gombrich (1961).
13 Oliver Sacks, An Anthropologist on Mars: Seven Paradoxical Tales
(New York: Vintage, 1996).

3 Choosing words and structuring sentences:

The first C: Clarity
1 Heider and Simmel (1944).
2 Michotte (1963).
192 ENDNOTES

3 Leslie and Keeble (1987).

4 Michael et al. (2001).
5 Anderson et al. (1978).
6 Fleischman (1990); Norman et al. (1991); Noordman et al. (1992);
Zwaan (1996); Ferreira et al. (2002); Ferreira (2003).
7 F. Ferreira and J. Stacey, unpublished manuscript, 2000.
8 Evans (2014): 75.
9 McWhorter (2001): 158–159.
10 Rhodes (1997).
11 Olson and Filby (1972).
12 Stelzner (1966).
13 Wydick (1998); Cordes (2002).
14 Anthony (2007).
15 Olson (1991).
16 Eisenstein (1979).
17 Ferreira and Clifton (1986); Ferreira (2003).
18 Greene and McKoon (1995); Bornkessel et al. (2005).
19 Clifton et al. (1984).
20 Clifton et al. (1984).
21 Bornkessel et al. (2005).
22 Klee and Legge (1976).
23 K. Kearns, unpublished manuscript, 1988; Bornkessel et al. (2002).
24 Berk and Whalen (1992); Billig (2008).
25 Van Dijk (2008).
26 Stallings et al. (1998).
27 Clark and Sengul (1979).
28 Frazier et al. (1984).
29 Researchers use the term NP shift or Heavy NP shift to indi-
cate sentences that begin with noun phrases and are also left-
branching, embedding complexity to the left of the main verb
in the major clause. Noun Phrase (NP) shift and left-branching
sentences make readers’ brains work overtime: Hagoort et al.
(1993); Just et al. (1996); Gibson (1998); Culicover and Lev-
ine (2001); Lieberman (2001); Clifton et al. (2007); Temperly
(2007).
30 Schank (1982).
31 Ferreira and Clifton (1986); Michael et al. (2001).
32 Pluchino et al. (2003): 688.
33 Steinman (2003): 671.
 ENDNOTES 193

4 Putting sentences together: The second C: Continuity

1 Britt (1994); Miyake et al. (1994); Ferreira et al. (2002); Baddeley
(2004).
2 The term cognitive overload crops up as a research topic with the
emergence of hyper- and multimedia. New technologies gave re-
searchers multiple lenses on these new media – as a curse, as an
everyday challenge, and as fodder for understanding how brains
cope with multiple demands. Unsurprisingly, studies through the
1990s focused on the downside of increased competition for our
attention, while studies after 2000 simply seized opportunities for
understanding our brains’ peak attentional capacity: Landauer
(1995); Mackay (2000); Fox et al. (2007).
3 Schank and Abelson (1977); Spiro (1980).
4 Smith (2012).
5 Anderson et al. (1978).
6 Bain (1890): 11; Atkinson and Shiffrin (1968); Glanzer (1972);
Baddeley and Hitch (1993); Thapar and Greene (1993); Davelaar
et al. (2005).
7 Murdock (1962); Just and Carpenter (1980); Williams et al. (1981);
Graesser et al. (1997).
8 This approximation of working memory avoids the impact of pre-
dictability, syntactic, or inferential difficulty, as well as the abil-
ities of the subjects doing the recalling: Miller (1956); Schnotz and
Kürschner (2007).
9 Daneman (1982).
10 Graesser et al. (2001).
11 Daneman and Carpenter (1983).
12 Garnham et al. (1982).
13 Just and Carpenter (1980); Trabasso and van den Broek (1985);
King and Just (1991).
14 Black and Bower (1979); Zwaan (1996).
15 Coulson et al. (1998); Mason et al. (2003); Davis (2005); Temperly
(2007); Smith (2012).
16 Chafe (1974).
17 Mayer (1976).
18 Sperber and Wilson (1985, 2002, 2004).
19 Just and Carpenter (1980); Daneman and Carpenter (1983);
Fahnestock (1983); Sloan (1988).
194 ENDNOTES

20 Garnham et al. (1982).

21 Dee-Lucas et al. (1982).
22 Daneman (1982).
23 Lupker (1984); Nicholas (1998); Chang et al. (2000).
24 Graf and Schacter (1985).
25 Kintsch and van Dijk (1978); Kintsch and Vipond (1979).
26 Tversky and Kahneman (1973).
27 Rayner et al. (2001).
28 Frazier and Rayner (1982); Rayner et al. (1983); Ferreira and
Clifton (1986).

5 Organizing paragraphs and documents: The third C:

Coherence
1 DeKay (2010).
2 Nielsen (2006).
3 Samuels and Kamil (1984).
4 Brown (1982).
5 Spiro (1980).
6 Yore and Shymansky (1985).
7 Chang et al. (2000).
8 Graesser et al. (1997).
9 Van Dijk (1988).
10 Dor (2003).
11 Duggan and Payne (2011). See earlier research on inferential pro-
cessing and comprehension: Kintsch and van Dijk (1978); Bower
et al. (1979); Miller and Kintsch (1980); Masson (1982).
12 Williams et al. (1981).
13 McCarthy et al. (2008).
14 Britton (1994); Kintsch (2002).
15 Cutrell and Guan (2007).
16 Cirilo (1981).
17 McCarthy et al. (2008).
18 Miller and Kintsch (1980).
19 Wikborg (1985).
20 Hakala and O’Brien (1995).
21 Miller and Kintsch (1980).
22 Williams (1995).
23 Anderson et al. (1978); Kintsch (2002).
 ENDNOTES 195

24 Fredric Jameson in the London Review of Books 23 (19) October 4,

2001.
25 Pickering and Branigan (1998); Pickering and Garrod (2013).
26 Baddeley (2004): 81. For the particularly strong impact of recency on
recall, see Glenberg and Swanson (1986); Baddeley and Hitch (1993).
27 McCarthy et al. (2008).
28 Kintsch and van Dijk (1978).
29 Kintsch and van Dijk (1978).
30 Frase (1969).
31 Kintsch and van Dijk (1978).
32 Graesser et al. (1997).
33 Zwaan (1994).
34 Kintsch (1988).
35 Graeme Wood, “The Future of College?” The Atlantic, September
2014: 50–60.
36 Spyridakis and Fukuoka (2002).
37 Lorch and Lorch (1985); Kintsch (1988); Murray and McGlone
(1997); Therriault and Raney (2002); Ritchey (2011).
38 Higgins et al. (2004).
39 Diederich et al. (1961).
40 Spyridakis and Fukuoka (2002); Therriault and Raney (2002);
Ritchey (2011).
41 Murdock (1962).
42 Spiro (1980); Brown (1982); Yore and Shymansky (1985).
43 Van den Broek et al. (2001).
44 Luchins (1958).
45 Kieras (1978).
46 Moreno and Mayer (2000).
47 Paivio (1990).
48 Van den Broek et al. (2001).
49 Pickering and Ferreira (2008).

