Lectures on Infinitary Model Theory 1st Edition

David Marker install download

https://ebookname.com/product/lectures-on-infinitary-model-
theory-1st-edition-david-marker/

Get the full ebook with Bonus Features for a Better Reading Experience on ebookname.com
Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

On Time Lectures on Models of Equilibrium Churchill

Lectures in Economics 1st Edition Peter A. Diamond

https://ebookname.com/product/on-time-lectures-on-models-of-
equilibrium-churchill-lectures-in-economics-1st-edition-peter-a-
diamond/

Logic Without Borders Essays on Set Theory Model Theory

Philosophical Logic and Philosophy of Mathematics 1st
Edition Asa Hirvonen

https://ebookname.com/product/logic-without-borders-essays-on-
set-theory-model-theory-philosophical-logic-and-philosophy-of-
mathematics-1st-edition-asa-hirvonen/

Social Theory Twenty Introductory Lectures 1st Edition

Hans Joas

https://ebookname.com/product/social-theory-twenty-introductory-
lectures-1st-edition-hans-joas/

Advanced Router Architectures 1st Edition Axel K. Kloth

https://ebookname.com/product/advanced-router-architectures-1st-
edition-axel-k-kloth/
Modern Digital and Analog Communication Systems 4th
Edition B. P. Lathi

https://ebookname.com/product/modern-digital-and-analog-
communication-systems-4th-edition-b-p-lathi/

Bilinguals Cognition Education and Language Processing

Cognition Education and Language Processing 1st Edition
Earl F. Caldwell

https://ebookname.com/product/bilinguals-cognition-education-and-
language-processing-cognition-education-and-language-
processing-1st-edition-earl-f-caldwell/

Siena the Heart of Tuscany 1st Edition Rebecca Ford

https://ebookname.com/product/siena-the-heart-of-tuscany-1st-
edition-rebecca-ford/

Lacey Wells and Quick Reconstructing Criminal Law Text

and Materials Law in Context 4th Edition Celia Wells

https://ebookname.com/product/lacey-wells-and-quick-
reconstructing-criminal-law-text-and-materials-law-in-
context-4th-edition-celia-wells/

Inspired by Biology From Molecules to Materials to

Machines 1st Edition Committee On Biomolecular
Materials And Processes

https://ebookname.com/product/inspired-by-biology-from-molecules-
to-materials-to-machines-1st-edition-committee-on-biomolecular-
materials-and-processes/
Teaching mysticism 1st Edition Parsons

https://ebookname.com/product/teaching-mysticism-1st-edition-
parsons/
 i i

i i

i i

i i
9781107181939pre CUP/MARKER-L4 June 16, 2016 17:52 Page-1

Lectures on Infinitary Model Theory

Infinitary logic, the logic of languages with infinitely long conjunctions, plays an
important role in model theory, recursion theory and descriptive set theory. This book
is the first modern introduction to the subject in 40 years, and will bring students and
researchers in all areas of mathematical logic up to the threshold of modern research.
The classical topics of back-and-forth systems, model existence techniques,
indiscernibles and end extensions are covered before more modern topics are
surveyed. Zilber’s Categoricity Theorem for quasiminimal excellent classes is proved
and an application is given to covers of multiplicative groups. Infinitary methods are
also used to study uncountable models of counterexamples to Vaught’s Conjecture,
and effective aspects of infinitary model theory are reviewed, including an introduction
to Montalbán’s recent work on spectra of Vaught counterexamples. Self-contained
introductions to effective descriptive set theory and hyperarithmetic theory are
provided, as is an appendix on admissible model theory.

D AV I D M A R K E R is LAS Distinguished Professor of Mathematics at the University

of Illinois at Chicago and a Fellow of the American Mathematical Society. His main
area of research is model theory and its connections to algebra, geometry and
descriptive set theory. His book, Model Theory: An Introduction, is one of the most
frequently used graduate texts in the subject and was awarded the Shoenfield prize for
expository writing by the Association for Symbolic Logic.
9781107181939pre CUP/MARKER-L4 June 16, 2016 17:52 Page-2

L E C T U R E N OT E S I N L O G I C

A Publication of The Association for Symbolic Logic

This series serves researchers, teachers, and students in the field of symbolic
logic, broadly interpreted. The aim of the series is to bring publications
to the logic community with the least possible delay and to provide rapid
dissemination of the latest research. Scientific quality is the overriding
criterion by which submissions are evaluated.

Editorial Board
Jeremy Avigad
Department of Philosophy, Carnegie Mellon University
Zoe Chatzidakis
DMA, Ecole Normale Supérieure, Paris
Peter Cholak, Managing Editor
Department of Mathematics, University of Notre Dame, Indiana
Volker Halbach
New College, University of Oxford
H. Dugald Macpherson
School of Mathematics, University of Leeds
Slawomir Solecki
Department of Mathematics, University of Illinois at Urbana–Champaign
Thomas Wilke
Institut für Informatik, Christian-Albrechts-Universität zu Kiel

More information, including a list of the books in the series, can be found at
http://www.aslonline.org/books-lnl.html
9781107181939pre CUP/MARKER-L4 June 16, 2016 17:52 Page-3

L E C T U R E N OT E S I N L O G I C 4 6

Lectures on Infinitary Model Theory

DAV I D MA R K E R
University of Illinois, Chicago

association for symbolic logic

9781107181939pre CUP/MARKER-L4 June 16, 2016 17:52 Page-4

University Printing House, Cambridge CB2 8BS, United Kingdom

One Liberty Plaza, 20th Floor, New York, NY 10006, USA

477 Williamstown Road, Port Melbourne, VIC 3207, Australia
4843/24, 2nd Floor, Ansari Road, Daryaganj, Delhi - 110002, India
79 Anson Road, #06-04/06, Singapore 079906
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/9781107181939
10.1017/9781316855560
Association for Symbolic Logic
Richard Shore, Publisher
Department of Mathematics, Cornell University, Ithaca, NY 14853
http://www.aslonline.org

c Association for Symbolic Logic 2016
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 2016
A catalogue record for this publication is available from the British Library.
ISBN 978-1-107-18193-9 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party Internet Web sites referred to in this publication
and does not guarantee that any content on such Web sites is, or will remain, accurate
or appropriate.
 i i

i i

To my mother Vivian and the memory of my father Charlie

i i

i i
 i i

i i

i i

i i
 i i

i i

CONTENTS

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Part 1. Classical results in infinitary model theory

Chapter 1. Infinitary languages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.1. Fragments and Downward Löwenheim–Skolem . . . . . . . . . . . . 10
1.2. Lù1 ,ù and omitting first order types. . . . . . . . . . . . . . . . . . . . . . . . . 13
Chapter 2. Back and forth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.1. Karp’s Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.2. Scott’s Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.3. Countable approximations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.4. Larger infinitary languages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Chapter 3. The space of countable models . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.1. Spaces of ô-structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.2. The number of countable models . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.3. Scattered sentences and Morley’s proof . . . . . . . . . . . . . . . . . . . . . 34
Chapter 4. The Model Existence Theorem . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.1. Consistency properties and model existence . . . . . . . . . . . . . . . . . 39
4.2. Omitting types and atomic models . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.3. The Interpolation Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4.4. The undefinability of well-ordering . . . . . . . . . . . . . . . . . . . . . . . . . 49
Chapter 5. Hanf numbers and indiscernibles . . . . . . . . . . . . . . . . . . . . . . 51
5.1. The Erdös–Rado Partition Theorem . . . . . . . . . . . . . . . . . . . . . . . 52
5.2. The Hanf number of Lù1 ,ù . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
5.3. Morley’s Two Cardinal Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.4. Completely characterizing ℵ1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Part 2. Building uncountable models

Chapter 6. Elementary chains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
6.1. Elementary end extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65

vii

i i

i i
 i i

i i

viii Contents

6.2. Omitting types in end extensions . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

6.3. Uncountable models realizing few types . . . . . . . . . . . . . . . . . . . . 73
6.4. Extending models of set theory. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
6.5. ℵ1 -categorical sentences have models in ℵ2 . . . . . . . . . . . . . . . . . . 79
Chapter 7. Vaught counterexamples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
7.1. Minimal counterexamples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
7.2. Harrington’s Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Chapter 8. Quasiminimal excellence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
8.1. Categoricity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
8.2. Covers of C× . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

Part 3. Effective considerations

Chapter 9. Effective descriptive set theory . . . . . . . . . . . . . . . . . . . . . . . 115
9.1. Recursion theory review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
9.2. Computable functions on ù ù . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
9.3. The arithmetic hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
9.4. The effective projective hierarchy . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
9.5. Recursive ordinals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Chapter 10. Hyperarithmetic sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
10.1. Borel codes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
10.2. Recursively coded Borel sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
10.3. Hyperarithmetic sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
10.4. The Effective Perfect Set Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . 132
Chapter 11. Effective aspects of Lù1 ,ù . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
11.1. Coding Lù1 ,ù -formulas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
11.2. Kreisel–Barwise Compactness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
11.3. Effective analysis of Scott rank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Chapter 12. Spectra of vaught counterexamples . . . . . . . . . . . . . . . . . . 151
12.1. Determinacy and Turing degrees . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
12.2. Montalbán’s Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Appendix A. ℵ1 -free abelian groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Appendix B. Admissiblility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
B.1. Kripke–Platek set theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
B.2. Admissible sets and HYP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
B.3. Model theory of admissible fragments . . . . . . . . . . . . . . . . . . . . . . 167
B.4. C1 = Q . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181

