Vol.:(0123456789)
American Journal of Cultural Sociology
https://doi.org/10.1057/s41290-019-00087-6
ORIGINAL ARTICLE
Representationalism and cognitive culturalism: riders
on elephants on turtles all the way down
Jason L. Mast1
© Springer Nature Limited 2019
Abstract
This article examines the influx of neurocognitive concepts in cultural sociology and
this development’s consequences for representationalism. In the first part, I examine
representationalism in two research programs that have shaped cultural sociology from
the cultural turn to the present: Jeffrey Alexander’s “strong program” and Ann Swidler’s
“tool kit” theory. I also briefly discuss the mixed and contradictory findings presented
in one of sociology’s most-cited cognitive works, Paul DiMaggio’s (Annu Rev Sociol
23:263–87, 1997) programmatic statement on cognitive psychology’s potential contributions
to sociology, which catalyzed the discipline’s cognitive turn. In part two, I
demonstrate how in working with and against these three pillars in cultural sociology,
figures such as Omar Lizardo, Steven Vaisey, and John Levi Martin have drawn on the
cognitive neurosciences to re-conceptualize culture in ways that may have profound
consequences for representationalism as it is practiced in the field. I conclude by arguing
that representationalism is present but suppressed in cognitive cultural theory and
its empirical investigations; that representationalism finds support in the neurocognitive
sources that cognitive culturalists cite; and by asserting that future general theories
of action will be predicated on a more interactive relationship between automatic and
deliberative cognitive domains than the cognitive culturalists currently allow.
Keywords  Representationalism · Cognition · Schemas · Dual process model ·
Habitus
Introduction
In this article, I examine the question, as cognitive and neuroscientific terms proliferate
in cultural sociology, what are the consequences for approaches that privilege
signification and representationalism? I define the latter terms broadly, as a range
*	 Jason L. Mast
	MastJasonL@gmail.com
1
	 Cluster of Excellence “Normative Orders”, Goethe University Frankfurt, Max‑Horkheimer‑Str.2,
60323 Frankfurt am Main, Germany
	J. L. Mast
of approaches that attribute varying degrees of relative autonomy to symbol systems,
and that attribute to these systems varying degrees of causal power. To examine
this question, I discuss two programmatic cultural sociological approaches that
have contributed significantly to shaping how representationalism is understood and
practiced in cultural sociology from the cultural turn to the present. First, I present
Jeffrey Alexander’s “strong program” as representative case of representationalism.
Second, I discuss Ann Swidler’s influential pragmatic alternative to representationalism,
which was predicated on demonstrating the limited degree to which symbolic
representations shape social action, on the one hand, and reducing representationalism’s
claims to explanation, on the other. I offer a new reading of Swidler’s empirical
investigations, and draw to the fore how representationalism permeates her work
while remaining relatively unacknowledged. To round out this section, I briefly
reconstruct Paul DiMaggio’s (1997) influential review article, Culture and Cognition.
I illustrate how even though he concluded that cognitive theory and empirical
findings strongly support the tool kit theory of culture in action, DiMaggio nonetheless
presented multiple examples of how the cognitive sciences were simultaneously
producing concepts and findings that substantiated representationalism’s theoretical
presuppositions.
Establishing a chronological dimension to my argument, in the article’s second
section, I investigate three figures who have incorporated cognitive neuroscientific
terms and findings into their works in ways that critically engage with the programs
described above. I examine the fate of representationalism in Omar Lizardo’s (2004,
2007) cognitive neuroscientific reconceptualization of the habitus concept, in Steve
Vaisey’s (2009) dual process model of culture in action, and in John Martin’s (2010)
readings of cognitive neuroscientific developments, which he presents as demonstrating
that representationalism is grounded on erroneous presuppositions.