6 Maximizing efficiency: The fourth C: Concision

1 Transcript of a recording of a meeting among the President,
H. R. Haldeman, and John Dean, on September 15, 1972, from
5:27 to 6:27 pm. Mary Ferrell Foundation, Watergate Recording
Transcripts. Available as of 2014 from www.maryferrell.org/
mffweb/archive/docset/getList.do?docSetId=1923.
196 ENDNOTES

2 Ron Grossman, “In May 1974, Tribune Delivered 2 Watergate

Bombshells,” Chicago Tribune, April 27, 2014.
3 Grice (1975); Limaye and Cherry (1987).
4 Britton (1982); Nell (1988); Bazerman (2003).
5 Plato (1973).
6 Grice (1975).
7 Ong (1982).
8 Limaye and Cherry (1987); Hyland and Tse (2004); Lebovits
(2006); Matsuda and Tardy (2007); Dermer et al. (2010).
9 Crystal (2004).
10 McCrum et al. (1992); Crystal (2004).
11 Clanchy (1979).
12 McCrum et al. (1992).
13 Crystal (2004).
14 Eisenstein (1979).
15 Bryson (1990).
16 Locker (1987).
17 Wydick (1998).
18 Costello and Keane (2000).
19 Crystal (2004).
20 Baddeley (2004).
21 Williams (1995).
22 Hyland and Tse (2004).
23 Handlin et al. (2003).
24 Hyland and Tse (2004). Metadiscourse is alive and everywhere in
academic medicine: Salager-Meyer (1994).
25 Harwood (2005).
26 Matsuda and Tardy (2007).
27 See the Oxford English Dictionary’s entry for intent, under Phrases
(2): www.oed.com. The abbreviation first crops up in Joseph Ad-
dison’s essay for Tatler (Number 96): “Whoever resides in the
World without having any Business in it . . . is to me a Dead Man
to all Intents and Purposes.” Addison, an elegant prose stylist, at
least takes us from the triplet of meaningless, cover-your-ass-ness
to a doublet of meaningless, cover-your-ass-ness.
28 Fowler (1965).
29 The Oxford New French Dictionary (New York: Berkley Books, 2003).
30 Herbert (1967).
31 Baddeley et al. (1975).
 ENDNOTES 197

7 Making music with words: The fifth C: Cadence

1 Erasmus (1978).
2 Fowler (1965): 526.
3 Qtd. in Manguel (1996): 42.
4 Fischer (2005).
5 Cobban (1975); Pedersen (1997).
6 Schwinges (1996).
7 Dehaene (2009).
8 Paulesu et al. (1993).
9 For a highly readable layperson’s guide to the brain and neuroanat-
omy, look no further than Rita Carter, Mapping the Mind (1998).
For one of the most informative guides you’d never imagined ex-
isted on what the most minute and insignificant aspects of our
language says about us – and an equally engaging transcription
of Broca’s and Wernicke’s aphasiacs in conversation, see James
W. Pennebaker, The Secret Life of Pronouns: What Our Words Say
about Us (2011).
10 Horwitz et al. (1998); Bernal and Ardila (2009).
11 Horwitz et al. (1998).
12 Büchel et al. (1998).
13 Nishimura et al. (1999).
14 Ernest Hemingway, “Three Shots,” The Nick Adams Stories (New
York: Scribner, 1972): 13.
15 Henry James, “The Aspern Papers,” The Great Short Novels of Henry
James. Ed. Philip Rahv (New York: Skyhorse Publishing, 2014): 453.
16 “Ernest Hemingway at 100: Star Style and Rules for Writing,” The
Kansas City Star, June 26, 1999. Available as of 2014 at www.
kcstar.com/hemingway/ehstarstyle.shtml.
17 Sean Michaels, “Rihanna and Michael Jackson Sued by African
Singer,” Guardian, February 4, 2009.
18 Heilman et al. (2004).
19 Ramachandran (2011).
20 Speedie et al. (1993).
21 Patel (2010).
22 Carpenter (1999).
23 Gordon et al. (2002).
24 Research on the role of reading in fostering better writing is sur-
prisingly scarce. However, some studies have made convincing
198 ENDNOTES

cases for the unquestionable impact of reading on writing in el-

ementary education. In this area, the most worthwhile findings
stem from a largely one-man crusade by researcher Timothy Shan-
ahan: Shanahan (1984); Shanahan and Lomax (1986); Crowhurst
(1990); Fitzgerald and Shanahan (2000); Shanahan (2006).
25 Pennebaker (2011).
26 Coh-Metrix, an automated tool for assessing the quality of writ-
ing, contains some admirable features correlated with the prin-
ciples tied to readability outcomes we’ve explored throughout The
Reader’s Brain. Most significantly for cadence, Coh-Metrix tied
perceptions of the quality of the writing directly to its syntactic
complexity. This outcome is unsurprising, given that some of the
brains behind Coh-Metrix include Arthur Graesser, Philip Mc-
Carthy, and Danielle McNamara, all of whom published insightful
research on many aspects of cognition, reading, and readability.
See McNamara et al. (2010).
27 Pennebaker (2011).
SELECT BIBLIOGRAPHY

Anderson, Richard C., Rand J. Spiro, and Mark C. Anderson. (1978).

“Schemata as Scaffolding for the Representation of Information
in Connected Discourse,” American Educational Research Journal
15/3: 433–440.
Anthony, Ted. (2007). Chasing the Rising Sun: The Journey of an
American Song. New York: Simon & Schuster.
Atkinson, R. C. and R. M. Shiffrin. (1968). “Human Memory: A Pro-
posed System and Its Control Processes,” in K. W. Spence and
J. T. Spence (eds.), The Psychology of Learning and Motivation:
Advances in Research and Theory. New York: Academic Press:
89–195.
Baddeley, Alan. (2004). Your Memory: A User's Guide. Buffalo, NY:
Firefly Books.
Baddeley, Alan D. and Graham Hitch. (1993). “The Recency Effect:
Implicit Learning with Explicit Retrieval?” Memory & Cognition
21/2: 146–155.
Baddeley, Alan D., Neil Thomson, and Mary Buchanan. (1975).
“Word Length and the Structure of Short-Term Memory,” Journal
of Verbal Learning and Verbal Behavior 14/6: 575–589.
Bain, Alexander. (1890). English Composition and Rhetoric. Enlarged
edn. London: Longmans, Green, and Company.
Bazerman, Charles. (2003). “Speech Acts, Genres, and Activity Sys-
tems: How Texts Organize Activity and People,” in Charles Bazer-
man and Paul Prior (eds.), What Writing Does and How It Does
It: An Introduction to Analyzing Texts and Textual Practices. New
York: Routledge: 309–339.
Berk, Robert N. and Elizabeth Whalen. (1992). “Impediments to Clar-
ity: An Annotated Glossary of Rhetorical Pratfalls and Pitfalls,”
AJR: American Journal of Roentgenology 159/5: 1115–1121.
Bernal, Byron and Alfredo Ardila. (2009). “The Role of the Arcuate
Fasciculus in Conduction Aphasia,” Brain 132/9: 2309–2316.
Billig, Michael. (2008). “The Language of Critical Discourse Analysis:
The Case of Nominalization," Discourse & Society 19/6: 783–800.
200 SELECT BIBLIOGRAPHY