i i

i i
 i i

i i

INTRODUCTION

This book grew out of two courses at the University of Illinois at Chicago. The
first was team taught with John Baldwin, Meeri Kesälä and David Kueker in
the fall of 2007 and the second I taught in Fall 2013. The main goal of my
lectures in 2007 was to provide background material on the classical results on
the model theory of infinitary languages needed to appreciate current work
on abstract elementary classes that would be the focus of the other lecturers.
When I returned to the material in 2013, this was still in my mind. One goal
of the course was to cover the introductory material on the model theory of
infinitary languages that would prepare someone interested in studying John
Baldwin’s Categoricity or the recent work of Zilber and his school around
quasiminimal excellent classes. My second goal was to explore some of the
connections between model theory and descriptive set theory, particularly
effective descriptive set theory. Montálban’s recent gem [59] convinced me it
was timely to include this material. To make this material more accessible, I
have included more-or-less self contained introductions to the needed effective
descriptive set theory and hyperarithemetic theory. A dominant theme related
to both goals is to study properties of counterexamples to Vaught’s Conjecture.
In more detail, Part 1 focuses on foundations of the model theory of
L∞,ù and Lù1 ,ù . In Chapter 1 we examine the expressive power of infinitary
languages and prove the downward Löwenheim–Skolem Theorem. We also
prove Chang’s results that Lù1 ,ù -axiomatizable classes can be characterized as
reducts of models of a first order theory omitting a set of first order types.
In Chapter 2 we give several back-and-forth criteria characterizing L∞,ù -
equivalence and introduce the fundamental notion of Scott analysis and prove
Scott’s Isomorphism Theorem. Section 2.3 is devoted to Kueker’s characteriza-
tion of L∞,ù -equivalence via countable approximations and Section 2.4 gives
the briefest taste of the model theory of fragments of L∞,κ for uncountable κ.
Descriptive set theory starts to play a role in Chapter 3 when we introduce a
Polish topology on the space of countable ô-structures. These ideas are the key
to Morley’s theorem on the number of countable models. This is the first place
we start to understand potential counterexamples to Vaught’s Conjecture, a
theme that will recur frequently.

i i

i i
 i i

i i

2 Introduction

While the Compactness Theorem fails in Lù1 ,ù , Henkin arguments still

play a useful role. Chapter 4 is devoted to the Model Existence Theorem
and its consequences. We will prove the appropriate version of Omitting
Types Theorem, discuss prime and atomic models, and prove Lopez-Escobar’s
Interpolation Theorem and the undefinability of well order. We will show
how the Interpolation Theorem can be used to show the spaces of models
of an Lù1 ,ù -sentence are exactly the invariant Borel subsets of the space of
ô-structures.
Chapter 5 introduces the Erdös–Rado partition theorem and uses it to
construct infinitary indiscernibles. Indiscernibles are used to prove that the
Hanf number for Lù1 ,ù is iù1 . In Section 5.4 we give Hjorth’s proof of Knight’s
result that there are complete Lù1 ,ù -sentences with models of size ℵ1 but no
higher cardinality. We also show how he uses these ideas to show that if there
are counterexamples to Vaught’s Conjecture, then there are counterexamples
with no models of size ℵ2 . Hjorth’s original proof used ideas from descriptive
set theory and dynamical systems. Here we give a straightforward model
theoretic proof from [4].
Part 2 is devoted to a number of constructions of uncountable models.
Chapter 6 begins with Keisler’s methods for constructing uncountable models
by building elementary chains of end extensions. These constructions are
then used to build uncountable models realizing few types. Here we closely
follow Keisler’s presentation from [37]. In Section 6.4 we give a nice proof by
Larson of a theorem of Keisler and Morley on extending models of set theory.
Section 6.5 contains Shelah’s elegant proof that an ℵ1 -categorical sentence
must have a model of size ℵ2 .
In Chapter 7 we consider uncountable models of Vaught couterexamples.
First we prove results of Harnik and Makkai showing that a Vaught counterex-
ample has models of size ℵ1 . We then prove Harrington’s result that a Vaught
counterexample has models of arbitrarily high Scott rank below ù2 . This is
a striking contrast to Hjorth’s result. Harrington’s original proof relied on
admissible model theory. Here we give a new proof from [4] that does not.
Much of the recent interest in infinitary languages has been generated by
the work of Zilber and others on complex-like exponential fields and universal
covers of commutative algebraic groups. Chapter 8 is devoted to the key model
theoretic result Zilber’s Categoricity Theorem for quasiminimal excellent
classes. Here we closely follow the proof given by Kirby in [39]. We also
give some details of the simplest interesting application of the method, the
categoricity for covers of the multiplicative group C× .
In Part 3 we look at some effective issues. Chapters 9 and 10 are intended to
be a self contained treatment of the results we need from effective descriptive set
theory and hyperarithmetic theory, though I’ve taken the liberty of including
some of my favorite results that are not needed in the sequel. In Chapter 11 we
use these tools to look at questions about bounds on Scott rank and effective

i i

i i
 i i

i i

Introduction 3

constructions of models. In particular, we prove Sacks’ theorem about Scott

ranks of Vaught counterexamples. We also prove a compactness theorem due to
Kreisel that anticipates the later work of Barwise. In Chapter 12 we prove one of
Montálban’s results on computability spectra of Vaught counterexamples after
first proving some prerequisite results of Martin on the effect of determinacy
on the Turing degrees.
When I first began this project my intention was to steer clear of admissible
model theory. My view has always been that, while admissible model theory
provides a fascinating bridge between set theory and recursion theory, it leads
to few real model theoretic insights that are unavailable by other means. In
the end, having developed the basics of hyperarithmetic theory, it seemed a
shame not to tie this all together. I decided, as a compromise, to include some
introductory material in an appendix.
There is of course much more that could have been included, but I have
mostly restrained myself to presenting material covered in my lectures. An
introduction to L(Q) and Lù1 ,ù (Q), for example, would be welcome, but I
leave that for others.
Prerequisites. The course this book was based on was taught as a topics
course to advanced graduate students most of whom had already taken
introductory courses in model theory, set theory and descriptive set theory, but
knew almost no recursion theory. While a broad preparation is ideal, much of
the book can be read with a more limited background.
• Model Theory: The reader should be familiar with basic model theory
including Henkin arguments, the Löwenheim–Skolem theorem, back-
and-forth arguments, omitting types, prime and atomic models and
indiscernibles. Chapters 1-5 of [56] are good preparation.
• Set Theory: Most of the book can be read with only a background in
naive set theory. Closed unbounded and stationary sets are useful at
a few points. Forcing and absoluteness arguments are used briefly in
Sections 3.2 and 6.4, and heavily in Section 7.2. Familiarity with models
of set theory and Gödel’s constructible universe is assumed in Appendix B.
Good references are [46] and [30].
• Descriptive Set Theory: In Chapter 3 I assume familiarity with Polish
spaces and Borel sets. Effective descriptive set theory plays a major
role in Part III. I introduce the effective ideas in Chapter 9, but assume
familiarity with classical results from the early chapters of [34].
• Recursion Theory: I attempt to give a self-contained introduction to
everything needed in Section 9.1 and Chapter 10.
In Section 8.2 we prove one case of Zilber’s Thumbtack Lemma. This argument
requires some comfortability with algebraic geometry.
Acknowledgments. This book would not have been written without the
influence of my colleague and friend John Baldwin, whom I thank for countless

i i

i i
 i i

i i

4 Introduction

conversations on infinitary model theory and many comments on earlier

versions of these lecture notes.
I have also learned a great deal from discussing these issues over the years
with a great many mathematicians. I would particularly like to thank Martin
Bays, Victor Harnik, Leo Harrington, Bradd Hart, Alekos Kechris, Meeri
Kesälä, Jonathan Kirby, Julia Knight, David Kuecker, Paul Larson, Chris
Laskowski, Angus Macintyre, Antonio Montálban, Mark Nadel, Gerald
Sacks, and Boris Zilber, as well as Alan Mekler and Greg Hjorth, both of
whom are greatly missed.
I am very grateful to the students in my 2013 course who gave detailed
comments on earlier drafts of these lectures: Gabe Conant, Maxwell Levine,
Tori Noquez, Caroline Terry, Phil Wesolek, Sam Ziegler and Joe Zelinski.
I first learned infinitary model theory from Keisler’s beautiful, but, sadly,
out of print, Model Theory of Infinitary Logic. There are many places where
I have found it difficult to improve on his exposition. Chapters 4 and 6, in
particular, closely follow Keisler’s presentation.
Chicago 2015

i i

i i
 i i

i i

Part 1

CLASSICAL RESULTS IN INFINITARY MODEL THEORY

i i

i i
 i i

i i

i i

i i
 i i

i i

Chapter 1

INFINITARY LANGUAGES

Let ô be a vocabulary, i.e., a set of function symbols, relation symbols and

constant symbols. In the logic L∞,ù (ô)V we build
W formulas using the symbols
from ô, equality, Boolean connectives ¬, and , quantifiers ∀ and ∃, variables
{vα : α an ordinal}.1
• terms and atomic formulas are defined as in first order logic;
• if φ is a formula, then so is ¬φ;
• if X is a set of formulas, then so are
_ ^
φ and φ;
φ∈X φ∈X

• if φ is a formula, so are ∃vα φ and ∀vα φ.

