Genera Mixta: Conceptual Blending and Mixed Genres in

Ulysses
Michael Sinding

New Literary History, Volume 36, Number 4, Autumn 2005, pp. 589-619
(Article)

Published by Johns Hopkins University Press

DOI: https://doi.org/10.1353/nlh.2006.0009

For additional information about this article

https://muse.jhu.edu/article/193114

[128.103.147.138] Project MUSE (2024-09-19 16:34 GMT) Harvard Library

 Genera Mixta: Conceptual Blending and
Mixed Genres in Ulysses*
Michael Sinding

G
eneric mixture interests critics of every stripe because it opens a
window on the processes that define literary history. Alastair
Fowler sketches a taxonomy of forms of generic change, distin-
guishing transformation and combination; Rosalie Colie analyzes the
mixture of forms in Renaissance literature; Tzvetan Todorov argues that
new genres arise out of combinations of old genres; Mikhail Bakhtin
considers so-called “secondary genres,” including the literary ones, to be
formed by the combination and transformation of the “primary genres”
of everyday speech and writing; and Jacques Derrida and others portray
deviation and mixture as rule rather than exception.1 I am persuaded of
David Duff’s claim, in his recent anthology of modern genre theory, that
to uncover the nature and principles of genre combination and change
is the central problem of genre theory today.2 The point has been
reiterated by Ralph Cohen, as editor of recent special issues of New
Literary History on “Theorizing Genres,” which he sees as pointing
[128.103.147.138] Project MUSE (2024-09-19 16:34 GMT) Harvard Library

“toward a generic reconstitution of literary study” through “the reexami-

nation of the nature, function, and significance of generic combina-
tions.”3 However, lacking analytic tools and vocabulary, studies of
specific mixtures are often rather helplessly casual.
A. Walton Litz describes the reader’s encounter with James Joyce’s
Ulysses in a way that can stand as an epitome of the encounter with mixed
genres: “We bring to Ulysses all the schemata available to us, both from
Joyce’s earlier works and from the literary tradition, since we know that
without such schemata we cannot begin to see or understand; but we
should also be prepared to modify and rearrange our preconceptions as
we go along.”4 Litz’s account of Ulysses here draws on E. H. Gombrich’s

* I thank the Social Sciences and Humanities Research Council of Canada for supporting
this research through a postdoctoral fellowship. I am grateful to Margaret Freeman and
Mark Turner for discussion of blending issues, and to Matt Laufer for organizing the panel
“Generic Interludes: Formal Mixture in Modern Literature” at the 2003 Central New York
Conference on Language and Literature in Cortland, New York, where I presented an
earlier version of this paper.

New Literary History, 2005, 36: 589–619

590 new literary history

use of the notion of “schemas” to analyze artistic perception and

interpretation. The notion is well attested in cognitive psychology and
linguistics. Schemas represent the simplified abstract elements and
relations that constitute concepts. I will be speaking of “frames,” which
are schematic models of situations, involving settings, roles, relations
among roles, and participants. Schemas for events and actions are known
as “scripts.” In particular, I will consider frames for rhetorical situations,
real and fictional, since they are a crucial dimension of genres. Gombrich’s
development of the schema idea in art theory has been immensely
productive for genre studies.5 What is still wanting, however, to really
make this investment pay is a model of the modification and rearrange-
ment of schemata Litz describes.
Over the past decade, just such a theory has evolved in cognitive
science and linguistics. Mark Turner and Gilles Fauconnier’s theory of
“conceptual blending” has roots in schema theory. Its express aim is to
explain the construction of meaning in terms of conceptual networks
created by the activation, interconnection, and manipulation of “mental
spaces” structured by schematic models. Bringing this theory to Ulysses,
we can see how Joyce’s remarkable play with genres, far from “denying
their validity” as Litz argues, instead opens wide that window on literary
evolution and history by exemplifying certain pervasive principles of
generic thought.6

Schemas

Schemas are skeletal frameworks of relations with “slots” for various

“elements.” The elements are variable, and particular examples of some
concept or category “fill in” the schema in various specific ways. There
are compulsory, default, and optional settings for the schema’s elements
and relations. To take a standard example, our schema for a “restaurant”
situation tells us that a restaurant is some defined space where people go
to buy meals and eat them. There are several places for people in groups
to eat, and there is a place where the food is made. There are prices for
the food, and the people who eat the food pay those prices. These are
compulsory settings: an entity that does not fit them is not a restaurant.
By default, the places to eat include tables and chairs and table settings,
and there is a menu with a range of choices for several courses. There
are certain expectations about the order and manner in which meals are
eaten. However, different kinds of restaurant have different defaults,
and there is much room for variation through the optional settings. The
restaurant schema does not specify a particular arrangement of tables
and chairs, décor, or type or quality of food. Some restaurants are
mixed genres in ulysses 591

unique, others belong to chains. Unlike prototypical restaurants, fast-

food restaurants typically have no waiters, and customers pay before
eating, seat themselves, and do not tip. Some have drive-through
options. Upscale restaurants may have an extensive wine list, exotic and
expensive dishes, a fixed daily menu, a small number of tables and
sittings, learned and fastidious waiters, and so on. Yet we still recognize
all this variety as within the possible meanings of “restaurant” and its
parts, “meal,” “menu,” and so on. Schemas for events are “scripts,” and
we have a script for going to a restaurant that includes entering, greeting
a host, waiting to be assigned a table, sitting down, ordering meals,
eating, requesting the bill, paying, and so on. Our schematic knowledge
also includes the range of familiar alternatives to such frames and scripts.
Much of our conceptual knowledge is structured in terms of such
schemas, and we draw on them continuously to understand language
and other forms of meaning, as well as to act in everyday life. Schematic
relations help define the meaning of schema parts and the words that
designate them. So the meaning of “menu” can only be grasped in terms
of its role in the restaurant schema. Schemas exist at many levels of
abstraction, and can embed. We have schemas for menus and for soups,
and for the general idea of “getting something to eat,” as well as for
restaurants.7

Genres as Schemas

Litz’s genre schemas are what E. D. Hirsch Jr. calls “extrinsic”

genres—an “a priori notion of what a ‘novel’ should be, that one brings
to the work of art”:8 genres as mental templates, in short, used by
authors and readers alike. (However, those templates are typically
acquired through exposure to a set of prototypical examples.) It seems
clear that many, if not all, genres are significantly defined by schemas for
the many possible dimensions of features that make up what Fowler calls
the “generic repertoire”: rhetorical situation, plot, character, and pat-
terns of rhetoric and form. These schemas are somehow linked together
in relatively coherent ways to form relatively stable literary genres. Such
links may be purely conventional associations; more likely they are
motivated by metaphor, metonymy, iconicity, and so forth.
Consider how the two main types of schema, scripts (for actions and
events) and frames (for situations), enter into genre structure. I believe
schemas for plot largely define most literary genres and subgenres. Plot
schemas convey a “sense of the whole” and help to motivate and to hold
together schemas for the other genre features. A certain kind of story
involves certain settings, actions, character types, and moods, and
592 new literary history

motivates a certain kind of narration, language, imagery, and so on.