Black, John B. and Gordon H. Bower. (1979). “Episodes as Chunks in

Narrative Memory,” Journal of Verbal Learning and Verbal Behav-
ior 18: 309–318.
Bornkessel, Ina, Matthias Schlesewsky, and Angela D. Friederici.
(2002). “Beyond Syntax: Language-related Positivities Reflect the
Revision of Hierarchies,” NeuroReport 13/3: 361–364.
Bornkessel, Ina, Stefan Zysset, Angela D. Friederici, D. Yves von
Cramon, and Matthias Schlesewsky. (2005). “Who Did What to
Whom? The Neural Basis of Argument Hierarchies during Lan-
guage Comprehension,” NeuroImage 26/1: 221–233.
Bower, Gordon H., John B. Black, and Terrence J. Turner. (1979).
“Scripts in Memory for Text,” Cognitive Psychology 11/2: 177–220.
Britt, M. Anne. (1994). “The Interaction of Referential Ambiguity and
Argument Structure in the Parsing of Prepositional Phrases,” Jour-
nal of Memory and Language 33/2: 251–283.
Britton, Bruce K. (1994). “Understanding Expository Text: Building
Mental Structures to Induce Insights,” in Morton Ann Gernsbacher
(ed.), Handbook of Psycholinguistics. San Diego, CA: Academic
Press: 641–674.
Britton, James. (1982). “A Reader's Expectations,” in Gordon M. Pradl
(ed.), Prospect and Retrospect: Selected Essays of James Britton.
Montclair, NJ: Boynton/Cook: 130–138.
Brown, Ann L. (1982). “Learning How to Learn from Reading,” in
Judith A. Langer and M. Trika Smith-Burke (eds.), Reader Meets
Author/Bridging the Gap: A Psycholinguistic and Sociolinguistic Per-
spective. Newark, DE: International Reading Association: 26–54.
Bruner, Jerome. (1986).Actual Minds, Possible Worlds. Cambridge,
MA: Harvard University Press.
Bryson, Bill. (1990). The Mother Tongue: English and How It Got That
Way. New York: William Morrow.
Büchel, Christian, Cathy Price, R. S. Frackowiak, and Karl Friston.
(1998). “Different Activation Patterns in the Visual Cortex of Late
and Congenitally Blind Subjects,” Brain 121/3: 409–419.
Carpenter, Ronald H. (1999). Choosing Powerful Words: Eloquence
That Works. Needham Heights, MA: Allyn & Bacon.
Carter, Rita. (1998). Mapping the Mind. Berkeley, CA: University of
California Press.
Chafe, Wallace L. (1974). “Language and Consciousness,” Language
50/1: 111–133.
 SELECT BIBLIOGRAPHY 201

Chang, Franklin, Gary S. Dell, Kathryn Bock, and Zenzi M. Griffin.

(2000). “Structural Priming as Implicit Learning: A Comparison
of Models of Sentence Production,” Journal of Psycholinguistic Re-
search 29/2: 217–230.
Cirilo, Randolph K. (1981). “Referential Coherence and Text Structure
in Story Comprehension,” Journal of Verbal Learning and Verbal
Behavior 20/3: 358–367.
Clanchy, Michael T. (1979). From Memory to Written Record 1066–
1307. Cambridge, MA: Harvard University Press.
Clark, Herbert H. and C. J. Sengul. (1979). “In Search of Referents for
Nouns and Pronouns,” Memory & Cognition 7/1: 35–41.
Clifton, Charles, Lyn Frazier, and Cynthia Connine. (1984). “Lex-
ical Expectations in Sentence Comprehension,” Journal of Verbal
Learning and Verbal Behavior 23/6: 696–708.
Clifton, Charles, Adrian Staub, and Keith Rayner. (2007). “Eye Move-
ments in Reading Words and Sentences,” in Roger P. G. van Gom-
pel (ed.), Eye Movements: A Window on Mind and Brain. Amster-
dam: Elsevier: 341–372.
Cobban, Alan B. (1975). The Medieval Universities: Their Develop-
ment and Organisation. London: Methuen.
Cordes, Jason L. (2002). “When Liability Can Attach for Remaining
Passive: Construction of the Term ‘Disposal’ under Cercla: Crofton
Ventures Limited Partnership v. G & H Partnership 1,” Missouri
Environmental Law & Policy Review 9: 67–154.
Costello, Fintan J. and Mark T. Keane. (2000). “Efficient Creativity:
Constraint-guided Conceptual Combination,” Cognitive Science
24/2: 299–349.
Coulson, Seana, Jonathan W. King, and Marta Kutas. (1998). “Expect
the Unexpected: Event-related Brain Response to Morphosyntactic
Violations,” Language & Cognitive Processes 13/1: 21–58.
Crowhurst, Marion. (1990). “Reading/Writing Relationships: An
Intervention Study,” Canadian Journal of Education 15/2: 155–172.
Crystal, David. (2004). The Stories of English. New York: Overlook
Press.
Culicover, Peter W. and Robert D. Levine. (2001). “Stylistic Inversion
in English: A Reconsideration,” Natural Language & Linguistic
Theory 19/2: 283–310.
Cutrell, Edward and Zhiwei Guan. (2007). “What Are You Looking
For?: An Eye-Tracking Study of Information Usage in Web Search,”
202 SELECT BIBLIOGRAPHY

in Proceedings of the SIGCHI Conference on Human Factors in

Computing Systems. New York: Association for Digital Machinery
(ACM).
Daneman, Meredyth. (1982). “The Measurement of Reading Com-
prehension: How Not to Trade Construct Validity for Predictive
Power,” Intelligence 6/4: 331–345.
Daneman, Meredyth and Patricia A Carpenter. (1983). “Individual
Differences in Integrating Information between and within Sen-
tences,” Journal of Experimental Psychology: Learning, Memory,
and Cognition 9/4: 561–584.
Davelaar, Eddy J., Yonatan Goshen-Gottstein, Amir Ashkenazi, Henk
J. Haarmann, and Marius Usher. (2005). “The Demise of Short-
term Memory Revisited: Empirical and Computational Investiga-
tions of Recency Effects,” Psychological Review 112/1: 3–42.
Davis, Kirsten. (2005). “Persuasion through Organization: The Final
Installment – Transitions,” Arizona Attorney: 31–33.
Dee-Lucas, Diana, Marcel Adam Just, Patricia A. Carpenter, and
Meredyth Daneman. (1982). “What Eye Fixations Tell Us about
the Time Course of Text Integration,” in Rudolf Groner and Paul
Fraisse (eds.), Cognition and Eye Movements. Leipzig: North-
Holland: 155–168.
Dehaene, Stanislas. (2009). Reading in the Brain: The Science and
Evolution of a Human Invention. New York: Viking.
DeKay, Sam H. (2010). “Focus on Business Practices: Designing Email
Messages for Corporate Readers: A Case Study of Effective and
Ineffective Rhetorical Strategies at a Fortune 100 Company,” Busi-
ness Communication Quarterly 73/1: 109–119.
Dermer, Marshall L., Shannon L. Lopez, and Paul A. Messling. (2010).
“Fluency Training a Writing Skill: Editing for Concision,” The
Psychological Record 59/1: 3–20.
Diederich, Paul, John W. French, and Sydell T. Carlton. (1961). “Fac-
tors in Judgments of Writing Quality,” Princeton, NJ: Educational
Testing Service: ERIC ED 002 172.
Dor, Daniel. (2003). “On Newspaper Headlines as Relevance Optimiz-
ers,” Journal of Pragmatics 35/5: 695–721.
Duggan, Geoffrey B. and Stephen J. Payne. (2011). “Skim Reading by
Satisficing: Evidence from Eye Tracking,” in Proceedings of the
SIGCHI Conference on Human Factors in Computing Systems. New
York: Association for Digital Machinery (ACM): 1141–1150.
 SELECT BIBLIOGRAPHY 203