We extend the usual definition of satisfaction by saying
_
M |= φ if and only if M |= φ for some φ ∈ X
φ∈X

and ^
M |= φ if and only if M |= φ for all φ ∈ X.
φ∈X
For notational simplicity, we use the symbols ∧ and ∨ as abbreviations for
binary conjunctions and disjunctions. Similarly we will use the abreviations →
and ↔ when helpful.
We can inductively define the notions of free variable, subformula, sentence,
theory and satisfiability in the usual ways.
Exercise 1.0.1. Suppose φ is an L∞,ù -sentence and ø is a subformula of φ.
Prove that ø has only finitely many free variables.
Let κ be an infinite cardinal. In the logic Lκ,ù (ô) we form formulas in
a similar way but we only use variables
W {vα :Vα < κ} and restrict infinite
conjunctions and disjunctions to φ∈X φ and φ∈X φ where |X | < κ. Thus
Lù,ù is just the usual first order logic. Throughout these notes we be focusing
1 When no confusion arises we omit the ô and write L∞,ù instead of L∞,ù (ô).

i i

i i
 i i

i i

8 1. Infinitary languages

primarily on Lù1 ,ù , the logic where we allow countable conjunctions and

disjunctions.2
Exercise 1.0.2. Show that if κ is a regular cardinal and φ is a sentence of
Lκ,ù , then φ has fewer than κ subformulas. Show that this fails for singular
cardinals. This is one reason it is customary to restrict attention to Lκ,ù for κ
a regular cardinal.
Definition 1.0.3. We say M ≡∞,ù N if
M |= φ if and only if N |= φ
for all L∞,ù sentences φ. The notion M ≡κ,ù N , is defined analogously.
Exercise 1.0.4. Show that if M ∼
= N , then M ≡∞,ù N .
When studying the model theory of infinitary logics there is one fundamental
and inescapable fact:
The compactness theorem fails for infinitary languages.
Exercise 1.0.5. Let ô be the vocabulary with constant symbols d, c0 , c1 , . . .
and let Γ be the set of sentences
 _ 
{d 6= ci : i ∈ ù} ∪ ∀v v = ci .
i∈ù

Show that every finite subset of Γ is satisfiable, but Γ is not satisfiable. Thus
the Compactness Theorem fails for L∞,ù and even Lù1 ,ù .
The failure of compactness will lead to many new phenomena and force us
to find new approaches and develop new tools.3
Exercise 1.0.6. (a) Give an example of structures M0 , M1 , . . . and φ ∈
Lù1 ,ù such
Q that Mi |= φ for all i, but if U is a nonprincipal ultrafilter on
ù then Mi /U |= ¬φ.
(b) Show that if U is a ó-complete ultrafilter on I , then
Y
Mi /U |= φ ⇔ {i ∈ I : Mi |= φ} ∈ U
i∈I

for φ ∈ Lù1 ,ù .
2 Logically it would make sense to call this logic L
ℵ1 ,ℵ0 but we will follow the historical precedent
and refer to it as Lù1 ,ù .
3 Though we will not focus on it, another fruitful approach is to look for more general forms

of the Compactness Theorem that hold in particular settings. One example of this is Barwise
Compactness for countable admissible fragments. We will see an avatar of these results in
Theorem 11.2.2 and give a quick treatment of the general result in Appendix B, but the interested
reader should consult [8].
Another important example is studying compactness in languages Lκ,κ where κ is a large
cardinal. See [31] or [30] for further information.

i i

i i
 i i

i i

1. Infinitary languages 9

If compactness fails, why then do we study the model theory of infinitary

languages? One reason is that we get new insights about first order model
theory. But the simplest answer is that there are many natural classes that are
axiomatized by Lù1 ,ù -sentences.
Exercise 1.0.7. Show that the following classes are Lù1 ,ù -axiomatizable for
appropriate choices ô.
(a) torsion abelian groups;
(b) finitely generated groups;
(c) non-finitely generated groups;
(d) linear orders isomorphic to (Z, <);
(e) archimedian fields;
(f) connected graphs;
(g) recursively saturated models of PA;
(h) ù-models of ZFC (i.e., models of ZFC where the integers are standard);
(i) models of T omitting p, where T is a first order theory and p is a type.
Exercise 1.0.8. Show by compactness that none on these classes is axioma-
tizable in first order logic.
Examples (ii), (iv) and (v) immediately show the failure of the Upward
Löwenheim–Skolem Theorem in infinitary languages. There are no uncount-
able linear orders isomorphic to (Z, <) and any archimedian ordered field is
isomorphic to a subfield of the real numbers and hence has cardinality at most
2ℵ0 . We give another example in Exercise 1.1.13.
Exercise 1.0.9. Show by induction that for ordinal α there is an L∞,ù -
sentence Φα describing (α, <) up to isomorphism.
κ
Exercise 1.0.10. Let κ be a regular cardinal. Show there are α, â < (22 )+
such that (α, <) ≡κ,ù (â, <).
Taken together these two exercises give examples of structures M, N with
M ≡κ,ù N but M 6≡∞,ù N for all κ.4
We will also, from time to time, look at pseudoelementary classes which are
simply reducts of elementary classes.
Definition 1.0.11. We say that a class K of ô-structures is a PCù1 ,ù -class if
there is a vocabulary ô ∗ ⊇ ô and φ ∈ Lù1 ,ù (ô ∗ ) such that
K = {M : there is a ô-structure M∗ expanding M with M∗ |= φ}
i.e., K is the class of ô-reducts of models of φ.
Similarly we say that K is a PC-class if φ is a first order sentence and K is a
PCä -class if φ is a first order theory.
4 For a specific example: suppose κ < ë are uncountable cardinals. Show that κ ≡
ù1 ,ù ë, but
κ 6≡∞,ù ë. See [43] for details.

i i

i i
 i i

i i

10 1. Infinitary languages

Exercise 1.0.12. Show that the following classes are PCù1 ,ù .

(a) orderable groups;
(b) free groups;
(c) 1-trainsitive linear orders, i.e., linear orders where for any a, b there is an
order automorphism taking a to b;
(d) fields that are pure transcendental extensions of Q;
(e) incomplete ordered fields;
(f) ordered fields with an integer part (i.e., an ordered subring with no element
between 0 and 1 such that every element of the field is within distance at
most one of some element of the subring).
Exercise 1.0.13 (Silver). Let ô = {U } where U is unary and let K = {M :
|M| ≤ 2|U (M)| ∧ |M \ U (M)| = |M|}.
(a) Show that K is PC-class.
(b) Show that K is κ-categorical if and only if κ = iα for some limit ordinal α.
This example shows that the straightforward generalization of Morley’s Cate-
goricity Theorem to PC-classes fails
Exercise 1.0.14. Let K be a PCù1 ,ù -class. Show that
K0 = {M ∈ K : M countable}
is also a PCù1 ,ù -class.

1.1. Fragments and Downward Löwenheim–Skolem

We will prove a useful and natural version of the Downward Löwenheim–
Skolem Theorem. The next exercise shows that even here we will need to be
careful.
Exercise 1.1.1. Give an example of a countable vocabulary ô and an Lù1 ,ù -
theory T such that every model of T has cardinality at least 2ℵ0 .
We will often restrict our attention to subcollections of the set of all Lù1 ,ù -
formulas. We will look at collections of formulas with some natural closure
properties. One of these will be formal negation an operation that shows how
we inductively move a negation inside a quantifier or a Boolean operation.
Closure under formal negation isn’t needed in this proof, but will be used in
later arguments.
Definition 1.1.2. For each formula φ we define ∼φ, a formal negation of φ
as follows:
(i) for φ atomic, ∼φ is ¬φ;
(ii) ∼(¬φ)
V is φ; W W V
(iii) ∼ φ∈X φ is φ∈X ∼φ and ∼ φ∈X φ is φ∈X ∼φ;
(iv) ∼∃vφ is ∀v∼φ and ∼∀vφ is ∃v∼φ.

i i

i i
 i i

i i

1.1. Fragments and Downward Löwenheim–Skolem 11

Exercise 1.1.3. Show by induction that

M |= ¬φ if and only if M |= ∼φ
for all φ ∈ L∞,ù .
Definition 1.1.4. We say that a set of L∞,ù -formulas A is a fragment if
there is an infinite set of variables V such that if φ ∈ A, then all variables
occurring in φ are in V and A satisfies the following closure properties:
(i) all atomic formulas using only the constant symbols of ô and variables
from V are in A;
(ii) if φ ∈ A and ø is a subformula of φ, then ø ∈ A;
(iii) If φ ∈ A, v is free in φ, and t is a term where every variable is in V , then
the formula obtained by substituting t into all free occurrences of v is
in A;
(iv) A is closed under ∼;
(v) A is closed under ¬, ∧, ∨, ∃v, and ∀v for v ∈ V .
Exercise 1.1.5. (a) Suppose κ is regular (in particular this holds for Lù1 ,ù ).
Prove that if T is a set of Lκ,ù -sentences with |T | < κ, then there is A a
fragment of Lκ,ù such that T ⊆ A and |A| < κ.
(b) Show that there is a smallest such fragment.
Exercise 1.1.6. Let T and A be as above. Show that every formula in A
has only finitely many free variables. Give an example showing that even
though are only finitely many free variables, there may be infinitely many
bound occurrences.
Exercise 1.1.7. We say that a formula is in negation normal form if the ¬ only
occurs applied to atomic formulas. Show by induction that for any fragment A
and any L∞,ù -formula φ ∈ A, there is ø ∈ A such that ø is equivalent to φ
and ø is in negation normal form.
Conjunctive and disjunctive normal form does not work as well.
Exercise 1.1.8. Let ô = {Pi,j : i, j ∈ ù} ∪ {c} where each Pi,j is a unary
predicate and c is a constant symbol. Consider the Lù1 ,ù -sentence
^ ^ ^ _
(Pi,j (c) → ¬Pi,k (c)) ∧ Pi,j (c).
i∈ù j6=k i∈ù j∈ù

Show that there is an equivalent sentence in disjunctive normal form in L(2ℵ0 )+ ,ù .