Thus the epic, film noir, and even supposedly nonnarrative genres like
the sonnet are all defined by the kind of action they involve, which
motivates many other features. Broader categories like the novel include
more specific subgenres defined by plot schemas, like the campus novel,
the factory novel, the romance, and the thriller. (I use “plot schema”
instead of “plot script” to avoid confusion with the usual literary-
dramatic sense of “script.”) Schemas for rhetorical situations also partly
define genres insofar as they also provide templates essential for
creating and interpreting the discourse. Frames for rhetorical situations
define gross distinctions among broad generic categories (for example,
the traditional triad of drama, epic, and lyric), as well as contributing to
more fine-grained distinctions (for instance, the epistolary novel, the
blazon, the hymn, and so on).9
To characterize specific genres (literary and nonliterary), Gerard
Steen offers a valuable analysis that recognizes multiple dimensions in
an overall schema, each of which has in turn its own schematic structure.
The dimensions include (communicative) function, mode, medium,
and channel; (discursive) code, type, form, subgenre, language; and
(topic) domain and content.10 There are profound benefits to this kind
of cognitive-schematic approach to genre theory. At the level of features,
it can account for the sense of a broad repertoire of optional features of
different kinds, without sacrificing the sense that some are more
important than others and that they often cohere to form an ordered
whole, not an unstructured list. Moving to the level of works, this
translates into a capacity to account for the loose “family resemblances”
among members of genre classes without sacrificing the sense that some
works are more central to the genre than others. In addition, of course,
its view of categories is backed by research in several scientific fields.
Given this explanatory power and elegance, there is no turning back for
genre theory.

Conceptual Blending Theory

Conceptual blending is a general cognitive process first posited to

solve certain problems in the theory of metaphor and related areas of
linguistics. It addresses, among other things, imaginative constructs that
look like straightforward metaphor but actually turn out to be much
more complex. According to the “conceptual theory of metaphor”
developed by George Lakoff and Mark Johnson, a metaphor projects a
“source” concept onto a “target” concept: language, imagery, and
inferential structure are systematically mapped across conceptual do-
mixed genres in ulysses 593

mains. This explains the systematicity of metaphoric language and

thought. In the “ship of state” metaphor, the state leader is the captain;
the governing party is the crew; the people are the passengers; the seas
are the political environment; rough seas are political trouble; calm seas
are political stability; the progress of the state is the forward motion of
the ship; and so on. This mapping structures the way we speak, imagine,
and reason about this topic. Metaphor is taken to involve one-way
projection, between exactly two concepts, and as a rule from the more
concrete and well-defined concept to the more abstract and subjective.11
Blending theory focuses on examples of metaphoric projection involv-
ing more than two concepts, and/or bidirectional projection (where
both concepts contribute to a new construct), and/or a reversal of the
usual concrete-to-abstract direction. Thus the ship-of-state metaphor
can enter into a blend when it gets combined with other metaphors: for
instance, the ship gets tossed from political Left to political Right; it may
start to sink in seas of debt. The specific targets of this metaphor may enter
into the blend, so that the captain is George W. Bush and his first mate
is Dick Cheney. A blend can project from abstract to concrete: a divisive
issue might split the ship of state in two. And blends can integrate details in
“impossible” ways: splitting the ship in two need not sink it. However,
blends need not be metaphoric: I can imagine a debate with Immanuel
Kant, where Kant tells me this, I counter with that, he replies thus. This
is a “counterfactual” scenario integrating roles and participants (that is,
other role-values) from contexts widely separated in space and time,
under one frame—that of “debate.” Here situations are linked up to do
conceptual work, but without using metaphor. Blending has been
studied in verbal, musical, and visual art; in advertising, cartoons, and
allegory; in counterfactual, mathematical, and scientific thought; in
word-formation, grammar, and rhetoric; and in action, ritual, and
everyday cultural artifacts like clocks and money. I hope its terms may
supply some of the needed analytic traction to study genre mixture.12
We may illustrate the three main processes of blend-construction—
composition, completion, and elaboration—with a relevant example.
One of the most basic facts about Ulysses is that it blends the realistic
story of the Dublin characters with Homer’s Odyssey. This blend projects
one story on to another in partial, selective, and transforming ways, such
that Leopold Bloom “is” Odysseus, Molly Bloom “is” Penelope, Stephen
Dedalus “is” Telemachus, and the other characters and events of this
Dublin day parallel the characters and events of The Odyssey. As is typical
of blends, the correspondences are not complete, and there is interfer-
ence among the various sets of parallels. Let us look closely at blending
processes in the “Circe” episode.13 Blending theory speaks of “input
spaces” instead of “source” and “target” concepts. For our present
 594 new literary history

purposes, the inputs to the blend are The Odyssey story and the Dublin
story.
Composition is the selective projection of structure from the input
spaces into the blend space. Composition is guided by “counterpart
connections” between the inputs. Counterpart connections get estab-
lished in virtue of abstract structure shared by the inputs, which is said to
reside in a “generic space.” Connections may be metaphorical, as in
ship-of-state blends; they may be based on frame structure, as when Kant
and the professor share “thinker” roles. In “Circe,” Bloom is matched to
his counterpart, Odysseus, Bella Cohen is matched to her counterpart,
Circe, and other events in the brothel are matched to the events of book
10 of The Odyssey.
The connections here do not fall easily into the class of metaphor,
analogy, or role-participant relations. Blending research shows that
those classes are points along a spectrum rather than definite discrete
categories that exhaust the possible options.14 The mapping across
stories invites a label of parable or allegory, but those are cases where
specific stories are mapped to abstract themes or named concepts,
whereas the parallel links two specific stories in detail. Analogy, on the
other hand, is generally a mapping of abstract relational structure—that
is, role-role connections are more analogical than role-participant
connections. But Ulysses maps both abstract relations and specific details
between stories. Bloom is “a modern Odysseus” in both senses. He takes
some Odysseus roles in his story. He takes the literary role of “hero” in
the sense that he is a main protagonist in the story. He shares more
[128.103.147.138] Project MUSE (2024-09-19 16:34 GMT) Harvard Library

specific story roles too: social-relations roles, in that he is a husband and

father; and action roles, in that he wanders from home, encounters
challenges, helps others, and returns victorious in some sense. He is a
hero in this “victorious” sense as well. His character and his story also
echo Odysseus’s character and story in descriptive details of events and
images. Is this mapping metaphorical? In metaphoric blends like the
ship of state, counterparts are “fused” into single elements: the ship is
the state, the captain is the president. By contrast, in nonmetaphorical
counterfactuals like the hypothetical debate with Kant, both counter-
parts are projected into the blend. There are two distinct instantiations
of a common “thinker” role, and they are not fused but projected
separately for purposes of interaction and comparison. Again, Ulysses
falls between these types, or, more to the point, achieves qualities of
both. Odysseus is not projected as a distinct element into the blend, so
it seems he is fused with Bloom. But he is not fused in the way that the
“president” role, or a specific president, is fused with the role of ship’s
captain. The Ulysses blend is not only metaphorical, where Bloom is the
Odysseus of Dublin; it is an implied comparison too, where the distance
mixed genres in ulysses 595