Eisenstein, Elizabeth L. (1979). The Printing Press as an Agent of

Change: Communications and Cultural Transformations in Early
Modern Europe. 2 vols. Cambridge: Cambridge University Press.
——(1983). The Printing Revolution in Early Modern Europe. New
York: Cambridge University Press.
Erasmus, Desiderius. (1978). Copia: Foundations of the Abundant
Style (De Duplici Copia Verborum ac Rerum Commentarii Duo).
Trans. Betty Knott. In Craig R. Thompson (ed.), Collected Works of
Erasmus. Vol. xxiv. Toronto: University of Toronto Press.
Evans, Vyvyan. (2014). The Language Myth: Why Language Is Not an
Instinct. Cambridge: Cambridge University Press.
Fahnestock, Jeanne. (1983). “Semantic and Lexical Coherence,” Col-
lege Composition and Communication 34/4: 400–416.
Ferreira, Fernanda. (2003). “The Misinterpretation of Noncanonical
Sentences,” Cognitive Psychology 47: 164–203.
Ferreira, Fernanda, Karl G. D. Bailey, and Vittoria Ferraro. (2002).
“Good-enough Representations in Language Comprehension,”
Current Directions in Psychological Science 11/1: 11–15.
Ferreira, Fernanda, and Charles Clifton. (1986). “The Independence
of Syntactic Processing,” Journal of Memory and Language 25:
348–368.
Ferreira, Fernanda and Janis Stacey. (2000). “The Misinterpretation of
Passive Sentences," unpublished manuscript.
Fischer, Steven Roger. (2005). A History of Reading. Amsterdam: Reak-
tion Books.
Fitzgerald, Jill and Timothy Shanahan. (2000). “Reading and Writing
Relations and Their Development,” Educational Psychologist 35/1:
39–50.
Fleischman, Suzanne. (1990). Tense and Narrativity: From Medieval
Performance to Modern Fiction. Austin, TX: University of Texas
Press.
Flesch, Rudolf. (1974). The Art of Readable Writing. New York:
Harper.
——(1979). How to Write in Plain English: A Book for Lawyers and
Consumers. New York: Harper.
Fowler, H. W. (1965). A Dictionary of Modern English Usage. 2nd edn.
New York: Oxford University Press.
Fox, Julia R., Byungho Park, and Annie Lang. (2007). “When Available
Resources Become Negative Resources: The Effects of Cognitive
204 SELECT BIBLIOGRAPHY

Overload on Memory Sensitivity and Criterion Bias,” Communica-

tion Research 34/3: 277–296.
Frase, Lawrence T. (1969). “Paragraph Organization of Written Mater-
ials: The Influence of Conceptual Clustering upon the Level and
Organization of Recall,” Journal of Educational Psychology 60/5:
394–401.
Frazier, Lyn and Keith Rayner. (1982). “Making and Correcting Errors
during Sentence Comprehension: Eye Movements in the Analysis
of Structurally Ambiguous Sentences,” Cognitive Psychology 14/2:
178–210.
Frazier, Lyn, Lori Taft, Tom Roeper, Charles Clifton, and Kate Ehrlich.
(1984). “Parallel Structure: A Source of Facilitation in Sentence
Comprehension,” Memory & Cognition 12/5: 421–430.
Garnham, Alan, Jane Oakhill, and P. N. Johnson-Laird. (1982). “Ref-
erential Continuity and the Coherence of Discourse,” Cognition
11/1: 29–46.
Gibson, Edward. (1998). “Linguistic Complexity: Locality of Syntactic
Dependencies,” Cognition 68: 1–76.
Glanzer, M. (1972). “Storage Mechanisms in Recall,” in G. H. Bower
and J. T. Spence (eds.), The Psychology of Learning and Motiv-
ation. New York: Academic Press: 129–193.
Glenberg, Arthur M. and Naomi G. Swanson. (1986). “A Temporal
Distinctiveness Theory of Recency and Modality Effects,” Journal
of Experimental Psychology: Learning, Memory, and Cognition
12/1: 3–15.
Gombrich, E. H. (1961). Art and Illusion: A Study in the Psychology of
Pictorial Representation. New York: Bollingen Foundation.
Gordon, Peter C., Randall Hendrick, and William H. Levine. (2002).
“Memory-load Interference in Syntactic Processing,” Psychological
Science 13/5: 425–430.
Graesser, Arthur C., Keith K. Millis, and Rolf A. Zwaan. (1997). “Dis-
course Comprehension,” Annual Review of Psychology 48/1: 163–189.
Graesser, Arthur C., Ashish B. Karnavat, Frances K. Daniel, Elisa
Cooper, Shannon N. Whitten, and Max Louwerse. (2001). “A Computer
Tool to Improve Questionnaire Design,” paper presented at the
Funding Opportunity in Survey Research Seminar, June 11, Bureau
of Labor Statistics, Washington, DC.
Graf, Peter, and Daniel L. Schacter. (1985). “Implicit and Explicit
Memory for New Associations in Normal and Amnesic Subjects,”
 SELECT BIBLIOGRAPHY 205

Journal of Experimental Psychology: Learning, Memory, and Cogni-

tion 11/3: 501–518.
Greene, Steven B. and Gail McKoon. (1995). “Telling Something We
Can't Know: Experimental Approaches to Verbs Exhibiting Impli-
cit Causality,” Psychological Science 6/5: 262–270.
Grice, H. Paul. (1975). “Logic and Conversation,” in P. Cole and J. L.
Morgan (eds.), Speech Acts. New York: Academic Press: 41–58.
Gunning, Robert and Richard A. Kallan. (1994). How to Take the Fog
out of Business Writing. Chicago: Dartnell.
Hagoort, Peter, Colin Brown, and Jolanda Groothusen. (1993). “The
Syntactic Positive Shift (SPS) as an ERP Measure of Syntactic Pro-
cessing," Language & Cognitive Processes 8/4: 439–483.
Hakala, Christopher M. and Edward J. O'Brien. (1995). “Strategies
for Resolving Coherence Breaks in Reading,” Discourse Processes
20/2: 167–185.
Handlin, Amy, Joseph B. Mosca, Dana A. Forgione, and Dennis
Pitta. (2003). “DTC Pharmaceutical Advertising: The Debate's not
Over,” Journal of Consumer Marketing 20/3: 227–237.
Harwood, Nigel. (2005). “‘Nowhere Has Anyone Attempted … in
This Article I Aim to Do Just That’: A Corpus-based Study of Self-
Promotional I and We in Academic Writing across Four Disciplines,”
Journal of Pragmatics 37/8: 1207–1231.
Heider, Fritz and Marianne Simmel. (1944). “An Experimental Study
of Apparent Behavior,” American Journal of Psychology 57/2:
243–259.
Heilman, Kenneth M., Susan A. Leon, and John C. Rosenbek. (2004).
“Affective Aprosodia from a Medial Frontal Stroke,” Brain and
Language 89/3: 411–416.
Herbert, A. S. (1967). Historical Catalogue of Printed Editions of the
English Bible 1525–1961. New York: American Bible Society.
Higgins, Derrick, Jill Burstein, Daniel Marcu, and Claudia Gentile.
(2004). “Evaluating Multiple Aspects of Coherence in Student
Essays,” paper Presented at the Higher Learning/Teaching-NAACL,
Boston, MA. Available as of 2014 via download at www.ets.org/
Media/Research/pdf/erater_higgins_dis_coh.pdf.
Horwitz, B., J. M. Rumsey, and B. C. Donohue. (1998). “Function-
al Connectivity of the Angular Gyrus in Reading and Dyslexia,”
Proceedings of the National Academy of Sciences USA 95: 8939–
8944.
206 SELECT BIBLIOGRAPHY