(See also Exercise 3.1.6.)
We write M ≡A N and M ≺A N for elementary equivalence and elementary
submodels with respect to formulas in A (where A could be Lù1 ,ù or L∞,ù ).
Theorem 1.1.9 (Downward Löwenheim–Skolem). Let A be a fragment of
L∞,ù such that any formula φ ∈ A has at most finitely many free variables.

i i

i i
 i i

i i

12 1. Infinitary languages

Let M be a ô-structure with X ⊆ M. There is N A M with X ⊆ N and

|N | ≤ max(|A|, |X |).
In particular, if ô is countable and A is a countable fragment of Lù1 ,ù (ô), then
every ô-structure has a countable A-elementary submodel.
The proof is a simple generalization of the proof in first order logic. It is
outlined in the following Exercise.
Exercise 1.1.10. (a) Prove there is ô ∗ ⊇ ô and A∗ ⊇ A and M∗ an ô ∗ -
expansion of M such that |ô ∗ |, |A∗ | = |A| and for each A∗ -formula
φ(v, w) with free variables from v1 , . . . , vn , w, there is an n-ary function
symbol fφ ∈ ô ∗ such that
M∗ |= ∀v (∃w φ(v, w) → φ(v, fφ (v))).
(b) Prove that if N is a ô ∗ -substructure of M∗ , then N A∗ M∗ .
(c) Prove that there is a ô ∗ -substructure N of M∗ with X ⊆ N and |N | ≤
max(|A∗ |, |X |).
Exercise 1.1.11. For ô and A define the appropriate notion of built-in Skolem
functions. Prove that for any ô and A and A-theory T there are ô ∗ ⊇ ô, A∗ ⊇ A
and T ∗ ⊇ T with |A∗ | = |T ∗ | = |A| where T ∗ has built-in Skolem functions
and any model of T has and expansion to a model of T ∗ .
Exercise 1.1.12. Let A be a fragment of L∞,ù where every formula has
finitely many free variables and let K be a PCù1 ,ù -class. Show that if M ∈ K is a
ô-structure and X ⊂ M, then there is N ≺A M such that |N | = max(|X |, |A|))
and N ∈ K.
Exercise 1.1.13. As mentioned above, in first order logic, the Upward
Löwenheim-Skolem is an easy consequence of Compactness. In infinitary
logics it is generally false. Let ô = {U, S, E, c1 , . . . , cn , . . . }, where U and S are
unary predicates, E is a binary predicate and each ci is a constant. Let φ be
the conjunction of
(i) ∀x U (x) ↔ ¬S(x),
(ii) ∀x∀y (E(x, y) → U (x) ∧ S(y)),
(iii) ci 6= cj , for i 6= j,
(iv) U (ci ) for all i,
(v) ∀y∀z ([S(y) ∧ S(z) ∧ ∀x ((E(x, y) ↔ E(x, z))] → y = z),
W∞
(vi) ∀x(U (x) → i=1 x = ci ).
Prove that every model of φ has size at most 2ℵ0 .
Elementary chains behave as they do in first order logic.

i i

i i
 i i

i i

1.2. Lù1 ,ù and omitting first order types 13

Exercise 1.1.14. Suppose A is a fragment of L∞,ù , (I, <) is a linear order

and (Mi : i ∈ I )Sis an elementary chain of ô-structures, i.e., Mi A Mj for
i < j. Let M = Mα . Then Mi A M for all i ∈ I .
Exercise 1.1.15. Give an example showing that PC classes need not be
preserved by elementary chains.
Exercise 1.1.16. Let ô be countable. Suppose
S (Mα : α < ù1 ) is a chain of
countable ô-structures such that Mâ = α<â Mα for â a limit ordinal. Let
S
M = α<ù1 Mα . Suppose φ ∈ Lù1 ,ù and M |= φ. Show that {α : Mα |= φ}
is closed unbounded.

1.2. Lù1 ,ù and omitting first order types

In Exercise 1.0.7 we noted that the class of models of a first order theory
omitting a type is expressible in Lù1 ,ù . The next result, due to Chang, shows
that any class axiomatizable by an Lù1 ,ù -sentence is the reduct of a class of
models of a first order theory omitting a set of types.
Theorem 1.2.1. Let ô be a countable vocabulary and let T be a countable
set of Lù1 ,ù -sentences. There is a countable vocabulary ô ∗ ⊇ ô, a first order
ô ∗ -theory T ∗ and a set of partial types Γ such that:
(i) if M |= T ∗ and M omits all types in Γ, then the ô-reduct of M is a model
of T ;
(ii) every model of T has an ô ∗ -expansion that is a model of T ∗ omitting all
types in Γ.
Proof. Let A be the smallest fragment containing all sentences in T . We
expand ô to ô ∗ so for each formula φ in A with free variables from v1 , . . . , vn
we have an n-ary relation symbol Rφ . T ∗ is formed by taking the following
sentences:
(i) if φ is atomic add ∀v (Rφ ↔ φ);
(ii) if φ is ¬è
V add ∀v (Rφ ↔ ¬Rè );
(iii) if φ is è∈X è add ∀v (Rφ → Rè ) for all è ∈ X and let ãφ be the type
{¬Rφ }W ∪ {Rè : è ∈ X };
(iv) if φ is è∈X è, add ∀v Rè → Rφ for all è in X and let ãφ be the type
{Rφ } ∪ {¬Rè : è ∈ X };
(v) if φ is ∃w è, then add ∀v (Rφ ↔ ∃wRè );
(vi) if φ is ∀w è, then add ∀v (Rφ ↔ ∀v∀w Rè );
(vii) for each sentence φ ∈ T , add Rφ to T ∗ .
Let Γ be the collection of types ãφ described above.5
5 As described here for each sentence φ we have added a 0-ary predicate symbol. If you are not
comfortable with this approach we could rephrase this by adding a single constant c. Then for
each sentence φ we could add unary predicate Rφ and make assertions about Rφ (c).

i i

i i
 i i

i i

14 1. Infinitary languages

Exercise 1.2.2. (a) Suppose M |= T ∗ and M omits every type in Γ. Prove

that
M |= ∀v φ ↔ Rφ
for all φ ∈ A.
Conclude that the ô-reduct of a model of T ∗ is a model of T .
(b) Prove that every M |= T has an expansion that is a model of T ∗ omitting
all types in Γ by interpreting Rφ as φ.
This completes the proof. a
We will examine a refinement of this result in Theorem 2.2.18.

i i

i i
 i i

i i

Chapter 2

BACK AND FORTH

2.1. Karp’s Theorem

We next work on several characterizations of ≡∞,ù due to Karp.

Definition 2.1.1. Let M and N be ô-structures. A partial isomorphism
system between M and N is a non-empty collection P of partial ô-embeddings
f : A → N where A ⊆ M such that:
(i) for all f ∈ P and a ∈ M there is g ∈ P such that g ⊇ f and a ∈ dom(g);
(ii) for all g ∈ P and b ∈ N there is g ∈ P such that g ⊇ f and b ∈ img(g).
We write M ∼ =p N if there is a partial isomorphism system between M and N .
Exercise 2.1.2. If M and N are countable and M ∼
=p N , show by a
back-and-forth argument that M ∼
= N.
Definition 2.1.3. Let M and N be ô-structures. The game G(M, N ) is
played as follows. At stage n, Player I plays an ∈ M or bn ∈ N . In the first
case Player II responds with bn ∈ N and in the later case Player II responds
with an ∈ M. Player II wins the play of the game if the map an 7→ bn is a
partial ô-embedding.
Theorem 2.1.4. The following are equivalent:
(i) M ≡∞,ù N ;
(ii) M ∼ =p N ;
(iii) there is a partial isomorphism system P between M and N where every
p ∈ P has finite domain;
(iv) Player II has a winning strategy in G(M, N ).
Proof. (iii) ⇒ (ii) is clear.
(ii) ⇒ (i). Let P be a system of partial isomorphisms. We prove that for all
φ(v1 , . . . , vn ) ∈ L∞,ù , f ∈ P and a1 , . . . , an ∈ dom(f), then
M |= φ(a) ⇔ N |= φ(f(a)).
We prove this by induction on formulas.
V This
W is clear for atomic formulas
and the induction step is obvious for ¬, and . Suppose φ(v) is ∃w ø(v, w).

15

i i

i i
 i i

i i

16 2. Back and forth

If M |= ø(a, c). There is g ∈ P with g ⊇ f and c ∈ dom(P). By induction,

N |= ø(f((a)), f(c)), so N |= φ(f(a)).
On the other hand, if N |= ø(f(a), d ). There is g ∈ P with g ⊇ f
and c ∈ dom(g) such that g(c) = d . By induction, M |= ø(a, c). Thus
M |= φ(a).
This proves the claim. In particular, if φ is a sentence M |= φ ⇔ N |= φ.
(iv) ⇒ (iii). Let ç be a winning strategy for Player II. Let P be the set of all
maps f(ai ) = bi where a1 , . . . , an , b1 , . . . , bn are the results of some play of the
game where at each stage Player I has played either an or bn and Player II has
responded using ç. Since ç is a winning strategy for Player II, each such f is a
partial ô-embedding. Since Player I can at any stage play any element from M
or N , P satisfies (i) and (ii) in the definition of a partial isomorphism system.
(i) ⇒ (iv). We need one fact.
Claim. Suppose (M, a) ≡∞,ù (N , b) and c ∈ M , there is d ∈ N such that

(M, a, c) ≡∞,ù (N , b, d ).

Suppose not. Then for all d ∈ N there is φd such that M |= φd (a, c) and
N |= ¬φd (b, d ). But then
^
M |= ∃v φd (a, v)
d ∈N

and
^
N 6|= ∃v φd (b, v)
d ∈N

a contradiction.
We now describe Player II’s strategy. Player II always has a play to ensure
(M, a1 , . . . , an ) ≡∞,ù (N , b1 , . . . , bn ). As long as Player II does this the
resulting map will be a partial L-embedding. a

Exercise 2.1.5. If M ≡∞,ù N , then for any countably generated substruc-

ture M0 ⊆ N there is an isomorphic N0 ⊆ N .

Exercise 2.1.6. We say that an abelian group is ℵ1 -free if every countable

subgroup is free.
Prove that if G is free abelian and H ≡∞,ù G, then H is ℵ1 -free. We will
show the converse in Corollary 2.2.9.