between kinds of worlds and heroes makes it a “mock epic.” Both epic
and realistic story are used as templates, providing structure at abstract
and specific levels, all the way down from large-scale events and themes
to details of description. This combination of abstract and specific
structure, metaphorical fusion and interactive comparison, seems typi-
cal of self-conscious literary uses of myth.
Joyce’s rearrangement of Homer’s episodes illustrates a clash in the
inputs between episode/action sequences and, therefore, causal structure.
In Homer, Odysseus’s encounter with Circe leads to the next episode
when she gives Odysseus directions for visiting Tiresias in Hades to learn
how to return home. In Ulysses, the “Circe” and “Hades” episodes occur
in reverse order, are separated by eight other episodes, and are not
causally related. Because Circe/Bella is a prostitute, Bloom’s meeting
with her must happen at night. But because funerals are held during the
day, if Joyce is to retain both his consistent diurnal chronological
scheme and his equation of the Odyssey necyia (underworld visit) with the
Dublin funeral, the Hades/funeral scene must precede the Circe/
brothel encounter.
Turning to composition of events in this blend, we find metaphorical
connections between counterparts. The magical transformation of men
into animals by the witch Circe is matched as a metaphorical source
domain with the sexual debauchery of men in Bella’s brothel. But the
details of composition show a number of further conflicts in matching
between inputs. Most importantly, Odysseus’s men get transformed, and
he does not, whereas the opposite is true in Dublin: Bloom gets
transformed into a pig, while the other patrons’ metamorphoses are
subtler—he sees “MALE BRUTES” and a man “with an ape’s gait.”15
Second, Joyce’s Bella acquires strong animal qualities (for example, her
foot is seen as a hoof) and, hence, appears to undergo metamorphosis
herself, unlike Homer’s Circe. Third, although Bloom has a talisman
against Bella’s power (the potato “is” the herb, moly, that Hermes gives
to Odysseus), what frees him from her power is not the potato-talisman
but the fact that he has recently vented his sexual desire by masturbat-
ing. Fourth, unlike Odysseus, Bloom has no crew to save from this Circe.
However, fifth, Joyce gives Bloom someone to save, filling that role
through further composition: he puts the drunk Stephen/Telemachus
in the brothel, though Telemachus is no part of Homer’s “Circe”
episode. Bloom must save Stephen from the consequences of his own
drunkenness. Stephen is not much animalized. His speech is full of
animal references, like the others, but he has just a few quick quasi
transformations. He is somewhat out of control due to drink, hunger,
and grief, and this degradation is perhaps analogous to Bloom’s sexual
beast-metamorphosis.16 Bloom defeats Bella’s attempt to swindle Stephen
596 new literary history

by taking charge of his money and protects him from the worst brunt of
violence from the English soldiers Stephen enrages.
Completion provides additional structure to the basic connections
created through composition by drawing on known conceptual frames.17
Such frames can apply in the blend to elements of either input, or to
some connection among them implied by composition. They are not
contained in the original inputs; they can come from any domain. The
debate frame can only exist in the blend that brings the two solitary
thinkers together and puts them in the role of intellectual opponents.
The most obvious and significant frame recruited for this scene is that
of dream or hallucination (Joyce calls his technic hallucination, but
since the time is midnight it is natural to link hallucination to dream-
ing). This frame provides a natural context for the metamorphoses and
other fantasies and helps to guide them—for example, it tends to root
them in subconscious concerns of the characters, and it tends to limit
their visibility to individual characters. Some other significant examples
are Joyce’s discovery and application of realistic frames to complete the
projection of elements from The Odyssey. For example, Joyce frames
Bloom’s transformation into a pig as a human scenario of sadomasoch-
istic degradation. There are examples of completion using frames
common to both inputs. Both Bloom and Odysseus are fathers, but
whereas the father-son frame applies literally to Odysseus and Telemachus,
the Bloom-Stephen relation fits the frame metaphorically, as empha-
sized in Bloom’s vision of his dead son Rudy appearing over Stephen (U
497). And the same relation can be framed in different ways. As we have
noted, in “Circe” the Bloom-Stephen relation is also framed as the
Odysseus-crew relation.
Elaboration “develops the blend through imaginative mental simula-
tion according to the principles and internal logic of the blend.”18 The
dream/hallucination frame allows for very free elaboration. Thus the
gender role-reversal which is part of Bloom’s sadomasochistic fantasy
frame is elaborated into literal gender reassignments: he becomes a
female whore and Bella becomes Bello the cigar-chomping masculine
cliché. Elaboration includes further completion, as new frames are
recruited to the blend. Thus, building on the Bloom-swine connection,
Joyce draws on other frames defining relations between humans and
animals, to have Bello threaten to slaughter, bake, and eat Bloom (U
434), and then put him up for sale as a farm animal (U 440).
Note that emergent conceptual structure, not existing in any of the
inputs alone, can be created in the blend through each of these
processes:
mixed genres in ulysses 597

• Composition creates new relations between elements. For example, there

are profound contrasts as well as parallels between many elements of the Odyssey
story and the Dublin story.
• Completion creates new role relations through new framing. For ex-
ample, the dream frame turns much of the fantasy into subconscious projections
from characters. Hence we infer that, as in dreams, the fantasy is disturbing but
not dangerous, and that characters are not aware they are dreaming.
• Elaboration creates new actions and events (that bring further relations,
inferences, emotions, and so on). For example, running the dream script leads
Joyce to bring both Bloom’s and Stephen’s dead parents into the scene to
interact with them.

The basic blend is diagrammed in figure 1. The diagram has the

conventional format of blending analyses, where the input spaces
presenting the details of their frames are set side-by-side, surmounted by
the generic space that contains the structure common to both, and
standing over the blended space that presents the new construct built
from the inputs. Figure 2 shows how the basic blend is elaborated
through mapping specific elements of the fictional story episode of
book 10 of The Odyssey to those of Joyce’s Dublin story. Figure 3 shows
how the basic blend with its initial input spaces links to its later
elaboration through these “subspaces” built out of the broader initial
spaces.

Challenges for Genre Blending Analyses

Studies of blending in literature rarely attempt to analyze textual

wholes (even of short texts), never mind the larger structures of genre.
It can be difficult to determine which genres are in play, and which
features of a text to associate with which genres. It seems to me that the
parodic blend of The Odyssey with the shabby Dublin story is part of
Joyce’s genre blending. The Odyssey is a (if not the) prototypical epic, and
Joyce draws on many of the features that make it epic. Although Ulysses
does not have a literal supernatural background or deal with the deeds
of a traditional or historical hero of high station, it is a long poetic
narrative profoundly concerned with nation, history, philosophy, my-
thology, and cosmology. The Dublin story, meanwhile, has many of the
features characteristic of the novel: its protagonists and their actions are
domestic and “low” (or middle class) rather than romantic and “high,”
it focuses on presenting multiple unresolved viewpoints of individual
consciousnesses, its style is generally realistic, and it is polyphonic in
voice. On this view, a great many elements of the text can be seen as part
of this genre-blend, from large-scale narrative structure, to the stylistic
598 new literary history

evocation of the history of prose narrative from saga to novel in the

“Oxen of the Sun” chapter, down to those parody-Homeric compound
epithets (presumably following the form of “winedark sea,” such as
“oakpale hair,” “snotgreen sea” [U 4], and so on). Blending the epic with
the novelistic story creates a large-scale emergent structure: as we have
seen, the work as a whole becomes ironic “mock-heroic” (and, as the
comparison evolves, “an ennoblement of the mock-heroic”).19 That
genre is emergent for this work, but it is not historically new, of course.
There is a long tradition of the novel understanding itself as mock-epic.
Indeed, Henry Fielding’s definition of his prototypical novel Joseph
Andrews as a “comic epic in prose” suggests that the novel (at least the
“comic” strain of it) arose through the blending of epic with traditional
lower forms like comedy and prose romance.20
Such judgments follow from focusing specifically on the blending of
genre elements, which means those rhetorical and formal features
clearly characteristic of certain genres. Of course there certainly seem to
be many kinds of blending in Ulysses that are independent of genre
blending in that they do not in themselves draw on genres as input
spaces. For example, the complex metaphoric blends in phrases like the
“heaventree of stars hung with humid nightblue fruit” (U 573), and
“Circe”’s many dream-figures of speaking animals and things, as well as
more involved monstrous mixtures like the hobgoblin and the End of
the World octopus (U 413).21

Genre Blending Along Other Dimensions:

Blending Rhetorical Situations

So far we have dealt chiefly with blending of characters and story

events, which, as instantiations of types, are important determinants of
genre. Let us turn now to blending in another important determinant of
genres—rhetorical situation, or what Northrop Frye calls “radical of
presentation” and other critics call enunciatory situation or stance.22
This is much clearer in the second half of the book, and especially in the
“Circe” chapter. As Michael Groden points out, the last nine of eighteen
chapters “transfer concern from character to technique”—what Arnold
Goldman calls the “drama of alternatives” of telling a story (UP 14).
Note that the two dimensions of genre are often linked. The links may
be relatively conventional: what makes “tragic drama” tragic has no
necessary association with the stage. Or they may be more fully moti-
vated. For example, the elegy tends to have a content, tone, and length
motivated by its original situation. A form read at funerals naturally
concerns death or some deceased, especially his or her relation to the
 mixed genres in ulysses 599

mourners. I will return to how “Circe”’s blended rhetorical situation

affects the story it tells.
The epic does have a distinct rhetorical situation. The epic bard or
rhapsode (professional reciter) performs alone to a public audience at
banquets or ritual events. He sings (or perhaps recites) the poem from
memory, improvising variations on an established narrative structure
familiar to the audience, using recurring formulaic epithets, lines, and
scene types as mnemonic aids.23 While this is marginally evoked by Joyce
in Buck Mulligan’s Homeric enthusiasms (he calls Stephen “bard,”
encourages “Hellenizing” Ireland, and so on), the actual dominant
rhetorical situation of the book as a whole is much more novelistic than
epic. It is prototypically read individually and silently, and it invites such
treatment. No one expects “Circe” to be performed—though it is read
aloud along with the rest of the text on Bloomsday celebrations, and
film and stage adaptations do perform it. Aspects of fictional rhetorical
situation(s), which also bear on judgments about genre, are more
novelistic than epic. Ulysses does not have a single bardlike narrator or
narrating voice that presents human, natural, and divine events from
the same omniscient point of view and in the same august and
authoritative tone, though Joyce clearly thinks of himself as continuing
the epic tradition, in terms of its large ambition, scope, and cultural
authority. Rather, it relies heavily on the first-person voice to render the
first-person point of view and develops the technique of interior
monologue to represent the stream of consciousness of a multiplicity of
focalizers. The implied author of Ulysses also plays with fictional rhetori-
[128.103.147.138] Project MUSE (2024-09-19 16:34 GMT) Harvard Library

cal situations (for example, newspaper, catechism, and so on, as well as

drama) in a way licensed by the novel but not by the epic.
The “Circe” chapter, however, represents two important aspects of
rhetorical situation blending. First, it illustrates how conventions of
dramatic presentation are blended with conventions of novelistic pre-
sentation to produce new conventions for the mixed-genre “drama-in-
novel.” Second, it illustrates how those emergent conventions can be
further blended with another fictional presentational situation, one that
is not a genre but has enough structure in common with genres to be
blended with them, to define more fully the chapter’s content, struc-
ture, and themes. We can treat this as a two-stage process: first, the basic
schemas of the rhetorical situations of the genres “drama” and “novel”
are blended to create a new form that coherently integrates key
elements of each, and the surface symbolic forms used to represent
them. Then a “dreaming” frame is connected to the “drama” frame,
which in turn allows the reader to see analogies to the rhetorical
situation and form of the “interior monologue” technique already
developed in the realistic story.
600 new literary history

Basic Blending for a Drama Embedded in a Novel

I expect this part of the analysis to apply to most (if not all) blends of
drama-in-novel. We start with the prototypical frames for the rhetorical
situations of drama and novel, and their textual marks.

The Novel Frame

I will regard the rhetorical situation of the “novel” frame as defined by

part of what Joyce called his “initial style”: that is, “a combination of
third-person, past-tense narration and direct first-person, present-tense
depiction of the characters’ thoughts” (UP 15). Groden observes that
Ulysses “to a large extent . . . is a novel in the traditional sense: it features
three major characters, Leopold and Molly Bloom and Stephen Dedalus.
. . . [Its interior monologue creates] the illusion of verisimilitude . . . ; it
contains several developing themes and conflicts, including the move-
ment of Stephen and Bloom toward each other in a spiritual father-son
relationship, Bloom’s need to deal with Molly’s adultery . . . , and
Stephen’s anguish over his mother’s recent death and his refusal to obey
her dying wish that he pray for her” (UP 13). This content could very
well be staged, but the third-person, past-tense narration defines the
novelistic side by defining the fictional rhetorical situation of the
traditional novel: the actual author takes the fictional stance of an
implied author communicating to the reader (perhaps to a certain
implied reader) in the present his or her (omniscient) observation of
events that occurred in the past.24 The interior monologue is the odd
feature out here. We will return to it.

The Drama Frame

We can take drama as defined by the conventions of stage directions

indicating the actions of characters; and character names introducing
the lines of dialogue they are to speak. The fictional rhetorical situation
here is that the audience is directly observing the action actually
occurring here and now.
Figure 4 indicates how these frames are blended to produce a new
abstract genre-frame. In each space, I set out a schema for the rhetorical
situation—the relations among writer, text, fiction, and audience—that
extensively (though not fully) defines each genre. I have added sepa-
rately the “surface symbolic form,” the formal textual cues we find on
mixed genres in ulysses 601

the page, which identify the genre for readers and guide their under-
standings.

Composition

Most basically, the novel’s third-person narration of action is fused

with the drama’s stage directions; and the novel’s third-person reporting
of speech and thought is fused with the drama’s presentation of
dialogue. Hugh Kenner shows how the drama frame implicitly selects
certain novelistic features for development: “Drama develops naturalism’s
externality, its linear progression in time, its refusal to ruminate and to
explain. What the stage shows us is what we can see and hear” (C 345).
Thus the central resulting inferences are that the novelist becomes a
kind of playwright with his narrator as the director; the novel characters
become play characters, but also the actors playing those roles, hence
they “play themselves”; and the novel reader becomes the play audience.
These inferences are unavailable in either input genre on its own.
But Joyce also creates certain curious features of rhetorical situation
that have become possible by virtue of the basic drama-in-novel blend
but are not naturally implied in it. First, he exploits the possibility of
presenting extremely non-naturalistic action to a putatively observing
audience.25

Completing and Elaborating the

Drama-in-Novel Blend: The Dream Frame

In order to see how the main forms of completion and elaboration

develop from the basic composition connections, we must turn to some
specifics of how a “dream” frame is added to the blend. Although Ulysses
is rhetorically mainly a novel rather than epic or anything else, this
chapter takes its own organizing genre frame from the drama input, as
is immediately clear from the “script” conventions. Yet despite those
comforting signs, we find ourselves in a surpassingly strange world.
Consider the “action.” Bloom is put on trial for past sexual misdemean-
ors; he then becomes Lord Mayor, then Messiah, then Martyr; then the
dominatrix appears to humiliate him; and later he acts as a hat rack for
his wife’s lover. The “speakers” include Bella’s “hoof”; a bar of soap that
becomes the sun; a dog; a horse; yew trees; the dead; a pianola; a gas jet;
a fly bill; a door handle; and abstractions like “The Hours,” and the
“Hue and Cry.” So what we discover as we read is how this drama frame
602 new literary history