Hyland, Ken. (2004). “Disciplinary Interactions: Metadiscourse in L2

Postgraduate Writing,” Journal of Second Language Writing 13/2:
133–151.
Hyland, Ken and Polly Tse. (2004). “Metadiscourse in Academic Writ-
ing: A Reappraisal,” Applied Linguistics 25/2: 156–177.
Just, Marcel Adam and Patricia A. Carpenter. (1980). “A Theory of
Reading: From Eye Fixations to Comprehension,” Psychological
Review 87/4: 329–354.
Just, Marcel Adam, Patricia A. Carpenter, Timothy A. Keller, William
F. Eddy, and Keith R. Thulborn. (1996). “Brain Activation Modu-
lated by Sentence Comprehension,” Science 274: 114–116.
Kieras, David E. (1978). “Good and Bad Structure in Simple Para-
graphs: Effects on Apparent Theme, Reading Time, and Recall,”
Journal of Verbal Learning and Verbal Behavior 17/1: 13–28.
King, Jonathan and Marcel Adam Just. (1991). “Individual Differences
in Syntactic Processing: The Role of Working Memory,” Journal of
Memory and Language 30/5: 580–602.
Kintsch, Walter. (1988). “The Role of Knowledge in Discourse Com-
prehension: A Construction-Integration Model,” Psychological
Review 95/2: 163–182.
——(2002). “On the Notions of Theme and Topic in Psychological Pro-
cess Models of Text Comprehension,” Thematics: Interdisciplinary
Studies 3: 157–170.
Kintsch, Walter and Teun A. van Dijk. (1978). “Toward a Model of
Text Comprehension and Production,” Psychological Review 85/5:
363–394.
Kintsch, Walter and Douglas Vipond. (1979). “Reading Compre-
hension and Readability in Educational Practice and Psycho-
logical Theory,” in Lars-Göran Nilsson (ed.), Perspectives on
Memory Research: Essays in Honor of Uppsala University's
500th Anniversary. Hillsdale, NJ: Lawrence Erlbaum Associ-
ates: 329–365.
Klee, Hilary and David Legge. (1976). “Estimates of Concreteness and
Other Indices for 200 Transitive Verbs,” Journal of Experimental
Psychology: Human Learning and Memory 2/4: 497–507.
Knoblauch, C. H. and Lil Brannon. (1984). Rhetorical Traditions and
the Teaching of Writing. Portsmouth, NH: Boynton/Cook.
Landauer, Thomas. (1995). The Trouble with Computers. Cambridge,
MA: MIT Press.
 SELECT BIBLIOGRAPHY 207

Lebovits, Gerald. (2006). “The Department of Redundancy Depart-

ment: Concision and Succinctness – Part II,” New York State Bar
Association Journal 78/7: 44–54.
Lepore, Jill. (2014). “Away from My Desk,” The New Yorker, May 12:
72–74.
Leslie, Alan and Stephanie Keeble. (1987). “Do Six-month-old Infants
Perceive Causality?” Cognition 25: 265–288.
Lieberman, Philip. (2001). “Human Language and Our Reptilian
Brain: The Subcortical Bases of Speech, Syntax, and Thought,”
Perspectives in Biology and Medicine 44/1: 32–51.
Limaye, Mohan R. and Roger D. Cherry. (1987). “Pragmatics, ‘Situ-
ated’ Language, and Business Communication,” Journal of Busi-
ness and Technical Communication 1/1: 68–88.
Locker, Kitty O. (1987). “‘As Per Your Request’: A History of Business
Jargon,” Journal of Business and Technical Communication 1/1:
27–47.
Lorch, Robert F. and Elizabeth P. Lorch. (1985). “Topic Structure
Representation and Text Recall,” Journal of Education Psychology
77/2: 137–148.
Luchins, Abraham S. (1958). “Definitiveness of Impression and
Primacy–Recency in Communications,” Journal of Social Psychol-
ogy 48/2: 275–290.
Lupker, Stephen J. (1984). “Semantic Priming without Association:
A Second Look,” Journal of Verbal Learning and Verbal Behavior
23/6: 709–733.
Mackay, Wendy E. (2000). “Responding to Cognitive Overload: Co-
adaptation between Users and Technology,” Intellectica 30/1:
177–193.
Manguel, Alberto. (1996). A History of Reading. New York: Viking.
Mason, Robert A., Marcel Adam Just, Timothy A. Keller, and Patricia
A. Carpenter. (2003). “Ambiguity in the Brain: What Brain Imag-
ing Reveals About the Processing of Syntactically Ambiguous Sen-
tences,” Journal of Experimental Psychology: Learning, Memory,
and Cognition 29/6: 1319–1338.
Masson, Michael E. (1982). “Cognitive Processes in Skimming Stor-
ies,” Journal of Experimental Psychology: Learning, Memory, and
Cognition 8/5: 400–417.
Matsuda, Paul Kei and Christine M. Tardy. (2007). “Voice in Academic
Writing: The Rhetorical Construction of Author Identity in Blind
208 SELECT BIBLIOGRAPHY

Manuscript Review,” English for Specific Purposes 26/2: 235–

249.
Mayer, Richard E. (1976). “Integration of Information during Prob-
lem Solving due to a Meaningful Context of Learning,” Memory &
Cognition 4/5: 603–608.
McCarthy, Philip M., Adam M. Renner, Michael G. Duncan, Nicholas
D. Duran, Erin J. Lightman, and Danielle S. McNamara. (2008).
“Identifying Topic Sentencehood,” Behavior Research Methods
40/3: 647–664.
McCrum, Robert, William Cran, and Robert MacNeil. (1992). The
Story of English. New York: Penguin.
McNamara, Danielle S., Scott A. Crossley, and Philip M. McCarthy.
(2010). “Linguistic Features of Writing Quality,” Written Commu-
nication 27/1: 57–86.
McWhorter, John. (2001). The Power of Babel: A Natural History of
Language. New York: HarperCollins.
Michael, Erica B., Timothy A. Keller, Patricia A. Carpenter, and Marcel
Adam Just. (2001). “fMRI Investigation of Sentence Comprehen-
sion by Eye and by Ear: Modality Fingerprints on Cognitive Pro-
cesses,” Human Brain Mapping 13: 239–252.
Michotte, A. (1963). The Perception of Causality. New York: Basic
Books.
Miller, George A. (1956). “The Magical Number Seven, Plus or Minus
Two: Some Limits on Our Capacity for Processing Information,”
Psychological Review 63: 81–97.
Miller, James R. and Walter Kintsch. (1980). “Recall and Readability
of Short Prose Passages.” paper presented at the American Educa-
tional Research Association Conference, Boston, MA. April 7–11:
32pp.
Miyake, Akira, Marcel Adam Just, and Patricia A. Carpenter. (1994).
“Working Memory Constraints on the Resolution of Lexical Ambi-
guity: Maintaining Multiple Interpretations in Neutral Contexts,”
Journal of Memory and Language 33/2: 175–202.
Moore, Gordon E. (1965). “Cramming More Components onto Inte-
grated Circuits,” Electronics, April 19: 114–117.
Moreno, Roxana and Richard E. Mayer. (2000). “A Learner-centered
Approach to Multimedia Explanations: Deriving Instructional
Design Principles from Cognitive Theory,” Interactive Multimedia
Electronic Journal of Computer-enhanced Learning 2/2: 12–20.
 SELECT BIBLIOGRAPHY 209

Murdock, Bennet B. (1962). “The Serial Position Effect of Free Recall,”