Exercise 2.1.7. For any abelian group A we say that a subgroup A0 is thin
if there is a subgroup A1 such that A = A0 ⊕ A1 and A1 is isomorphic to A.
Let G be the free abelian group on countably many generators. Let P be the
set of isomorphisms between thin subgroups of G and thin subgroups of Zù .

i i

i i
 i i

i i

2.1. Karp’s Theorem 17

(a) Prove that P is a partial isomorphism system. Conclude that G ≡∞,ù Zù ;

(b) The group Zù is known not to be free abelian (see for example [32] Theorem
21). Conclude that the class of free abelian groups is PCù1 ,ù but not L∞,ù -
axiomatizable. In Appendix A we show that there are 2ℵ1 non-isomorphic
ℵ1 -free abelian groups of cardinality ℵ1 .
Exercise 2.1.8. Prove that M ≡∞,ù N if and only if there is a forcing
extension of the universe V[G] where M ∼
= N.
For another view, we begin by defining a deceptively simple sequence of
equivalence relations. For each ordinal α, we will have a relation
(M, a) ∼α (N , b)
where M and N are ô-structures, a ∈ Mn and b ∈ N n and n = 0, 1, 2, . . . .
• (M, a) ∼0 (N , b) if M |= φ(a) if and only if N |= φ(b) for all atomic
ô-formulas φ.
• For all ordinals α, (M, a) ∼α+1 (N , b) if for all c ∈ M there is d ∈ N
such that (M, a, c) ∼α (N , b, d ) and for all d ∈ N there is c ∈ M such
that (M, a, c) ∼α (N , b, d ).
• For all limit ordinals â, (M, a) ∼â (N , b) if and only if (M, a) ∼α (N , b)
for all α < â.
We show that ∼α captures the more complicated notion of equivalence for
sentences of quantifier rank at most α.
Definition 2.1.9. Define qr(φ) the quantifier rank of L∞,ù -formulas as
follows:
• If φ is atomic, then qr(φ) = 0;
• qr(¬φ)
V = qr(φ); W
• qr( φ∈X φ) = qr( φ∈X φ) = supφ∈X qr(φ);
• qr(∃v φ) = qr(∀v φ) = qr(φ) + 1.
Definition 2.1.10. We say M ≡α N if and only
M |= φ ⇔ N |= φ
for any L∞,ù -sentence φ of quantifier rank at most α.
Exercise 2.1.11. If α is a limit ordinal, show that M ≡α N if and only if
M ≡â N for all â < α.
Exercise 2.1.12. If α is a limit ordinal, then M ≡α N if and only if
M ≡â N for all â < α.
Theorem 2.1.13. M ≡α N if and only if M ∼α N .
We leave the proof as an exercise.

i i

i i
 i i

i i

18 2. Back and forth

Exercise 2.1.14. Prove by induction on α that for all ô-structures M, N ,

a ∈ M and b ∈ N , (M, a) ∼α (N , b) if and only if
M |= φ(a) ⇔ N |= φ(b)
for all formulas φ(v) of quantifier rank at most α.

2.2. Scott’s Theorem

If ô is any vocabulary and M is an ô-structure we define a sequence of

L∞,ù -formulas ΦM
a,α (v) for a ∈ M
<ù
and α is an ordinal as follows:
^
ΦM
a,0 (v) = ø(v),
ø∈X

where X = {ø : M |= ø(a) and ø is atomic or the negation of an atomic

ô-formula}.
If α is a limit ordinal, then
^
ΦMa,α (v) = ΦM
a,â (v).
â<α

If α = â + 1, then
^ _
ΦM
a,α (v) = ∃w ΦM
a,b,â (v, w) ∧ ∀w ΦM
a,b,â (v, w).
b∈M b∈M

Lemma 2.2.1. Let M and N be ô-structures, a ∈ Ml , and b ∈ N l . Then,

M
(M, a) ∼α (N , b) if and only if N |= φa,α (b).
Proof. We prove this by induction on α. Because (M, a) ∼0 (N , b) if and
only if they satisfy the same atomic formulas, the lemma holds for α = 0.
Suppose that ã is a limit ordinal and the lemma is true for all α < ã. Then
(M, a) ∼ã (N , b) ⇔ (M, a) ∼α (N , b) for all α < ã
⇔ N |= ΦM
a,α (b) for all α < ã

⇔ N |= ΦM
a,ã (b).
M
Suppose that the lemma is true for α. First, suppose that N |= φa,α+1 (b).
Let c ∈ M. Because
^
N |= ∃w ΦM a,x,α (b, w),
x∈M

there is d ∈ N such that N |= ΦM a,c,α (b, d ). By induction, (M, a, c) ∼α

(N , b, d ). If d ∈ N , then because
_
N |= ∀w ΦM a,c,α (b, w)
c∈M

i i

i i
 i i

i i

2.2. Scott’s Theorem 19

there is c ∈ M such that N |= ΦM a,c,α (b, d ) and (M, a, c) ∼α (N , b, d ). Thus

(M, a) ∼α+1 (N , b).
Suppose, on the other hand, that (M, a) ∼α+1 (N , b). Suppose that c ∈ M,
then there is d ∈ N such that (M, a, c) ∼α (N , b, d ) and N |= ΦM a,c,α (b, d ).
Similarly, if d ∈ N , then there is c ∈ M such that N |= ΦM a,c,α (b, d ). Thus,
M
N |= Φa,α+1 (b), as desired. a
A similar induction shows that even more is true. The formula ΦM
a,α depends
only on the ∼α -class of (M, a).
Exercise 2.2.2. Prove that if (M, a) ∼α (N , b) if and only if ΦM
a,α (v) =
ΦN
b,α
(v)

Lemma 2.2.3. For any infinite ô-structure M, there is an ordinal α < |M |+

such that if a, b ∈ Ml and (M, a) ∼α (M, b), then (M, a) ∼â (M, b) for all â.
The least such α is called the Scott rank of M and denoted sr(M).
Proof. Let Γα = {(a, b) : a, b ∈ Ml for some l = 0, 1, . . . and (M, a) 6∼α
(M, b)}. Clearly, Γα ⊆ Γâ for α < â.
Claim 1. If Γα = Γα+1 , then Γα = Γâ for all â > α.
We prove this by induction on â. If â is a limit ordinal and the claim
holds for all ã < â, then it also holds for â. Suppose that the claim is
true for α ≤ â and we want to show that it holds for â + 1. Suppose
that (M, a) ∼â (M, b) and c ∈ M. Because (M, a) ∼α+1 (M, b), there
is d ∈ N such that (M, a, c) ∼α (M, b, d ). By our inductive assumption,
(M, a, c) ∼â (M, b, d ). Similarly, if d ∈ M, then there is c ∈ M such that
(M, a, c) ∼â (M, b, d ). Thus, (M, a) ∼â+1 (M, b) as desired.
Claim 2. There is an ordinal α < |M |+ such that Γα = Γα+1 .
Suppose not. Then, for each α < |M|+ , choose (a α , b α ) ∈ Γα+1 \ Γα .
Because Γα ⊆ Γâ for α < â, the function α 7→ (a α , b α ) is one-to-one. Because
there are only |M| finite sequences from M this is impossible. a
A more subtle notion of Scott rank is sometimes useful
Definition 2.2.4. For a ∈ M define r(a) as the least ordinal α such that
for all b
(M, a) ∼α (M, b) ⇒ (M, a) ∼α+1 (M, b).
Then
sr(M) = sup{r(a) : a ∈ M}.
We also define
SR(M) = sup{r(a) + 1 : a ∈ M}.

i i

i i
 i i

i i

20 2. Back and forth

Exercise 2.2.5. Calculate SR for the following structures.

(a) (ù, <);
(b) (ù + 1, <);
(c) (ù + ù, <);
(d) (Z, +, 0).
We will see later that at limit ordinals α there can be interesting distinctions
between SR(M) = α and SR(M) = α + 1 while in both cases sr(M) = α.
Exercise 2.2.6. Formulate and prove a version of Lemma 2.2.3 for finite
ô-structures.
Exercise 2.2.7. (a) Show that for all â < ù1 and n ∈ ù there is an Lù1 ,ù (ô)-
formula Sân (v, w) where |v| = |w| = n such that
M |= Sân (a, b) ⇔ (M, a) ∼α (N , b)
for any ô-structure M.
(b) Show that for all α < ù1 there is óα ∈ Lù1 ,ù (ô) such that M |= óα if and
only if SR(M) ≥ α.
We conclude this section with Scott’s Isomorphism Theorem
Let M be an infinite ô-structure of cardinality κ, and let α be the Scott rank
of M. Let ΦM be the sentence
∞ ^
^
ΦM ∅,α ∧ ∀v(ΦM M
a,α (v) → Φa,α+1 (v)).
l =0 a∈M l

Because all of the conjunctions and disjunctions in ΦM M

a,â are of size κ, Φa,â ∈
Lκ+ ,ù for all ordinals â < κ + . Thus ΦM is an Lκ+ ,ù -sentence. We call ΦM
the Scott sentence of M. If ô is a countable vocabulary and M is countable
ô-structure, then ΦM ∈ Lù1 ,ù .
Theorem 2.2.8 (Scott’s Isomorphism Theorem). Let M and N be ô-struct-
ures, and let ΦM ∈ L∞,ù be the Scott sentence of M. Then, N ≡∞,ù M if and
only if N |= ΦM .
In particular, if M and N are countable then φ ∈ Lù1 ,ù and, if N |= ΦM ,
then N ∼ = M.
Proof. Clearly if N ≡∞,ù M, then N |= ΦM .
For the other direction, suppose N |= ΦM . We will show that N ∼ =p M.
Let α be the Scott rank of M. Let P be the set of partial functions f with
finite domain from M to N such that if a is the domain of f and f(a) = b
then N |= ΦM M
a,α (b). Since N |= Φ , the empty map is in P so P 6= ∅.
Suppose we have f in P with domain a with f(a) = b. Let c ∈ M. Since
N |= ΦM ∧ ΦM a,α (b), N |= Φ
M
∧ ΦM M M
a,α+1 (b). But Φa,α+1 (v) → ∃w Φa,c,α (v, w).
M
Thus there is d ∈ N such that N models Φa,c,α (b, d ) and we can extend f by
sending c to d .