is used to present subconscious content, in dreamlike fashion. But the

drama frame cannot supply that content. That must come from some-
where else. In our terms, a dream frame is introduced here and blended
with the new drama-novel genre-frame. (Technically, the projection of
new frames into the blend is completion, and the “running” of the
blend by simulation according to its various frames is elaboration.) A
close look shows that in fact the frame is used not to supply content,
exactly, nor to supply form, exactly, but rather to provide a basis for
accessing, selecting, and expressing a certain content. That is, Joyce mines
the subconscious minds of Bloom and Stephen and works their deep-
seated desires and fears into hyperbolic fantasies. This is blend elabora-
tion, which in this case works according to two interacting logics: that of
dreams and that of the stage.
This reveals an important principle of frame blending: what counts as
frame and what counts as content are partly relative to the level of
organization we examine. What is content at one level may be frame at
another level—as “menu” is an element of the “restaurant” frame, but is
also itself a frame with slots for elements like “appetizer section,” “dish
description,” and “price.” Here, at the level of the dream frame, the
content comes from the elements of the realistic story. At the “higher”
level of the drama frame, the content comes from the elements of the
dream.26 The complex interdependency that exists among frames and
elements at different levels of specificity also suggests some of the
difficulty of distinguishing genre blending from other kinds of blend-
ing. In fact, there may be no hard-and-fast dividing line. Blending in
features like character or style will have an effect on judgments about
genre, and may well inspire emulation by later authors and thence shape
the history of the genre. Rather, it is reasonable to view genre as a level
of an artistic structure we choose to examine. To grasp its workings, we
may need to refer to lower and higher levels of structure and refer them
back to the level (or levels?) of genre.
Kenner remarks that Joyce’s extreme development of nonnaturalistic
fantasy content is logically connected to interior monologue: “What
‘Circe’ shows us is what we could see and hear were everything pertinent
to the goings-on translated into terms of seeing and hearing, a method
already latent in the naturalism of the early chapters. . . . [T]he logic of
the method . . . is to place all figures, all analogies, all ruminations, on
the plane of the visible and audible” (C 345–46). Our analysis does not
foresee this development, but the emergent feature does not embody
any conflict with blending processes. More strikingly, Dermot Kelly
notes some (less prominent) features that are not only not implied in
the drama-in-novel blend, but seem to run contrary to it. Chiefly, there
is some character speech and thought in the stage directions.27 This
mixed genres in ulysses 603

conflict or paradox seems to derive from connecting individual ele-

ments of one genre to several elements of the other, both counterparts
and noncounterparts: novelistic speech is matched to both dialogue and
stage directions. This raises the challenge to our genre-blending analysis
more acutely. Kelly and other critics have observed that these paradoxi-
cal features also have some continuity with the interior monologue
technic, which also mixes thought with action and speech (whereas in
the traditional novel, speech, thought, and action tend to be clearly
partitioned).28
I think there is a unified explanation for all these features that go
beyond the new drama-in-novel frame. They come from blending the
drama-in-novel frame with a dream frame, which has a kind of rhetorical
situation very much like that implied by interior monologue. Kenner
also discusses how interior monologue connects quite naturally with
qualities of dream, since dream makes manifest the principle latent in
the interior monologue’s method of seeing the internal in the external
(C 355–56). A blending analysis can develop and refine this point by
specifying how it follows from genre blending: how dream connects
readily with drama-in-novel, and how the connection relates to the
features Kenner associates with interior monologue.
A dream has its own kind of rhetorical situation. Inner personal
concerns (desires, fears, and hatreds, linked to memories and fantasies)
are “projected” onto a mental stage for the mind’s eye (and ear) to
witness. Normally in drama the dialogue tells us what the audience
hears, and the stage directions tell us what the audience sees on stage.
But that the “audience” “sees and hears” more than what is on stage is
already implicit in the drama-in-novel blend: not only do readers
become a kind of audience, the purported audience members become
readers (and therefore have access to the stage directions). That is, the
audience/readers see and hear the authorial thinking and planning
behind the characters’ action, as well as the performance of it. The
third-person objective description of character interiority is a normal
part of the novel, but absent the novel’s reading audience, such
information is available only to writer, director, and actors. This fact is
part of the common structure of dreams and drama-in-novel that helps
them to blend: the audiences of both have access to the inner life of
characters. However, the phenomenon in question goes beyond this.
The direct expression of character thought and speech in stage direc-
tions indicates role connections between characters and writer. It is true
that this is a quality of interior monologue. But Joyce’s “Circe” has a
further rhetorical feature that it shares with dreams but not with interior
monologue: the lack of awareness by the character that he is also taking
the role of writer-creator by expressing stream-of-consciousness in the
 604 new literary history

stage directions. The characters in this chapter show no consciousness

of the formal frame in which they are cast, though Joyce might easily
have created this inference if he wanted to stress the links to interior
monologue. Hence a genre-blending analysis helps explain the most
incongruous as well as the most prominent features of the chapter’s
content and form. The drama-in-novel blend sets the stage (so to speak)
for further blending with the dream frame, which allows for the
integration of the contents, techniques, and themes of the earlier
chapters in a new way, and creates the paradoxical features noted.
Kenner cites other features deriving from blending drama-in-novel
with dream that partially clash with the dream input. First, it is primarily
what is already visible that gets exaggerated into hallucination: “Out-
and-out hallucination, irreducible naturalism, these are extremes of a
continuum, extremes seldom visited” (C 347). Second, some of the
material of Bloom’s fantasies comes from episodes from which he is
absent (C 357). Third, emergent in the same blend, the hallucinatory
action proceeds not only from Bloom’s mind, but also from Stephen’s—
his vision of his dead mother is “the only genuine hallucination in the
chapter” (C 351–52). That is, Stephen’s vision causes him to act in an
erratic way noticed by others. His hallucination actually breaks the
general hallucinatory atmosphere because it leads to genuinely con-
fused and therefore dangerous action.29
Figures 5 and 6 indicate in somewhat more detail how some aspects of
the content of the “Circe”-Dublin blend are completed and elaborated
through blending occurrences of “transformation” from the Odyssey, the
[128.103.147.138] Project MUSE (2024-09-19 16:34 GMT) Harvard Library

Dublin story, and a dream input. Figure 7 shows how the basic drama-
novel genre blend is elaborated through blending with genre-like
elements of the dream input. Finally, figure 8 gives an approximate idea
of how the various categorial schemas and blends involved in “Circe” are
connected in an overall network of spaces at various levels of abstrac-
tion. It also indicates how genre-blending interacts with blending at
other levels. (It should not be taken to suggest that creativity occurs
simply in a top-down fashion.)

Inclusions and Conclusions

“Circe” has external genre relations with the other genres in other
chapters of the book, and with the genre of the book as a whole. The
former impinge on the latter. The chapter also has internal genre
relations among the genres and genre features it evokes. We have
primarily addressed chapter-internal genre relations, but I will comment
briefly on issues raised by chapter-external genre relations.
mixed genres in ulysses 605