I chose these articles because they have been influential in building cognitive
culturalism into the energetic movement it is today. These articles have an enviable
number of citations. Their authors’ names frequently dot cognitive culturalist texts,
and oftentimes appear together, cited side-by-side. It is suggestive of the development
of an iconic status, as if the citations signify the promise of a consolidated
paradigm with settled foundations (e.g. Patterson 2014). While cognitive culturalism
is by no means a settled field, these particular figures and works have exercised
considerable organizing power. Foundational statements  may influence research
programs long after the details of each particular argument have grown obscure. It
is worthwhile to examine these articles in depth lest they become to some extent
symbols of shared knowledge, totemic emblems of a paradigm’s formative statements
(Alexander 1987; Collins 1998). While I chose these because they have been
influential, I also chose them for more specific purposes. The articles in the first case
present examples of neurocognitive literatures shaping theory and concept development.
The main article in the second case presents an example of empirical testing
interpreted as justifying non-representational theory. Finally, the article in the third
case marshals neurocognitive literature in effort to disprove the foundations of representationalist
approaches to social action.
In selecting these three figures, I do not mean to suggest that a pattern of influence
exists between them, or that they represent a unified core of consensus within
Representationalism and cognitive culturalism: riders on…
contemporary cognitive cultural work. The latter remains diverse, and disagreements
over fundamental issues continue to animate it (see Brekhus and Ignatow [2019] for
a comprehensive overview). In fact, emerging debates suggest its organizing theories
and key concepts remain far from settled, much as they remain unsettled in the
fields of cognitive psychology and the cognitive neurosciences from which these
scholars draw. Cognitive culturalism is a vibrant program in the field. My intention
is to encourage the reconsideration of the role representationalism (in all its varieties)
will play in its development. The authors discussed here have identified tensions
between the two. In the pages that follow, I examine these tensions as they are
specified and developed in these particular cognitive cultural works. I identify ways
that representationalism seems at first to yield only then to reappear in significant
epistemological spaces, such as in theoretical logic, and in the research that is cited
to build or sustain the arguments.1
In my analysis, I argue that representationalism lurks within these contemporary
cognitive works, as well as within the sources they cite to build empirical tests and
to buttress their arguments and conclusions. Also, more than linking representationalism
with discursive consciousness, these works explicitly bracket signification and
symbol systems from automatic cognition in terms of both processes of internalization
as well as those shaping motives and action. I draw on cognitive theorists and
scholars to argue, to the contrary, that during socialization automatic schemata are
shaped by symbol systems, and that during internalization the dispositions encoded
in automatic schemata come to bear these symbolic forms (D’Andrade 2002). Likewise,
these cognitive culturalist works represent automatic and deliberative cognition
as if they are entirely distinct, and as if interactions between these cognitive
domains are so inefficient as to be extraordinarily rare. As a consequence, they
represent action as the product of an excessively pure automatic cognition. I argue,
rather, that interactions between automatic and deliberative cognition are more frequent
and complex than presented in cognitive culturalism (Kennett and Fine 2009;
Moore 2017; Shepherd 2011),2
and that a general theory of social action should give
significant attention and causal weight to these inter-domain relations and outputs.
Part I: Representationalism in the cultural turn, and the cognitive
intervention
Jeffrey Alexander’s strong program presents us with a theoretical framework that
exemplifies the general presuppositions animating much of contemporary representational
cultural sociology. Alexander has not engaged in a direct and sustained way
1
 I do not exhaust the cognitive or neuroscientific works these authors reference. Instead, I focus
on some of the central figures as they are presented in the articles, many of which—Bellah, D’Andrade,
Gallesse, Lakoff, Miller, Piaget—have been influential beyond the domains and concerns presented here.
2
  See Brekhus and Ignatow (2019, pp. 12–13) for a brief discussion of recent efforts to revise and refine
the dual-process model of cognition.
	J. L. Mast
with cognitive neuroscientific theory.3
On the other hand, Alexander’s work and the
strong program’s version of representationalism have played the foil for cognitive
culturalists in recent years (see Lizardo and Strand 2010; Martin 2010). Critics have
argued that new developments in cognitive neuroscientific theory and conceptual
modeling reveal fallacies in some of the strong program’s and representationalism’s
foundational premises. I investigate these arguments below.