i i

i i
Exploring the Variety of Random
Documents with Different Content
 least some estimates claim that by the end of this century
two percent of our national population will be able to produce the
basic food and goods needed by the other ninety-eight percent. And
so, coming from the honking expressways of Chicago by way of hot
dog and soda pop world of Florida and also Atlanta, Georgia, where
God is demonstrably dead according to its resident theologian, T.
Altizer, I find myself in a tranquil atmosphere in the midst of a
campus which, at least physically, suggests an environment of
single-mindedness, of comprehensiveness, of "the old days" when
one could walk for ten minutes and see the whole world, when the
institutional Church kept the rumor of God alive, where people did
their own weaving and justice was as simple as six guns and
woodshed whippings, where everyone knew where he stood, where
Norman Rockwell rather than Pablo Picasso would have made a
decent living, where there were no Bob Dylans, no Rolling Stones,
where woman's place was in the — you guessed it — in short, where
change was slow, where positions were clear if not always in
agreement, and most important perhaps, where the structures of
existence were coherent. This is an entirely too simple view of the
past admittedly, and I do not wish to suggest that this restored
campus is either irrelevant or somehow cut off from the world. I
trust that during the next few days all of us will be thrust perhaps
against our wills into areas of thought that are so overwhelming in
their implications that we might wish to plummet swiftly back into
some womb of simplicity, some fraction of the world where we might
live in peace. Nevertheless, I regard our task, if I understand it, as
that of raising questions — raising questions not merely for the sake
of increasing confusion — but raising questions that, I would
contend, must be answered by all of us if we are even to hope for
continued existence on this small planet. This morning I shall try to
do two things — first: attempt to outline something of the challenge,
threat, and excitement of what I shall call the world metropolis. In
doing this I shall attempt to offer a vision which I believe applies to
all of us. My second task will be an attempt to look at the challenges
facing the religious community — in particular, the Church that
claims to have its basis in Jesus Christ ■ — in the context of this
world metropolis. I would like to change the title of this lecture to
"How to be double-minded and like it" or, perhaps more to the point,
"Can Batman and Robin beat the Jester.'" Nevertheless, I shall stick
with the original title, "Religious Dimensions of the New Metropolis."
In his Notes from Underground, the great Dostoevsky claims that
man will go to any lengths, even to outrageous acts, to prove to
himself that he is a man and not a piano key. His claim was made in
the context of a society which he perceived as increasingly hostile to
human self-expression. Indeed, the rise of the metropolis,
particularly in the last century and spectacularly in the last decades,
has led to a variety of reactions, all of them suggesting that man is
being degraded, is becoming a piano key, an automaton. Whether in
the paranoic half-truths of the John Birch Society, the self-conscious
railings of a Dwight McDonald at "mass culture", or even the rioting
of Watts or Rochester, regardless of other factors that may be
involved, there is a growing sense that it is impossible to speak of a
moral dimension, a dimension within which man is able to forge his
own destiny, to overcome the grey bureaucracy, to locate himself in
the world. It is not difficult to substantiate the view that the
metropolis is the focus of man's present and future existence. We
can substantiate it first at the level of the observable physical
environment. Let me quote from a justpublished report on urban
renewal and the future of the American city by the well-known Greek
city planner C. A. Doxiadis: On the basis of general data and criteria,
we foresee that the earth's present population ratio of 40 per cent
urban and 60 per cent rural is going to change to a ratio of 95.7 per
cent urban and 4.3 per cent rural in a hundred years; and that by
the end of the next century, 98 per cent of the population will be
living in urban settlements and 2 per cent will be living in minor
settlements oriented toward agricultural production. Even this 2 per
cent will have many of the characteristics of urban residents, for
production will be completely mechanized and the agricultural
settlements will serve as overhaul bases for mechanical equipment.
Futhermore, the Doxiadis report estimates that the world population
will reach virtually imimaginable proportions. In the judgment of
most population experts, it is probable that the population of the
earth, which is now approximately 3 billion people, will be by the
end of the present century somewhere between 6 and 7 billion . . .
If we take the most conservative population estimates, and if we
assume that strict policies of birth control will be implemented, the
population of the earth will be more than 12 billion a hundred years
from now. But it is not very likely that strict birth control policies will
be implemented immediately in any area of the world, so the earth's
population is hkely to reach a minimium of 25 billion people a
century from now and 50 billion people before the end of the
following century. There are even estimates that the world might
feed this many persons. When one considers the effect that the
growth of the metropolis has already had on American life, when we
consider that such issues as reapportionment and even the
guaranteed annual income are primary reflections of both the assets
and the liabilities of the metropolis, one is hard pressed to imagine
what might be the result of Doxiadis' projection. Suffice to say that
the Eighteenth century metaphysical poets were somehow able to
see in the smallest object of nature a microcosm of the whole world.
Today the scale has changed. We must look at the metropolis as it
presently is ( Chicago, for example, with almost 300 separate
municipalities in a metropolitan area, living on the basis of legal
precedents that were appropriate to a rural economy), to find even
an approximate microcosm of the world as metropolis — the world
metropolis. I should like to point to a passage of St. Paul. . . . the
body does not consist of one member, but of many. If the foot
should say, "Because I am not a hand, I do not belong to the body,"
 that would not make it any less a part of the body. And if
the ear should say, "Because I am not an eye, I do not belong to the
body," that would not make it any less a part of the body . . . If all
were a single organ, where would the body be? As it is there are
many parts, yet one body. The eye cannot say to the hand, "I have
no need of you," nor again the head to the feet, "I have no need of
you." ... If one member suffers, all suffer together; if one member is
honored, all rejoice together. (From I Cor. 12:24-26) On one level
this is an exhortation, a call to harmony and love within the Church.
On quite another level, it suggests stark, almost brutal, scientific
reality, a reality which emerges as the main characteristic of the
metropolis and as such the main characteristic of the world. John
Donne wrote, "No man is an island." Accelerating population and
technology revises this observation: "No human settlement is an
island. Indeed no nation is an island. In fact, no continent is an
island." For, inexorably, the world metropolis is an organism, a body,
which cannot function without comprehending its own
interdependence, without raising to the level of consciousness the
new necessity of self-understanding relative to the whole, a far more
complex whole than the New England village, sufficient unto itself,
represented. The world metropolis — where a disturbance in Asia
can trigger cancer in urban ghettoes five thousand miles away,
where one organ can dispose of its obsolete weapons, not by
evacuation, but by distributing them, for a price, around the globe,
where sickness anywhere daily becomes sickness everywhere — the
world metropolis becomes the only point of reference for meaning.
The moralistic impulse toward one-worldism of past decades has
now become a practical matter of life and death. And the ultimate
historical question becomes not. In a discussion group following his
lecture, Dr. Rose responds to a question put by Ann Wilson, senior
from Jacksonville, Fla. "Can our little suburb survive.-*" or "Can our
nation survive.'" but "Can the world metropolis survive?" In the light
of the emerging world metropolis, the alternative is world
anihillation. I personally find it difficult to be terribly optimistic. The
emergence of the conditions of a world metropolis suggest to me
unimaginable strains on the human psyche, strains so unimaginable
that I can conceive of a basic evolutionary leap within the human
community as the only alternative to chaos. What once seemed to
us moral causes — things that we could support at our will to win a
few merit badges — have now become the essence of survival. No
longer do we confront one another with the moral request to help
the poor, for we know that the poor in the world are now strong
enough ultimately to clobber the rich. No longer do we entreat
people to be concerned for racial justice, for we know that racial
justice deferred too long simply leads to the probability of new racial
injustices only this time the language will be reversed and our
children will have to grow used to hearing words like "nigger"
applied to the diminishing white minority. Institutionally speaking,
the adventurer in the new world metropolis has perhaps two
choices. He can become a subversive within the establishment. If he
is a good subversive he will lose no opportunity to challenge in
fundamental ways the policies by which his institution impedes the
necessary development of the world metropolis. If he is in the
institutional Church he does not hesitate to expose the investment of
his church in institutions that perpetuate war or supply war efforts.
If he is in government he does not hesitate to expose the way that
the bureaucracy subverts the intent of its own legislative mandates.
If he is a lawyer he does not hesitate to file the suits that are
needed to bring the law into some approximate harmony with the
legal requirements of the world metropolis. To these remarks on the
challenge of the new metropolis, I would add two final qualifications,
or rather two items of baggage that are needed by the venturer into
the challenging world I have sought to describe. First, he will need a
sense of politics, secondly a sense of inventiveness, of imagination.
The sense of politics is really no diflf^erent than the sense of
inventiveness . . . indeed an essential recipe for the emerging
metropolis is the wedding of the two. Such a marriage might result
in the following attributes, attributes that have been unusual in the
past: — a capacity to relate scientific discovery to political and social
implications, even to devise means by which such discoveries can be
equally distributed or, if destructive, controlled; — a capacity to
interpret ethics in the context of group relationships in an era when
it is more appropriate to fight for decent legislation than to be a
good Samaritan, at least quantitatively; — a developing capacity to
sort out those elements of life which require centralization and those
who do not. Vis a vis this point I hope that all of you will take a look
at Paul Goodman's recent book. People or Personnel: — an openness
to change, self-renewal • — understanding. Is there something to be
said that requires a Church to say it? I believe so, though in this age
the saying cannot be too final or comprehensive. We live in a time
when in Paul's phrase, we "see through a glass darkly." 6
 Perhaps the most important theological fact of our time is
that we are still here. The early Christians were convinced that the
world was doomed, that the Judgment was at hand. But the world
remains, doomed to ultimate extinction by natural causes quite apart
from any eventual intervention from beyond. It is true that the Christ
event as understood by the New Testament did not foreclose human
history, but which of Christ's apostles imagined that the wicked
world would last as long as it has? The entrance of God into history
in the person of the man Jesus, His crucifixion and resurrection, was
an extraordinary revelation of the divine nature. It revealed for all