The relation of “Circe,” with its particular genre-frame, to the rest of

the text is an example of what Fowler calls “inclusion,” where a smaller
genre is inserted into a larger genre, but does not go to defining the
prevailing genre frame of the whole work (KL 241–43).30 As we have
seen, the book’s containing genre is more novel than drama, despite the
inclusion of a chapter that uses dramatic conventions of presentation. As
a whole, Ulysses may not be a prototypical novel, but it fits naturally
enough within the larger category of prose fiction. Although the drama
of the “Circe” chapter is included, it is not wholly absorbed. Its dramatic
aspect does have an effect on judgments about the book’s genre because
it is part of the book’s overall formal playfulness, its frequent ironic use
of many literary and nonliterary genres and subgenres and their styles,
and these qualities help motivate its classification as satire or Menippean
satire.31
Fowler is interested in how inset genres may become conventionally
linked with their matrical genres as a cause of generic change: “Change
is only likely to occur if the inset form is structurally assimilated; or if its
proportion to the matrical work is large; or if it is regularly linked to the
matrical genre” (KL 243). By these measures, the drama-in-novel
inclusion has been less a factor in generic evolution than, say, the letter-
in-novel. Ulysses probably helped in making dramatic inclusions more
conventional in novels influenced by satirical genres, though it is still
not conventional in realistic novels.
Inclusion, as the insertion of elements from one frame into another
overarching organizing frame, is just one of many kinds of blend.
Fauconnier and Turner call it “single-scope.” They have set out a
taxonomy of blends,32 which might compare with Fowler’s taxonomy of
forms of genre change (KL 170–90) in interesting ways, but we cannot
go into that here. On the other hand, at the level of the chapter, both
novel and drama organizing frames contribute significantly to the blend
(both are part of the presentation of this part of the story), which is
therefore “double-scope.”33
A range of related topics might be considered to flesh out the
implications of my approach. How does Joyce’s blend of drama and
novel compare with other examples? With other possible ways to
combine these genres?34 Joyce thought of Ulysses as an “encyclopedia”
rather than a novel or other traditional genre. As an example of
inclusion in the service of encyclopedism, how does “Circe” compare
with precedents in François Rabelais, Robert Burton, and Laurence
Sterne? They too are known for including other genres (often miniature
versions) within their books. More generally, how are genres collected
and contained in larger forms (sometimes known as “compendium
genres”)? How does genre blending relate to other aspects of the Ulysses
606 new literary history

schema? What is the role in the “Circe” blend of other dramatic

elements in the text? There are dramatic features in “Scylla and
Charybdis” (U 171–72) and in Buck Mulligan’s obscene “play for the
mummers,” “Everyman His Own Wife/or/A Honeymoon in the Hand.”
He lists the cast of characters, and the language associates Mulligan with
a “clown” role (U 178). How is the genre blending clarified by study of
Joyce’s specific influences from drama, and his thinking about it?
Shakespeare haunts Ulysses, and Joyce’s admiration for Ibsen is well
known, as is his enthusiasm for music-hall, burlesque, and vaudeville
stages.35 Joyce’s critical writings regard literature as developing from
lyric to epic to a pinnacle in drama. We see strong dramatic elements in
Dubliners and A Portrait of the Artist as a Young Man (realistic scenes, plain
style, portrayal of raw human conflict), but it seems obvious that the
drama would have blocked the exercise of some of his greatest talents, of
pyrotechnical verbal invention, lyricism, and fantasy. What does his own
play, Exiles, reveal about these attitudes to drama?
In short, we should look at genre mixtures of all degrees of complex-
ity and scale in the light of blending theory. We should examine how
they embody and clarify its basic processes (what counts as composition,
completion, and elaboration) and its constitutive and governing prin-
ciples. The present study is a beginning in the necessary effort to move
in this promising direction.

Social Sciences and Humanities Research Council of Canada

Case Western Reserve University
mixed genres in ulysses 607

Fig. 1. The basic epic-novel blend for Ulysses.

The diagram indicates counterpart connections between elements in the input spaces, but
for the sake of clarity it shows links between inputs and blend in terms of categories of
elements (settings, characters, actions) rather than specific elements. In fact, all of the
specific elements listed in the input spaces and more are projected into the blend.
608 new literary history

Fig. 2. The elaboration of the basic Odyssey-Dublin blend in the “Circe” episode.
Again, the diagram indicates specific counterpart connections between elements in the
input spaces, but only categories of projections to the blend space. Note that each input
has elements with dual counterparts in the other input.
 mixed genres in ulysses 609
[128.103.147.138] Project MUSE (2024-09-19 16:34 GMT) Harvard Library

Fig. 3. Connections between basic epic-novel blend and its elaboration in the “Circe”
episode.
610 new literary history

Fig. 4. Genre blend of drama conventions and novel conventions.

mixed genres in ulysses 611

Fig. 5. The blend creating Joyce’s “hallucination technic.”

Note that more than two inputs contribute to the blend, and that inputs and blend are
processes (conceptualized as “procedural schemas”), not static concepts.
612 new literary history

Fig. 6. Transformational blend creating the dreamlike aspects of the “Circe” content.
Note that the hallucinated elements and events occur alongside the real actions of the
chapter, and interact with them.
mixed genres in ulysses 613

Fig. 7. Elaboration of drama-novel genre blend through blending with dream input.
This produces the paradoxical rhetorical-generic features of the chapter as emergent
structure.
 614 new literary history
[128.103.147.138] Project MUSE (2024-09-19 16:34 GMT) Harvard Library

Fig. 8. Organization of categorial schemas and blends in “Circe” network in terms of

abstraction hierarchy.
mixed genres in ulysses 615

NOTES

1 Alastair Fowler, Kinds of Literature: An Introduction to the Theory of Genres and Modes
(Oxford: Clarendon Press, 1982), chap. 10 (hereafter cited in text as KL); Rosalie Colie,
The Resources of Kind: Genre-Theory in the Renaissance, ed. Barbara K. Lewalski (Berkeley and
Los Angeles: University of California Press, 1973); Tzvetan Todorov, “The Origin of
Genres,” in Modern Genre Theory, ed. David Duff (New York: Longman, 2000), 193–209, rpt.
from Todorov, Genres in Discourse, trans. Catherine Porter (New York: Cambridge Univer-
sity Press, 1990); Mikhail Bakhtin, “The Problem of Speech Genres,” in Modern Genre
Theory, ed. Duff, 82–97, rpt. from Bakhtin, Speech Genres and Other Late Essays, trans. Vern W.
McGee, ed. Caryl Emerson and Michael Holquist (Austin: University of Texas Press, 1986),
60–67, 78–81, 96–101; Jacques Derrida, “The Law of Genre,” trans. Avital Ronell, in On
Narrative, ed. W. J. T. Mitchell (Chicago: University of Chicago Press, 1981), 51–77.
2 Duff, Modern Genre Theory, 232.
3 Ralph Cohen, “Introduction: Notes Toward a Generic Reconstitution of Literary
Study,” in “Theorizing Genres II,” special issue, New Literary History 34, no. 3 (2003): xvi.
4 A. Walton Litz, “The Genre of Ulysses,” in James Joyce: A Collection of Critical Essays, ed.
Mary T. Reynolds (Englewood Cliffs, NJ: Prentice Hall, 1993), 114.
5 Colie, in Resources of Kind, and Adena Rosmarin, in The Power of Genre (Minneapolis:
University of Minnesota Press, 1985), also cite Gombrich’s use of the term in Art and
Illusion. Others invest in the concept on the basis of other sources. See Daniel Chandler,
“An Introduction to Genre Theory,” 1997, The Media and Communications Studies Site,
University of Wales, Aberystwyth, http://www.aber.ac.uk/media/Documents/intgenre/
intgenre.html (accessed September 19, 2000); David Fishelov, Metaphors of Genre: The Role
of Analogies in Genre Theory (University Park: Pennsylvania State University Press, 1993); E.
D. Hirsch Jr., Validity in Interpretation (New Haven: Yale University Press, 1967); Howard
Mancing, “Prototypes of Genre in Cervantes’ Novelas ejemplares,” Cervantes: Bulletin of the
Cervantes Society of America 20, no. 2 (2000): 127–50; Michael Sinding, “After Definitions:
Genre, Categories and Cognitive Science,” Genre: Forms of Discourse and Culture 35, no. 2
(Summer 2002): 181–220; Gerard Steen, “Genres of Discourse and the Definition of
Literature,” Discourse Processes 28, no. 2 (1999): 109–120; Peter Stockwell, Cognitive Poetics:
An Introduction (London: Routledge, 2002); Mark Turner, Reading Minds: The Study of
English in the Age of Cognitive Science (Princeton: Princeton University Press, 1991);
Reinhold Viehoff, “Literary Genres as Cognitive Schemata,” Empirical Approaches to
Literature: Proceedings of the Fourth Biannual Conference of the International Society for the
Empirical Study of Literature—IGEL, Budapest, August 1994, ed. Gebhard Rusch (Siegen,
Germany: Siegen University Press, 1995), 72–76; and Anne Williams, Art of Darkness: A
Poetics of Gothic (Chicago: University of Chicago Press, 1995). Among discourse analysts,
John M. Swales, Genre Analysis: English in Academic and Research Settings (New York:
Cambridge University Press, 1990), and Brian Paltridge, Genre, Frames and Writing in
Research Settings (Philadelphia: J. Benjamins, 1997) have discussed schemas in genre
theory. For excellent studies of the general applicability of cognitive theory to literary
studies, and schema theory in particular, to literary studies, see Turner, Reading Minds;
Patrick Colm Hogan, Cognitive Science, Literature, and the Arts: A Guide for Humanists (New
York: Routledge, 2003); and Stockwell, Cognitive Poetics, chap. 6. For scripts, frames, and
schemas in narrative theory, see David Herman, Story Logic: Problems and Possibilities of
Narrative (Lincoln: University of Nebraska Press, 2002), chap. 3. Alan Richardson’s
website, Literature, Cognition & the Brain, http://www2.bc.edu/~richarad/lcb/home.html,
provides a thorough annotated bibliography and ongoing contact with research in this
area. Hogan’s Joyce, Milton, and the Theory of Influence (Gainesville: University of Florida
Press, 1995) shows how a thoroughgoing cognitive approach to Joyce can address specific
interpretive issues while advancing a research program into a general topic.
616 new literary history