Journal of Experimental Psychology 64/5: 482–488.
Murray, John D. and Chadd McGlone. (1997). “Topic Overviews and
Processing of Topic Structure,” Journal of Educational Psychology
89/2: 251–261.
Nell, Victor. (1988). Lost in a Book: The Psychology of Reading for
Pleasure. New Haven, CT: Yale University Press.
Nicholas, Serge. (1998). “Perceptual and Conceptual Priming of Indi-
vidual Words in Coherent Texts,” Memory 6/6: 643–663.
Nielsen, Jakob. (2006). “F-shaped Pattern for Reading Web Content,”
Alertbox: Current Issues in Web Usability. Available as of 2014 from
http://www.useit.com/alertbox/reading_patternhtml.
Nishimura, Hiroshi, Kazuo Hashikawa, Katsumi Doi, Takako Iwaki,
Yoshiyuki Watanabe, Hideo Kusuoka, Tsunehiko Nishimura, and
Takeshi Kubo. (1999). “Sign Language ‘Heard’ in the Auditory
Cortex,” Nature 397: 116.
Noordman, Leo, Wietske Vonk, and Henk J. Kempff. (1992). “Causal
Inferences during the Reading of Expository Texts,” Journal of
Memory and Language 31/5: 573–590.
Norman, Suzanne, Susan Kemper, Donna Kynette, Hintat Cheung, and
Cheryl Anagnopoulos. (1991). “Syntactic Complexity and Adults’
Running Memory Span,” Journal of Gerontology 46/6: 346–351.
Olson, David R. (1991). “Literacy and Objectivity: The Rise of Modern
Science,” in David R. Olson and Nancy Torrance (eds.), Literacy
and Orality. New York: Cambridge University Press: 149–164.
Olson, David R. and Nikola Filby. (1972). “On the Comprehension of
Active and Passive Sentences,” Cognitive Psychology 3/3: 361–481.
Ong, Walter J. (1982). Orality and Literacy: The Technologizing of the
Word. New York: Methuen.
Orwell, George. (1986). “Politics and the English Language,” in M.
H. Abrams (ed.), The Norton Anthology of English Literature.
London: W. W. Norton: 2260–2270.
Paivio, Allan. (1990). Mental Representations: A Dual Coding
Approach. New York: Oxford University Press.
Patel, Aniruddh D. (2010). Music, Language, and the Brain. New York:
Oxford University Press.
Paulesu, Eraldo, Christopher D. Frith, and Richard S. J. Frackowiak.
(1993). “The Neural Correlates of the Verbal Component of Work-
ing Memory,” Nature 362/25: 342–345.
210 SELECT BIBLIOGRAPHY

Pedersen, Olaf. (1997). The First Universities: Studium Generale and

the Origins of University Education in Europe. Trans. Richard
North. Cambridge: Cambridge University Press.
Pennebaker, James W. (2011). The Secret Life of Pronouns: What Our
Words Say about Us. New York: Bloomsbury Press.
Perfetti, Charles A. (1999). “Comprehending Written Language: A
Blueprint of the Reader,” in C. M. Brown and P. Hagoort (eds.),
The Neurocognition of Language. Oxford: Oxford University Press:
167–208.
Pickering, Martin J. and Holly P. Branigan. (1998). “The Representa-
tion of Verbs: Evidence from Syntactic Priming in Language Pro-
duction,” Journal of Memory and Language 39: 633–651.
Pickering, Martin J. and Victor S. Ferreira. (2008). “Structural Prim-
ing: A Critical Review,” Psychological Bulletin 134/3: 427–459.
Pickering Martin J. and Simon Garrod. (2013). “Forward Models and
Their Implications for Production, Comprehension, and Dialogue,”
Behavioral and Brain Sciences 36/4: 377–392.
Pinker, Steven. (1994). The Language Instinct: How the Mind Gener-
ates Language. New York: Harper.
Plato. (1973). Phaedrus. Trans. Walter Hamilton. London: Penguin.
Pluchino, Stefano, Angelo Quattrini, Elena Brambilla, Angela Gritti,
Giuliana Salani, Giorgia Dina, Rossella Galli et al. (2003).“Injec-
tion of Adult Neurospheres Induces Recovery in a Chronic Model
of Multiple Sclerosis,” Nature 422: 688–694.
Ramachandran, V. S. (2011). The Tell-Tale Brain: A Neuroscientist's
Quest for What Makes Us Human. New York: W. W. Norton.
Rayner, Keith, Marcia Carlson, and Lyn Frazier. (1983). “The Inter-
action of Syntax and Semantics during Sentence Processing: Eye
Movements in the Analysis of Semantically Biased Sentences,”
Journal of Verbal Learning and Verbal Behavior 22/3: 358–374.
Rayner, Keith, Barbara R. Foorman, Charles A. Perfetti, David
Pesetsky, and Mark S. Seidenberg. (2001). “How Psychological
Science Informs the Teaching of Reading,” Psychological Science in
the Public Interest 2/2: 31–74.
Rhodes, Susan. (1997). “The Active and Passive Voice Are Equally Com-
prehensible in Scientific Writing,” PhD thesis, University of Wash-
ington, 1997. Available as of 2014 at https://digital.lib.washing-
ton.edu/researchworks/bitstream/handle/1773/9033/9819294.
pdf?sequence = 1&isAllowed = y.
 SELECT BIBLIOGRAPHY 211

Richards, I. A.(1930). Practical Criticism: A Study of Literary Judgment.

London: Kegan Paul Trench Trubner.
Ritchey, Kristin A. (2011). “How Generalization Inferences Are Con-
structed in Expository Text Comprehension,” Contemporary Edu-
cational Psychology 36/4: 280–288.
Rumelhart, David. (1975). “Notes on a Schema for Stories,” in David
G. Bobrow and Alan Collins (eds.), Representation and Under-
standing. New York: Academic Press: 211–235.
Salager-Meyer, Françoise. (1994). “Hedges and Textual Communica-
tive Function in Medical English Written Discourse,” English for
Specific Purposes 13/2: 149–170.
Samuels, S. Jay and Michael L. Kamil. (1984). “Models of the Reading
Process,” Handbook of Reading Research 1: 185–224.
Schank, Roger C. (1982). Dynamic Memory: A Theory of Reminding
and Learning in Computers and People. Cambridge: Cambridge
University Press.
Schank, Roger and Robert Abelson. (1977). Scripts, Plans, Goals and
Understanding. Hillsdale, NJ: Lawrence Erlbaum Associates.
Schnotz, Wolfgang and Christian Kürschner. (2007). “A Reconsidera-
tion of Cognitive Load Theory,” Educational Psychology Review 19:
469–508.
Schwinges, Rainer Christoph. (1996). “Student Education, Student
Life,” in Hilde de Ridder-Symoens (ed.), Universities in Early Mod-
ern Europe, 1500–1800. Cambridge: Cambridge University Press:
195–243.
Shanahan, Timothy. (1984). “Nature of the Reading–Writing Relation:
An Exploratory Multivariate Analysis,” Journal of Educational
Psychology 76/3: 466–477.
——(2006). “Relations among Oral Language, Reading, and Writing
Development,” in C. A. MacArthur, S. Graham, and J. Fitzgerald
(eds.), Handbook of Writing Research. New York: Guildford Press:
171–186.
Shanahan, Timothy and Richard G. Lomax. (1986). “An Analysis and
Comparison of Theoretical Models of the Reading–Writing Rela-
tionship,” Journal of Educational Psychology 78/2: 116–123.
Sloan, Gary. (1988). “Relational Ambiguity between Sentences,” Col-
lege Composition and Communication 39/2: 154–165.
Smith, Frank. (2012). Understanding Reading: A Psycholinguistic Analy-
sis of Reading and Learning to Read. 6th edn. New York: Routledge.
212 SELECT BIBLIOGRAPHY