time that the absolute nature of God is to forgive ihe man who
repents and turns to Him. But it was not the End. It was not the
completion of some celestial poker game in which the only remaining
task was to gather in the chips in the form of souls won. No, God
was and is still involved in history and Christ is comprehensible as
one whom we see through a glass darkly, sporadically, but not yet
face to face, eternally. Christ lives as an indestructible aspect of God
whose victory is foregone but not yet achieved. The Church ought
really to be the New Israel, meaning that it ought to be a people
committed to the working out of God's purposes within history. The
purpose is to serve, to evangelize by showing forth the struggle, not
to form a nice little club that revels in benefits already won and
takes refuge from the world in sanctuaries and personal testimonials.
We live between partial victory and what we trust and hope will be
total victory. The Christ event widened the struggle to include the
Gentiles. It clarified the nature of the battle. But it did not relegate
the God of the Old Testament to a far off cloud, which is where far
too many academic theologies have placed Him. We need to recover
the spirit of the prophets of Israel, the concreteness of the
Psalmists, the deep spiritual wrestling of the Patriarchs and Job. We
need to recover the sense of the God who is active in history,
contemporary, emotional, argumentative, cajoling, loving,
thundering. God is not distant! He is present, restless, discontented
with human sin, commiserating with the oppressed, fighting with
man and the Devil. God's victory is a matter of faith looking forward,
not of absolute certainty looking backward. One who would involve
himself with the God who is revealed in Jesus Christ will be sensitive
to the contemporary clash of good and evil in the real and present
world. The Church is related to God as Israel was related to Him in
the Old Testament: called to serve, given a covenant with the Lord,
and under divine mandate to do justice in the world. But for too long
the Church has acted as if the war were over, as if man were a
species of pawn. At her worst the Church has outdone Israel in
idolatry, inspiring the wrath of an Amos: "I hate, I despise your
feasts. I take no delight in your solemn assemblies." The Hebrev\'s
were less ingenious than we. They lacked adding machines. They
were unfamiliar with the intricate logic of pensions, building funds
and job security. We have learned to dedicate our primary energies
to such important questions as which comes first, the sanctuary or
the educational wing, the suave preacher or the go-go administrator,
the leadership gifts campaign or the grass roots appeal. We send
ministers out not to preach the Gospel but to 7 insure the financial
success of a foundering new Church development project. So it is
not Christianity that is irrelevant; it is its institutional expression, the
Church, mired in its moralism and ceaselessly contemplating
everything but the One who calls it into being. What, then, is the
mission of the Church.^ The common answer of many who have
spoken and written of the renewal of Protestantism is that the
Church's mission is "to be in the world." Because this phrase has
gained such wide currency it is important that we raise the question
of what being "in the world" means. For many partisans of Church
renewal, the world is "where the action is." It is where the "big "
decisions are made, where "history" is unfolding, whether before the
TV cameras or behind the closed doors of the powerful. It is where
the Church "ought to be," but most often is not. The distance
between the Chiu-ch and the world, so defined, is the distance
between dishpan evangelism and the sizzling issues of the planning
commission. To speak of being in the world, in this sense, is perhaps
to lament the era in Western history when the Church was a
temporal power. There is the haunting sense that the Church is no
longer in the center of the stage. But there are few partisans of
renewal who would advocate the seizure of political power as a
solution to the problem of irrelevance. If the Chiurch is to be in the
world today, it will arrive not as master but as servant, minus the
trappings of ecclesiastical authority. Very often the "world" is much
less glamorous than it seems to be. There is the perhaps
subconscious tendency to feel that if we could just infiltrate Madison
Avenue, General Motors, or even the White House, we would
perforce be "in the world. " But business, advertising, and politics
can be more isolated and impotent than one might During a tea
break students continue reacting to Mr. Rose's lecture as they
discuss it ivith Chaplain Clark Thompson.
 suspect. The very institution that seems crucial may prove
peripheral in terms of its actual impact on events. History, and the
making of it, is too complex to yield to easy generalizations about
where power lies or who is the crucial decision-maker. History often
makes posthumous heroes of contemporary unknowns. The
Montgomery Bus Boycott, which determined the future character of
much of the civil rights movement, had its genesis when a
courageous but certainly unprestigeous Negro lady, Mrs. Rosa Parks,
refused to give up her seat on a bus. Her refusal precipitated the
modern nonviolent movement in the United States. A local minister,
the Rev. Martin Luther King, Jr., emerged, following Mrs. Parks'
decision, as a national leader. Who could have foreseen that a
woman's stubbornness on a bus would be the focus of "the action?"
This is not to deny that there are many visible power centers in our
society that are in need of humanization and where the Church
should be present. But there are times when we fail to consider that
"the action" is much closer to home. We may run toward the world
without seeing the world at our doorstep. We shall err in our
definition of the Church's mission if we see the world only as the
stage on which the dramatic scenes of history are enacted. There is
also the world of the constant, where change is slow and where pain
and sorrow are unpublicized. The world changes, but it does not
change. As a consequence the Church must be in the world as priest
as well as prophet. If Christians support revolution without healing
the wounds of both friend and enemy, they foster the illusion that
the conflicting truth of this life can be reconciled painlessly. So while
the Church abandons herself to "the action," she must remember
also the lonely, the hung up, the sick and the dying. It is true that
Christ said, "Let the dead bury the dead." But he also said, "Thy sins
are forgiven thee. Rise up and walk." St. Paul had ample opportunity
to compare the drama of jail with the smaller dreams of weakness
and pain, and he used the word "love" to describe the style of life
needed both on the picket line and in the hospital corridor. It was
not to be an insipid, never angry, sentimental love. But it ivas to be a
love which was aware of needs and sympathetic to human
weakness. Without love, one could prophesy and be "nothing" — a
clanging cymbal. The Church loves the world when she is oriented
both to Dr. Hubbard and Dr. Martin (on floor), prior to their oivn
lectures, get the feel of their future audience as they join a
discussion group. revolution and to constant, unglamorous needs. To
be preoccupied with one and neglect the other is to lose balance:
love then becomes not love but escape. But Paul also recognized
that it was humanly impossible to combine revolutionary zeal and
priestly concern within each individual. The Church was described by
Paul as a Body whose several members performed different
functions, all under the imperative of love. Protestantism has lost
this sense of the Church. J. B. Phillips translates Paul's first letter to
the Corinthians (Chapter 12, verses 27-30) as follows: As we look at
the Body of Christ do we find that all are Special Messengers, all are
preachers, or all teachers.'' Do we find all wield ers of spiritual
power, all able to heal, all able to speak with tongues.' No, we find
God's distribution of gifts is on the same principles of harmony that
He has shown in the human body. Paul is speaking of specialization.
He is recognizing that different persons are called to different tasks.
The true unity of the Church is expressed in the term Body. Each
functioning organ in the Body is essential to the life and unity of the
whole. But one sees few such distinctions within the Protestant
Church today. We lump most of the specific functions of which Paul
speaks into the job description of the professional, ordained ministry.
The laity emerges as the severed portion of an amputated body. The
result is a Church which is neither priestly nor prophetic, a Church
which cannot minister effectively in the world. For in the last analysis
it is obvious that the Church is inextricably in the world. It cannot
escape. The Priest and the Levite were on the Jericho Road when
the wounded man cried for help. The Church was in the world when
the Inquisitor's rack was nudged to the extreme. She was there
when St. Francis spoke to animals and indigents. And she is in the
world tody — weak, divided, concerned and self-serving, helping and
hindering, worshipping God and worshipping idols, but there
nonetheless. So the question is really bow the Church is to be in the
world. How is the Church — 2000 years after the Christ event — to
organize itself? The Bible provides ample cause for discouragement
in the face of this question. At one level, the Scriptures can be seen
as the story of man's abortive attempts to capture God, to freeze
Him into an inflexible and easily handled mold. Man continually
attempts to perform surgery on God — usually of the heart or brain
— in order to avoid the passion of his restless activity in all the world
and to escape the endless depths of His mysterious presence. But
the Bible is also the story of God's response. He demonstrates the
folly of these attempted operations in a supreme act of self-
revelation culminating in the crucifixion of the God-man who
consorted with prostitutes, cast out money-changers, and suggested
the possibility of resurrection. He unleashes His Spirit in the world
and when man is seized by this Spirit he can no longer remain
content with his little efforts to confine God, to remove all the
elements of risk and joy and suffering that the Spirit opens up.
Finally one must affirm that the Church that is deaf to the Spirit is
not the Church. And with that affirmation, one rests the case for
renewal not on human plans and notions, but upon the Biblical call
for continued openness to the Will of God. If our only achievement is
to remove the cotton from the ears of the Church, that Christians
might stand ready to obey, that is satisfaction enough. 8
 Government In An Urban Age DR. ROSCOE COLEMAN
MARTIN DR. MARTIN Dr. Roscoe Coleman Martin, our speaker on
the role and problems of government in the city, was born in Silsbee,
Texas, in 1903. After receiving his A.B. and M.A. degree at the
University of Texas in 1924 and 1925, where he was Phi Beta Kappa,
he taught in the field of government at the university from 1926 to
1937 as instructor, adjunct professor, assistant professor, and
professor. In 1932 he completed his Ph.D. at the University of
Chicago. He was professor of political science at the University of
Alabama from 1937 to 1949 and head of the department of political
science at Syracuse University from 1949 to 1956. It is from
Syracuse that he came to our symposium. The numerous
publications of Dr. Martin's work include People's Party in Texas
(1933), A Budget Manual for Texas Cities (1934), The Defendant and
Criminal Justice ( 1934), From Forest to Front Page ( 1956), TV A:
The First Twenty Years (1956), Grass Roots (1957) and Water for
New York (I960 ) . At various times. Dr. Martin has held the following
positions: director of the Bureau of Municipal Research at the
University of Texas; chief research technician of the National
Resources Committee; and director of the Bureau of Public
Administration at the University of Alabama. He has been a member
of the U. S. National Committee of UNESCO, the American Political
Science Association, the National Civic Service League, and the New
York Political Science Association. Some years ago the present under
secretary for the newly established Department of Housing and
Urban Development, Mr. Robert Wood, wrote a book he titled 1400
Governments. This book illustrated the fact that metropolitan areas
have many governments within themselves by describing the