6 Litz argues that the image of the work as a totality in the end supersedes whatever
genre knowledge we use in attaining it. In my view, this argument does not necessarily
imply denying the validity of genres. It only implies that the new mixed genre is more than
the sum of its parts. In fact, even a piece of formulaic “genre fiction” is more than its genre,
and only a naïve genre critic would claim that a work’s meaning is exhausted by its genre
or genres, or how it uses them.
7 The facts of schematic idealization and default settings indicate a major point
distinguishing the cognitive from the traditional analysis of concepts and categories.
Under the traditional analysis, categories are defined by necessary and sufficient condi-
tions: anything that meets the criteria belongs to the category, and hence all members
belong equally. Cognitive theory makes much of “prototype effects” in judgments about
category structure. Some members are considered more “central” or prototypical for the
category in various ways, while others are “noncentral.” For example, robins are central to
the category “bird,” while ostriches and penguins are noncentral. Eleanor Rosch and her
colleagues first studied such effects in the 1970s. George Lakoff, in Women, Fire and
Dangerous Things: What Categories Reveal About the Mind (Chicago: University of Chicago
Press, 1987), provides a thorough overview of the research and issues arising out of it and
explains prototype effects as resulting from the structuring of categories in terms of
Idealized Cognitive Models. Gregory Currie, “The Film Theory That Never Was: A Nervous
Manifesto,” in Film Theory and Philosophy, ed. Richard Allen and Murray Smith (New York:
Oxford University Press, 1997), 42–59; Mancing, “Prototypes”; Sinding, “After Defini-
tions”; Steen, “Genres of Discourse”; Turner, Reading Minds; and Williams, Art of Darkness,
discuss prototype issues in genre theory.
8 Litz, “The Genre of Ulysses,” 111.
9 Carolyn R. Miller’s paper “Genre as Social Action,” Quarterly Journal of Speech 70
(1984): 151–67, has been extremely influential in defining rhetorical genres as types of
social action structured as responses to recurring rhetorical situations. Thus more fine-
grained characterizations of rhetorical situation provide more fine-grained distinctions
from neighboring genres, as well as more extensive links to rhetorical structure. For
example, the discursive distinctions between acceptance and rejection letters for job
candidates stem from the fact that they enact different kinds of social action demanding
distinct kinds of social action in response. For an analysis of nonliterary genres in terms of
“frame semantics,” see Paltridge, Genres, Frames and Writing, especially chap. 3.
10 Thus, “a novel might be characterized as follows: Its content would be fictional and
portray a significant action or process. Its form could be one of suspense, surprise, or
curiosity. . . . Its type would be narrative. . . . Its function would be to positively affect the
mood of the reader. Its medium would be printed matter for a mass readership. Its domain
would be . . . the arts. Its language could be characterized . . . as ‘extremely narrative,
moderately involved, situated, nonabstract, and not marked for persuasion’” (Steen,
“Genres of Discourse,” 114). Steen admits his model is simplified in that its differentia do
not capture the subtleties of literary history. It does not distinguish between subgenres
such as sonnets, ballads, and sestinas, or novels of detection, romance, or science fiction.
Nor does it indicate how the multiple schemas are interconnected, although it is clear that
to some extent they form a hierarchy running from abstract to specific. I suspect that any
dimension of this schema can be blended with other schemas at the same level (that is,
type can be some combination of narrative and exposition; medium can be some
combination of print and recorded speech, and so on).
11 George Lakoff and Mark Johnson, Metaphors We Live By (Chicago: University of
Chicago Press, 1980); Lakoff and Johnson, Philosophy in the Flesh: The Embodied Mind and Its
Challenge to Western Thought (New York: Basic Books, 1999), part 1. The theory is developed
mixed genres in ulysses 617

in relation to poetic metaphor in Lakoff and Turner, More Than Cool Reason: A Field Guide
to Poetic Metaphor (Chicago: University of Chicago Press, 1989).
12 See Turner’s Blending and Conceptual Integration website, http://markturner.org/
blending.html. Gilles Fauconnier and Turner, The Way We Think: Conceptual Blending and
the Mind’s Hidden Complexities (New York: Basic Books, 2003), is the most complete
statement of the theory. Fauconnier and Turner, “Conceptual Integration Networks,”
Cognitive Science 22, no. 2 (April–June 1998): 133–87 (expanded web vers. February 10,
2001, Blending and Conceptual Integration, ed. Turner, http://markturner.org/cin.web/
cin.html, accessed March 2001) is an excellent introduction. See also Fauconnier,
Mappings in Thought and Language (New York: Cambridge University Press, 1997), and
Turner, The Literary Mind: The Origins of Thought and Language (New York: Oxford
University Press, 1996).
13 On aspects of genre in “Circe” (including discussions of narrative, rhetoric, style, and
composition), see Hugh Kenner, “Circe,” in James Joyce’s “Ulysses”: Critical Essays, ed. Clive
Hart and David Hayman (Berkeley and Los Angeles: University of California Press, 1974),
341–62 (hereafter cited in text as C); Michael Groden, “Ulysses” in Progress (Princeton:
Princeton University Press, 1977), 13–18, 52–61, 166–75 (hereafter cited in text as UP);
Karen Lawrence, The Odyssey of Style in “Ulysses” (Princeton: Princeton University Press,
1981), chap. 6; Dermot Kelly, Narrative Strategies in Joyce’s “Ulysses,” Studies in Modern
Literature 96 (Ann Arbor: University of Michigan Press, 1988), chap. 6; and Selected Letters
of James Joyce, ed. Richard Ellmann (New York: Viking Press, 1966), 265–77.
14 Turner, Literary Mind, 89–96.
15 James Joyce, Ulysses, The Corrected Text, ed. Hans Walter Gabler et al. (Harmondsworth,
UK: Penguin, 1986), 409 (hereafter cited as U).
16 Stephen’s apparent animal metamorphoses are brief and tangential, and linked to his
intellectual and emotional concerns. They are not sexual, and are far from Bloom’s deep
and consistent mutation. First, he shares with Bloom a reflection of a beardless Shakespeare
in the mirror, “crowned by the reflection of the reindeer antlered hatrack in the hall” (U 463). This
suggests that Stephen identifies with Shakespeare as an artist, and Bloom identifies with
him as a cuckold, and perhaps there is some connection between the roles. But the
moment is only a brief intrusion into Bloom’s fantasy of encouraging Molly’s liaison with
Boylan. Then Stephen stands at the fireplace with “finny hands outspread” (U 464), but this
is just prior to Stephen’s transformation into a marionette who celebrates the sensual
charms of Paris, and presumably indicates that his hands, like a marionette’s, have no
individual fingers. The marionette continues Stephen’s lewd mockery of religion and
leads into another Shakespeare reference (“omlet”). As Stephen briskly recalls a dream
involving a watermelon, flying, triumph, and freedom, he becomes birdlike: “he cries, his
vulture talons sharpened” (U 466). Fittingly, Dedalus senior appears, flying “on strong
ponderous buzzard wings” (U 466), to coach him in a contest that appears to start as
foxhunting and then becomes a horse-race. The allusion to the Icarus-Dedalus story
suggests that his father’s pride is leading him towards a fall, and the use of carrion birds
suggests they are preying on the weak or dead. Then he jumps like a frog (“frogsplits in
middle highkicks”) as he dances with the group (U 472), but this is a minor detail next to his
overseeing an animal-filled dance of the dead, just before he totters and sees his dead
mother.
17 Turner, Literary Mind, 84.
18 Turner, Literary Mind, 84.
19 Ellmann, qtd. in Hogan, Joyce, 48.
20 Northrop Frye, Anatomy of Criticism: Four Essays (Princeton: Princeton University Press,
1957), and Mikhail Bakhtin, The Dialogic Imagination: Four Essays, trans. Caryl Emerson and
Michael Holquist (Austin: University of Texas Press, 1981), among many others, have
articulated this view.
618 new literary history