In this section, I outline pertinent themes within the strong program that cognitive
culturalists have addressed. Given that Alexander’s strong program offers both
a well elaborated theoretical framework as well as a number of empirical investigations
undertaken based on its premises, I discuss it here as a representative case of
representational cultural sociology. To be clear, the strong program by no means
exhausts the theoretical orientations, conceptual frameworks, or methodological
strategies being used by cultural sociologists practicing some form of representationalism
today. Nor do the cognitivist works I discuss below represent the entire
range of cognitivism currently being practiced in cultural sociology, or the range
of the work being developed by the particular authors I cite. I chose the following
works because either they have been highly influential in the field or because their
findings are of particular relevance and consequence for the themes explored herein.
The“strong program in cultural sociology”as representative case
of representationalism
In his critique of Talcott Parsons, Clifford Geertz argued that the theorist tended to
reduce cultural forms to being the product of strains between other systems, which
prevented him from “ever seriously examining [cultural forms] as systems of interacting
symbols, as patterns of interworking meanings” (1973a, p. 207). The works
of both of these figures, Parsons and Geertz, informed Jeffrey Alexander’s development
of the “strong program in cultural sociology,” as did those of their critics. In
“Thick Description,” Geertz (1973b) outlined a structural or semiotic theory of culture.
In “Notes on the Balinese Cockfight,” Geertz (1973c) argued that social action
should be treated as a text and subjected to hermeneutic analysis. Alexander calls
the strong program’s principal method structural hermeneutics, a name that aptly
signifies its influences. As we will soon see, however, Alexander would depart from
Geertz in significant ways, and the theoretical and empirical work he created in the
latter half of his career can accurately be described as post-Geertzian.
While Geertz played an important role during the strong program’s formative
period, it was also during this time that other theorists were turning to pragmatism
and phenomenology to offer trenchant criticisms against determinisms of all
kinds. Alexander recognized that excessive idealism and structuralism would create
residual categories of local, interactional contexts, reduce actor subjectivity to
cultural form, and fail to account for both rational and strategic actions, as well
3
  See Alexander (2006, pp. 569–570, note 4) for an indication of how the author interprets the cognitive
linguistics of George Lakoff in a manner that substantiates the culture-structural dimensions of Alexander’s
research program.
Representationalism and cognitive culturalism: riders on…
as habitual, typified practices that real social actors engage in as they navigate
their ways through their daily lives. Building on insights of figures like Garfinkel
and Goffman, Alexander (1988 [1987], pp. 311–316) developed a theory of social
action that conceptualized actors as experiencing their environments through a
combination of interpretation and strategization. He added that interpretation is
itself a combination of typification and invention. To the extent that the research
programs discussed here privilege one of these dimensions over the others, we
will note the centrality of interpretation in representationalism, strategization in
the tool kit theory, and typification in cognitive culturalism. What will become
increasingly clear in the following is that while irreducible to one another, conceptually
interpretation and typification share a number of significant themes.
One of the central points of contention between representationalists and cognitive
culturalists is the question of what forces predispose the mechanisms of typification
to intuit particular patterns in experience (dispositions). We will return to
this theme below.
In a programmatic statement outlining the strong program’s version of representational
cultural sociology, Alexander and Smith (1993) conceptualize culture
in structural and narrative terms while simultaneously attributing to social actors
dimensions of autonomy and capacities for acting in strategic, inventive, and even
intuitive and habituated ways. This sentence outlines the theoretical ambitions:
Bringing contingency and institutional effects back into our understanding
of how culture works is a vital task. In achieving this micro-macro link,
however, one must not overlook the reality of emergent properties, which
demands that the integrity of different levels of analysis be maintained. Neither
the importance of attitudes and actions, nor the significance of organization
and environment, negates the existence at still another level of a cultural
system (p. 154).
The authors conceptualize symbolic structures in these terms:
[B]eneath narrative there lie structures of a more basic kind which organize
concepts and objects into symbolic patterns and convert them into signs…
Complex cultural logics of analogy and metaphor, feeding on differences,
enable extended codes to be built up from simple binary structures…
Because meaning is produced by the internal play of signifiers, the formal
autonomy of culture from social structural determination is assured… [S]
igns sets are organized into discourses. These discourses not only communicate
information, structuring reality in a cognitive way, but also perform a
forceful evaluative task (pp. 156–157).