governmental paraphernalia of the New York Metropolitan Area. We
are blessed with a plethera of local governments in metropolitan
areas. We will continue, baring some catastrophe, to have these
scores of governments in the metropolitan area. So the question is
how can we improve the business of government in the metropolitan
area with the government we have. I earnestly wish, with the
purists, that we didn't have all these governments — or even that
we had only 10 per cent of them — but we have them. And because
these governments have been going on for many years and nothing
has happened to remove them, something must happen by way of
reorganization. Therefore, I have turned my interest in the direction
of proposals for the amelioration of governmental problems in
metropolis. Along what lines shall we attempt to ameliorate these
problems? Along the lines of working out more sensible relationships
among the governments we have. If you ask, how do you go about
getting rational government in metropolitan areas, a realist has to
say you can't get there from here. We are here with 90,000 local
governments, and we want to get there. There means, for our
purposes, a more inventive system of government in the
metropolitan area. I do not rule out the possibility of vast
reorganizations but is there any real prospect of progress here?
History plainly shows not much has been done in the areas of
reorganization. What then are the possibilities for improving
government in the metropolitan area? I am talking here about local
level not about government up and down. What devices can be
found to improve government at the local metropolitan area? One
device is the Council of Governments idea. Here representatives of
existing governments get together and talk over common problems.
This is a very useful device for identifying problems, for adjuration
and admonition, but this doesn't get anything very tangible done.
There are a number of these councils around in places like
Washington and Los Angeles, but they are essentially "Associations
of Alien Governments". They represent a most tentative step
forward. Even though they are one more kind of government to add
to the rest, I would not call them a step backward because they are
at least talking about their problems.
 A second approach to the rationaUzation of metropolitan
government is found in the Metropolitan Planning Agency. There are
many of these around today; I saw the other day an estimate of
125. They are often found in twin cities like Minneapolis-St. Paul.
The Metropolitan Planning Agency is limited pretty much to drawing
pictures on paper; that is, making blueprints. This is a good thing to
do; it employs young architects. But, again, it doesn't achieve many
real results. This is because the Metropolitan Planning Agency is not
related to any important decisionmaking force in the metropolitan
area. That's not surprising, seeing there is no decision-making force.
So the Metropolitan Planning Agencies make the plans for
metropolitan areas, but their magnificent specifications are not
recorded to have made much difference in the governmental
situation. One day maybe they will, but this project doesn't round off
many rough edges at the moment. Next there is the Inter-local
Contract, which is a means by which one local agency provides
service for another. There are thousands of such contracts around
today. In the metropolitan county of Los Angeles there are hundreds
of such contracts. Indeed, there are cities of 25-50 thousand people
in Los Angeles County which, though incorporated, have no
employees. They have a small city hall with a shingle out front, but
no employees. They buy all their services from Los Angeles County.
An ingenious arrangement, it gets the job done and keeps a cover
on the multiplication of services and the multiplication of agencies.
Frequently, you will find Special Districts created to handle
metropolitan problems. As an example, the city of Seattle is built
around beautiful Lake Washington. Sewage was deposited by Seattle
and other cities around the lake until eventually the lake was
murdered — for you can murder a body of water. Portions of it
became an eye-sore and an offense to the nose for all. In other
portions people went swimming, and one summer 600 of them
suddenly became ill. This kind of thing compels cities to action. So
Hai'ing lectured, Dr. Martin becomes the willing target of questions
from students and faculty. Seattle and its neighboring cities acted
and set up a metropolitan sewage district. It was a very good thing,
thirty years late, but a very good thing. Sometimes a function of
local government can be transferred to a larger unit of government.
The big city often lies near the center of the county so that the
metropolitan area is bounded pretty well in terms of county lines.
Sometimes one of the functions of the city which is also a function of
the county will be transferred to the larger government body.
Sometimes, however, you have quite a different problem: there are
five counties involved in the Atlanta metropolitan area, so there is no
larger unit to shift functions to. But this Transfer of Functions to a
larger entity is simplest device for the amelioration of metropolitan
government. When you use it, you have gradually emerging a
metropolitan county. There are three such governmental units in the
United States : Dade County, Florida, Birmingham and Nashville in
Davidson County, Tennessee. This is not a very good batting average
considering that my friends and co-workers over the past years must
have recommended this device for at least 200 metropolitan
developments. Finally, among the devices for ameliorating
metropolitan problems there is the Regional Agency. Let me take just
a moment to remind you that there is developing before our very
eyes a predisposition toward the Regional Agency. Let me give you
an example, the Appalachia Agency, the newly established regional
approach to ameliorate our economic problems. The emphasis by
the government on River Basins to protect our water resources is
another. Let me describe one Regional Agency. Five years ago there
was established the Delaware River Commission. While the Delaware
River is a very small river by comparison with giants like the
Colorado, the Columbia and the Missouri it nevertheless flows
through the industrial heartland of the northeast. It drains an area
where millions of people live and work. A lot of cities are dependent
on the Delaware for water and a lot of cities dump sewage into the
Delaware. The Delaware flows through four states. No one of these
states is competent to deal with the problems of the Delaware. So a
regional government was established to deal with the problem. It
was established by agreement between the four states and the
national government. In time to come we shall hear more, a good
deal more, about regional agreements as a solution to metropolitan
and regional problems. This is a very significant device. It is
significant because it reflects one of the very imaginative approaches
to the problems of government in metropolitan areas. It is significant
because it can deal with problems which cannot be resolved by
individual local government. I have presented several devices for the
amelioration of metropolitan problems. Now let us ask, do these
approaches, either singly or all together, promise to solve
metropolitan problems. No. The metropolitan problem has no easy
set of solutions. What we can do is ameliorate the problem, and we
can do it by joint action among local governments. If there were an
individual or agency which had carte blanch authority to deal with
the metropolitan problem, it would scarcely have developed in the
first place. The metropolitan problem exists in part as the price of
oiu: insistence on pluralism. The metropolitan problem could be
simplified providing we were willing to pay the price of centralization.
Really, at the heart of the problem 10
 is the American preference for pluralism, the American
preference for pandemonium. We want to do it the hard way. Why
can't we eliminate all these local governments? My answer is that I
am not siae that this would be a good thing. I am sure that this
would be next to impossible. We are making some progress in
metropolitan government. It is slow and torturous. So far we have
been talking about local governments. Let's look now at the relations
among governments at various levels. When a city finds that it is no
longer able to handle its problems what may it do? What can the
mayor or city manager do? What can he turn to? For one thing he
can look across and see what other cities are doing. He may get an
idea for interlocal action. For another thing he can look up. The first
thing he will see is the state capital. He can seek help there. What
kind of help will he get? Sometimes none. This is the kind of help
the mayor of Chicago gets when he goes to Springfield or the mayor
of New York City gets when he goes up to Albany. He meets there a
stone wall of indifference to the metropolitan problem. State
governments are oriented to a rural past, an agrarian life, a nostalgic
time when things were lovely and there were no real problems.
There never was such a time, but this is our way of shutting off the
problems of today, and tomorrow. The city just does not get much
help from the state government, for the state just plain doesn't love
the cities. Nothing would please Colorado more than to have Denver
dematerialize into thin air; and nothing, I suspect, would please
Denver more than to be able to remove itself by vote to nearly any
other place. I'm speaking facetiously, but I am expressing my deep
conviction that the cities don't get much help of a material or
emotional value from the state. The newly elected mayor of New
York City, when flatly refused by Albany, made the statement, "I'll
have to go to Washington." Mayor Lindsey reached a conclusion to
which mayor after mayor has been forced: to seek help from the
federal government. Increasingly, they have done this since 1933.
So, this isn't a new thing. If that date seems to call for some
explanation, may I remind you that that date marked the end of a
laissez-faire era in this country. It marked the end of negative
national government, and it marked the beginning of a concerned
and involved national government. In the background we have a
new federal system evolving. But urgent problems require us to
concentrate on foreground things. The significance of the
modification of the federal system is secondary. Primary in this
framework is the significance of the national government as
alleviator of these metropolitan problems. The federal government
makes great contributions ( 1 ) by providing expert assistance which
a city might not be able to locate or afford, { 2 ) by identifying
problems which may be insidious, ( 3 ) by providing standards of
municipal performance, (4) and, most of all, by providing financial
assistance. There, in this last step, is where the cities' abilities break
down. Here, too, is where the states break down. The national
government has access to sources of funds that neither states nor
cities have. After all, the most important sources of funds are
national in nature and not local. This involvement of the federal
government in the metropolitan problem has the effect of
nationalizing local problems, of elevating them to a national status,
of making the whole country take a look at the problems. Water
preservation in New York City forces every city to look ahead to the
time when even citizens who have supposed they were located in
water-rich parts of the country will have to control, safeguard and
regulate their use of water. Participation by the federal government
in making proposals for joining hands with local government is a
great hope for ameliorating the metropolitan problem. What can be
said in conclusion about governing the metropolis? It is an
extraordinarily complex problem. It is a problem of great variety. It is
a problem of a longtime trend in the growth and movement of
American population. A city manager said the other day, "I know
what I would do. I would limit the growth of the city." To do this,
you would have to limit the freedom of the American people to
move. We may come to this, but we haven't yet. It is extraordinarily
difficult to devise a system of government that governs and does not
inhibit freedom. Some things can be done at a local level and some
by government up and down at all levels. It isn't easy. Movements
toward solutions must be incremental. They must be evolutionary
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookname.com