21 A difficulty here is that knowledge of genres and intuitions about what counts as
generic vary from person to person, and what should be included in input spaces of genre
varies accordingly. Thus some authors might connect even these specific details with
genres—perhaps the metaphoric phrases with lyrical poetry and the poetic novel, and
fantasy beings with gothic novels, fairy tales, or romance generally. The best we can do is
to find out all we can about the genres Joyce consciously knew and used, and make
informed guesses about how he may have linked specific features with genre schemas.
22 Frye, Anatomy of Criticism, 246–47.
23 Bernard Knox, “Introduction,” The Odyssey, by Homer, trans. Robert Fagles (New York,
1996), 5–10, 14–17. Knox discusses the question of when and how The Odyssey was written
down, and what effect writing may have had on its composition and structure (3–7, 18–22).
The Homeric epics are very much longer than what could be recited at a single sitting, so
they may be expansions or aggregations of shorter oral versions or episodes. But that does
not affect the point that there existed a familiar recurrent rhetorical situation for the epic
bard.
24 This characterization of the genre frame in fact applies to prose fiction generally, but,
as we have indicated above, the content is particularly novelistic, and Joyce and his readers
would have known the frame best from traditional novels.
25 The significance of such creative choices appears best in contrast with other available
but unrealized options. Joyce does not exploit, for example, the potential for the narrator
to assume a godlike role scripting or directing or even commenting on the action for an
audience—which might have been motivated from the epic input.
26 Miller sets out a hierarchy of interdependent categories of form, substance, and act. A
fusion of substance and form at one level acquires meaning when it serves as substance for
a higher-level form, and thereby enables a kind of action that has pragmatic force and is
interpretable. For example, the fusion of a grammatical form with a lexical substance
serves as the substance for a proposition and enables an act of locution. The fusion of a
propositional substance with a form of illocutionary force serves as the substance for a
speech act, and enables that speech act (“Genre as Social Action,” 158–62). The analogy
with my sense of the interdependence and relativity of frame and content at various levels
of organization suggests a way towards a unified analysis of the form-content associations
that hold at the level of whole genres as well as at the level of their constitutive rhetorical
features. That is, genres often have a “worldview” or theme, fused with a complex of
features of narrative, character, imagery, discourse, style, and so on, each of which is a
fusion of form and meaning. If such an analysis of tightly interconnected cross-level
structure were possible for literary genres, it might answer the challenge to demonstrating
literary genre coherence posed by the fact that literary genres are not associated with
structure-constraining definite recurrent rhetorical situations and social actions, as are
nonliterary genres.
27 Kelly, Narrative Strategies, 57–58. Also, silent thoughts are rendered as speech. This also
happens in dramatic soliloquies, but in a very different way. “Circe” thoughts are
fragmentary and fleeting, more like the preceding interior monologues than well-formed
soliloquy speeches for the benefit of the audience.
28 Kelly, Narrative Strategies, 57.
29 Some such effects might be possible in a straight novel or drama. But the drama could
not present the details of the hallucinations, or show how perception shapes what is
observed and how it changes. Drama’s visuality could present either the realistic basis or
the hallucination, but not both. And the novel could not present the metamorphoses and
remain both dreamlike and “realistic.” It might achieve a similar hallucinatory effect
through simile and metaphor (“Bloom felt he had become like a swine,” and so on), or by
presenting the scene as an actual dream, but either of these options would downplay the
mixed genres in ulysses 619

power of hallucination. In short, the drama would have no way of distinguishing real from
false hallucination, whereas the novel would have no way of connecting them. Neither
could present both the continuity between the hallucinatory imaginings of Bloom and
Stephen’s “real hallucination,” and the distinctions between them and their conse-
quences.
30 Fowler takes the term from Colie’s insightful discussion in Resources of Kind, chap. 3.
31 See Frye’s discussion of this genre and Joyce’s relation to it. Frye, Anatomy of Criticism,
308–14.
32 Fauconnier and Turner, The Way We Think, 119–35.
33 The blend taxonomy works in terms of the various kinds of possible relations among
the “organizing frames” and “elements” of input spaces. The difficulty in applying this to
literary genre blending lies in finding a suitable equivalent for “organizing frame.” The
reason for the difficulty is that as multidimensional schemas, genres have various kinds of
overall organization—form, action, style, and so forth. I think that plot schemas are the
best candidates for the job, but I cannot argue this here. Fowler distinguishes between
change within single genres, which includes Topical Invention, Combination, Aggrega-
tion, Change of Scale, Change of Function, and Counterstatement; and change involving
connections between two or more genres, which includes Inclusion, Selection, and
Mixture.
34 A complementary kind of blending of novel-in-drama can be seen in George Bernard
Shaw’s highly novelistic long passages describing and analyzing characters as they are
introduced. These may be stage directions by virtue of their location, but they are clearly
intended for the benefit of readers, directors, and actors, rather than audience. They
include details of physique, bearing, psychology, personal history, and cultural allusion
that an actor cannot hope to convey. See for example the introductory account of Major
Sergius Saranoff early in act 2 of Arms and the Man. Several major novelists have been tacitly
influenced by drama’s rhetorical situation. The dramatic qualities of Henry Fielding’s
novels, emerging from his work in that genre, have been discussed; and David Lodge, in
After Bakhtin: Essays on Fiction and Criticism (London: Routledge, 1990), has studied “scenic”
methods in Jane Austen (116–28) and in Henry James’s Spoils of Poynton (129–42).
35 See B. J. Tysdahl, Joyce and Ibsen: A Study in Literary Influence (New York: Humanities
Press, 1968).