Speedie, Lynn J., Eliahu Wertman, Judy Ta'ir, and Kenneth M. Heil-
man. (1993). “Disruption of Automatic Speech Following a Right
Basal Ganglia Lesion,” Neurology 43/9: 1768–1774.
Sperber, Dan and Deirdre Wilson. (1985). “Loose Talk,” Proceedings of
the Aristotelian Society 86/1: 153–171.
——(2002). “Pragmatics, Modularity and Mind Reading,” Mind &
Language 17/1–2: 3–23.
——(2004). “Relevance Theory,” in Laurence Horn and Gregory
Ward (eds.), Handbook of Pragmatics. Oxford: Blackwell: 607–
632.
Spiro, Rand J. (1980). “Prior Knowledge and Story Processing: Inte-
gration, Selection, and Variation,” Poetics 9/1: 313–327.
Spyridakis, Jan H. and Waka Fukuoka. (2002). “The Effect of Induc-
tively versus Deductively Organized Text on American and Japan-
ese Readers,” IEEE Transactions on Professional Communication
45/2: 99–114.
Stallings, Lynne M., Maryellen C. MacDonald, and Padraig O'Seaghdha.
(1998). “Phrasal Ordering Constraints in Sentence Production:
Phrase Length and Verb Disposition in Heavy-NP Shift,” Journal
of Memory and Language 39: 392–417.
Steinman, Lawrence. (2003). “Collateral Damage Repaired,” Nature
422: 671–672.
Stelzner, Hermann G. (1966). “‘War Message,’ December 8, 1941:
An Approach to Language,” Communications Monographs 33/4:
419–437.
Temperly, David. (2007). “Minimization of Dependency Length in
Written English,” Cognition 105: 300–333.
Thapar, Anjal and Robert L. Greene. (1993). “Evidence against a
Short-term Store Account of Long-Term Recency Effects,” Memory
& Cognition 21/3: 329–337.
Therriault, David J. and Gary E. Raney. (2002). “The Representa-
tion and Comprehension of Place-on-the-Page and Text-Sequence
Memory,” Scientific Studies of Reading 6/2: 117–134.
Trabasso, Tom and Paul van den Broek. (1985). “Causal Thinking and
the Representation of Narrative Events,” Journal of Memory and
Language 24/5: 612–630.
Tversky, Amos and Daniel Kahneman. (1973). “Availability: A Heuris-
tic for Judging Frequency and Probability,” Cognitive Psychology
5: 207–232.
 SELECT BIBLIOGRAPHY 213

Van den Broek, Paul, Robert F. Lorch, Tracy Linderholm, and Mary
Gustafson. (2001). “The Effects of Readers’ Goals on Inference
Generation and Memory for Texts,” Memory & Cognition 29/8:
1081–1087.
Van Dijk, Teun A. (1988). The News as Discourse. Hillsdale, NJ:
Lawrence Erlbaum Associates.
——(2008). “Critical Discourse Analysis and Nominalization: Problem
or Pseudo-Problem?” Discourse & Society 19/6: 821–828.
Wikborg, Eleanor. (1985). “Unspecified Topic in University Student
Essays,” Text – Interdisciplinary Journal for the Study of Discourse
5/4: 359–370.
Williams, Joanna P., Maravene B. Taylor, and Sonia Ganger. (1981).
“Text Variations at the Level of the Individual Sentence and the
Comprehension of Simple Expository Paragraphs,” Journal of
Educational Psychology 73/6: 851–865.
Williams, Joseph M. (1995). Style: Toward Clarity and Grace. Chicago:
University of Chicago Press.
Wydick, Richard C. (1998). Plain English for Lawyers. 4th edn.
Durham, NC: Carolina Academic Press.
Yore, Larry D. and James A. Shymansky. (1985). “Reading, Under-
standing, Remembering and Using Information in Written Science
Materials,” in Association for the Education of Teachers in Science.
Cincinnati, OH: ERIC: 1–59.
Zwaan, Rolf A. (1994). “Effect of Genre Expectations on Text Compre-
hension,” Journal of Experimental Psychology: Learning, Memory,
and Cognition 20/4: 920–933.
——(1996). “Processing Narrative Time Shifts,” Journal of Experimen-
tal Psychology: Learning, Memory, and Cognition 22/5: 1196–1207.
Zwaan, Rolf A., Mark C. Langston, and Arthur Graesser. (1995). “The
Construction of Situation Models in Narrative Comprehension: An
Event-Indexing Model,” Psychological Science 6: 292–297.
Zwaan, Rolf A. and Gabriel A. Radvansky. (1998). “Situation Models
in Language Comprehension and Memory,” Psychological Bulletin
123/2: 162–185.
INDEX

action verbs, 37, 43, 44, 45, 47, 50, Baddeley, Alan, 140
51, 58, 172 Berbice Dutch, 39
active construction, 5, 32, 35, Big Bang Beginnings, 110, 121,
36, 37, 38, 39, 44, 47, 51, 58, 122
84, 173 bottom-up processing, 88, 89
actor, 39, 51, 56 Braille, 147, 154
Adams, Scott, 8, 191, 197 Brannon, Lil, 5, 206
adjective clauses, 55 Broca, Pierre Paul, 151
defining, 169 Broca’s aphasia, 151, 159
definition and use of, 183 Broca’s area, 151, 153, 159
non-defining, 169
punctuation, 184 cadence, 74, 142–63
use of that or which, 170 transition, 143
adjectives California Penal Code, 7, 11, 190
function, 174 Carroll, Lewis, 131, 153
adjectives and adverbs, 174–5 causal, 46
adverb clauses causal transitions, 71, 72
definition and use of, 182 causation, 22, 23, 35, 38, 43, 44,
punctuation, 182–3 51, 71. See cause and effect
adverbs cause and effect, 22, 29, 30, 35,
comparative form, 174 38, 140, 176, 182. See
function, 174 causation
superlative form, 175 cerebellum, 150, 151
Anglo-Saxon, 31, 32, 33, 41, clarity, vii, 12, 28, 29–60, 82, 83,
126 84, 87, 89, 104, 106, 199, 213
angular gyrus, 17, 153, 154 clause, 33
appositive definition, 175
definition and use of, 180–1 clauses, 176–82
appositive phrases definitions and uses of, 182–5
function, 180 function, 175
arcuate fasciculus, 153 major, 175, 176
Aristotle, 2 minor, 176
Artificial Intelligence (AI), 6, 10, 11, types, 176
56, 76 cognitive load, 17, 20, 63
Aspern Papers, The, 157 cognitive overhead, 92
availability, 82 cognitive overload, 56, 63, 193
 215 INDEX

coherence, 85 emphasis, 66, 67, 69, 84, 91, 114,

common grammatical subjects, 115
82, 83 secondary, 80
common wording emphasis position, 70. See
use in continuity, 82, 83 emphasis
complex sentence, 145, 156 English, 31, 35, 120, 126, 127, 128,
compound-complex sentence, 130, 131
145 Erasmus, 129, 146
concision, 118–40 event related potentials (ERPs),
conclusion 16
sentence, paragraph, 115–16 eye-tracking, 15, 68, 78, 82, 86,
consistent grammatical subjects, 96, 97
84
consistent subjects, 78–82 Flesch Reading Ease, 11, 12, 13,
construction 191
active, 5 Flesch–Kincaid, 11, 12, 13, 14,
passive, 5 191
continuity, 61–84, 87, 88, 89, 92, FOG Index, 12, 13, 14, 191, 205
107 Fowler, H. W., 139, 146, 147
French, 15, 31, 35, 41, 120, 126,
dangling modifier, 176 127, 128, 130, 131, 139, 140,
Dead Zones, 99, 106 196, 202
Denby, David, 13, 14, 19, 191 French doublets, 132
diffusion tensor imaging (DTIs), French, Norman, 131
16 French, Parisian, 131, 132
Dilbert, 8, 14 frequency effect, 82
document body, 108 front-loading, 99, 109
document conclusion paragraphs, functional magnetic resource
115, 116 imaging (fMRI), 7, 16, 17,
document head, 109, 110, 112, 114 34, 43, 208
components of, 110 functional words, 152, 159
length, 109
document structure, 108–10 garden path, 13, 19, 25, 56, 75,
Douay Bible, 139 179
Duala, 159 Germanic words, 127
gerund phrase, 54, 55, 56
Eats, Shoots & Leaves, 14, 191 punctuation of, 179
electroencephalograms (EEGs), gerunds, 54, 179
15, 17, 34, 43, 56 definition and use of,
Elements of Style, 3 179–80
email, viii, 86, 102, 133, 136 function, 179
Emerson, Ralph Waldo, 69, 82 Gombrich, E. H., 24, 191
216 INDEX

grammatical subject, 13, 14, 32, Latin, 31, 32, 33, 35, 41, 74, 75,
37, 39, 42, 48, 51, 53, 54, 56, 126, 127, 128, 130, 131, 139,
57, 58, 72, 74, 79, 80, 83, 143, 140, 146, 148
145, 165, 179, 187 lead, 96
common, 79, 80, 81, 84 lectio, 148
graphemes, 149 lexical processing, 25, 34, 35, 43,
Grice, Paul, 122, 124 44, 45, 46, 53, 70, 147, 149
light verbs, 45
handling items in a series, lingual gyrus, 151
160–1 linguistic fossil, 128, 138
headline lists
creating coherence, 95, 96 items in, 68–9
Hemingway, Ernest, 155, 156, 157 London Review of Books, 105
history of English, 120 long sentences, 69–70, 91
history of the written word, 147
Hitchcock, Alfred, 89 macrostructure, 79, 103, 108,
Hyland, Ken, 134, 136 109
magnetoencephalograms (MEGs),
iconicity assumption, 22 16
inference building, 20. See major clauses, 33, 55, 75, 145,
inferential processing 165, 182
inferential processing, 20, 21, 23, metadiscourse, 133–8
24, 25, 26, 27, 28, 34, 35, 70, attitude markers, 134, 135, 137
76, 79, 82, 93, 95, 96, 105, boosters, 134, 135, 138
108 emphatics, 138
infinitive phrase, 55 engagement markers, 134, 135,
infinitives, 172 137, 138
definition and use of, 181–2 evidentials, 134, 135, 136, 137
interactive processing, 89, 114 frame markers, 134, 135, 137
hedges, 135, 136
“Jabberwocky,” 131, 153 hedging self-mention, 136
James, Henry, 156, 157 interactional, 134, 135
interactive, 134, 135
Kahneman, Daniel, 81 self-mentions, 134, 135, 136,
Kansas City Star, 157 137, 138
Keats, John, 94 self-reference, 135
klimax, 160, 162 throat-clearing, 138, 139
Knoblauch, C. H., 5 transitions, 135, 137
Middle Dutch, 130
Language Instinct, The, 38 Miller, George, 68
language style matching, minor clauses, 33, 145, 165, 176,
162 182
 217 INDEX

misplaced modifiers, 177 paragraph conclusion, 115, 125

Modern English Usage, 139, 146 paragraph head, 92, 96–103, 104,
106, 107, 108, 109, 111
negatives, 132–3 paragraphs
neuroplasticity, 154 journalism conventions, 90, 91
New Yorker, The, 13, 17, 19, 207 parietal lobe, 153
Nick Adams Stories, 155 participles, 172
Nixon, Richard M., 119 definition and use of, 179
nominalization, 33, 46, 47, 48, 49, passive construction, 31, 32, 35,
199, 213 36, 37, 38, 39, 40, 42, 43, 45,
nominative case, 164 48, 49, 50, 82, 83, 84, 125,
non-action verbs, 32, 37, 42, 43, 132, 161, 173
44, 45, 172 passive sentence, 38. See passive
Norse, 128, 132 construction
noun clauses Pennebaker, James, 162
definition and uses, 184 Phaedrus, 119, 120
punctuation, 184–5 phonemes, 149
nouns, 32, 53, 165, 166–7 phonological processing, 147, 149
collective, 166 phonological-lexical processing,
common, 166 154
definition and use of, 178–9 phrases, 176
proper, 166 definition, 175
nut graph, 96, 111 types, 176
Pinker, Steven, 38
occipital lobe, 153 Plato, 119, 120
“Ode on a Grecian Urn,” 94 point-last paragraph, 104
Old Danish, 128 “Politics and the English
Old English, 126, 127, 128, 130, Language,” 32, 140
140 positron emission tomography
Old High German, 128, 130 (PET), 16, 150, 153
Old Norse, 31, 127 posterior inferior frontal gyrus, 151
online text, 86 Practical Criticism, 26, 211
Orwell, George, 32, 86, 140 prediction, 64, 65, 66, 70, 74, 90,
102, 108, 109, 119
paragraph preliminary thesis, 113–14
focus, 91, 92 preposition, 33, 36
main point, 88 prepositional phrases, 36, 45, 55,
optimal organization, 91 67
structure, 87 definition and uses of, 178
paragraph body, 92, 99, 101, function, 178
103–8 primacy effect, 99, 104, 106, 115
minimum length, 107 priming, 79, 93
218 INDEX

priming effect, 76, 79, 81, 82, 93, schema, 23, 24, 25, 27, 35, 64–5,
99, 103, 104, 106, 108, 111, 79, 88, 89, 107, 111, 191
112, 116 sequencing, 76–8, 80, 81, 82,
pronoun, 32, 53, 168 83, 84
number, 166 placement of material, 78
referents, 168, 169 sign language, 147, 154
pronouns, 167–72 silent reading, 148
gender, 166 simple sentence, 145
object, 171 speech–auditory–visual loop,
possession, 171 151
self-reference, 172 split infinitives, 74, 75, 177
subject, 170 St. Ambrose, 148
prospect, 119 St. Augustine, 148
Psycho, 88, 89 Strunk, William and White, E. B.,
punctuation 3, 4, 126, 177
colon, 188–9 Style: Toward Clarity and Grace,
comma, 186–7 104
dash, 189 subject pronoun, 164
periods, 185 subjects, 165–72
question mark, 185–6 subject–verb agreement, 166
semicolon, 187–8 subject–verb–object order, 38, 39,
42, 55, 182
readability formulas, 11, 12, 14 subvocalization, 140
readability outcomes, 26, 27 supplementary motor area (SMA),
recency effect, 68, 106, 112, 114, 150
115, 116 supratemporal gyri, 154
redundant modifiers, 129–31, syntactic, 34. See syntactic
141 processing
redundant pairs, 127–8, 129, 130, syntactic processing, 18, 19, 20,
131 23, 25, 34, 35, 44, 45, 46, 53,
referential continuity, 77, 78 68, 70, 94, 144, 161, 162, 193,
relevance 198
assumptions of, 76
repetition, 110, 116, 125 temporal lobe, 153
retrospect, 119 textual evidential, 137
Richards, I. A, 25, 26, 93, 94, 116, textual frame marker, 137
211 there is, 40, 41, 42, 48
Roosevelt, Franklin Delano, 40 thesis, 110, 111, 112, 113, 114,
115
saccades, 15, 34 definition, 113
scala, 160 importance of, 113
Schank, Roger, 56 types, 113
 219 INDEX

thesis sentence, 110–13 varying sentence lengths, 155–6

throat-clearing, 121, 140 varying sentence structure,
top-down processing, 88, 89, 114 143–5
topic sentence, 92, 93, 97, 100 verbs, 32, 37, 172–5
training wheels effect, 75, 81 default mood, 173
transitions, 66, 70–5, 77, 78, 80, imperative mood, 173
81, 83, 84, 102, 145 subjunctive mood, 173
absence of, 78
additive, 75 Watergate, 118
amplification, 72 weasel words. See metadiscourse:
causal, 71 hedges
conclusion, 72 web pages, 97
conditions, 72 Wernicke, Karl, 151
continuity, 71 Wernicke's aphasia, 152, 153,
contrast or exception, 71 159
example, 72 Wernicke's area, 17, 151, 153,
frequency or time, 72 159
journalism, 73, 74 Williams, Joseph, 104, 133, 134,
optimal number, 74 138
order, 72 word-level processing, 20. See
placement, 72, 73 lexical processing
result, 72 working memory, 68, 70, 103,
time-shifts, 72 107, 110, 114, 161
Truss, Lynne, 14, 191
Tversky, Amos, 81 zombies, 37, 48, 50