While signification informs the cognitive apprehension of reality, the authors
emphasize that actors and action have degrees of autonomy vis-à-vis these structures.
Collective cultural forms do not  exhaust subjective capacities, nor do
they determine action in an absolute way. Perhaps anticipating the critiques presented
by contemporary cognitive culturalists, Alexander and Smith explain that
from “a semiotic perspective, cultural codes are elastic because there is only a
	J. L. Mast
conventional, not a necessary, relation between signifier and referent” (p. 158).
This means that no single, fixed meaning inheres in a collective representation
or sign. Meanings become fixed to varying degrees through convention. Consequently,
there is no presumption that social actors internalize perfectly a perfectly
ordered and coherent and stable system of representation. Alexander and Smith
continue, asserting that “cultural codes are elastic because individuals can ad-hoc
from event to code and from code to event” (ibid.). Not only are meanings presented
as variable, as existing in between fixed and unfixed states, but actors are
conceptualized as orienting toward them in contingent ways.
The authors continue, explaining:
There is no inconsistency, then, between speaking of cultural structures and of
the contingency of action. Accountability and symbolic classification are different
theoretical levels—emergent properties—of the same empirical process;
they are concepts that explain the reciprocal interaction of structure and action.
Culture, in our understanding, is one of the internal environments of action
(158–159).
Given that theory requires abstraction and reduction, it is no surprise that when theorists
characterize their theoretical opposition, they tend to abstract from their opposition’s
arguments and reduce them in advantageous ways. Critics of representationalism
tend to gloss over some these details. As a result, actual representationalist
works fail to demonstrate the features—excessive coherence and stability, perfect
internalization—attributed to them by influential cognitive culturalist critics.
Tool kit theory and the displacement, suppression, and reinvigoration
of representationalism
Ann Swidler (1986) developed her tool kit theory of culture in action as an explicit
rejection of such approaches as the strong program’s version of representationalism.
She states that her tool kit theory of culture in action is intended to be a resource for
“those interested in cultural explanation, (as opposed to ‘thick description’”) (p. 273,
italics and parentheses in original), or that which Clifford Geertz argued the social
scientist should strive to produce. This binary code in cultural sociology’s epistemological
foundation imposes on the field’s practitioners a choice between relentlessly
pursing cures for one’s explanation envy or learning to live with one’s interpretation
anxiety. The promise of achieving explanation in a strict causal sense and of becoming
a cumulative, predictive science (see DiMaggio 1997), runs through and also
ahead of the cognitive culturalist texts analyzed below.
Culture, Swidler argues, “is more like a ‘tool kit’ or repertoire… from which
actors select differing pieces for constructing lines of action” (1986, p. 277). Tool
kit culture is comprised of styles, skills, and habits that people use routinely. These
tools are more local, immediate, or tangible to actual social actors than are abstract
symbol systems, and they greatly influence how actors navigate their everyday lives.
From this formulation, Swidler draws two important conclusions: a) these tools and
practices represent the essential, most causally powerful materials in what we call
Representationalism and cognitive culturalism: riders on…
culture, and b) sociologists of culture should train their attention on these tools and
practices, and dispense with attending to broader cultural forms. She does not deny
the existence of cultural structures, however: Strategies “of action incorporate, and
thus depend on, habits, moods, sensibilities, and views of the world,” she acknowledges,
citing Geertz. “People do not build lines of action from scratch,” she continues,
but instead “construct chains of action beginning with at least some pre-fabricated
links” (ibid.). Yet she denies these symbolic forms much causal relevance.
This kind of one-dimensional theorizing produces residual categories that
weaken a theory’s explanatory purchase. Swidler demonstrates that she is aware
that by limiting cultural causes to strictly local, or micro-interactional manifestations,
her theory is incapable of explaining what Geertz’s work appears to explain
so well. She wrestles with this tension in her subsequent work, Talk of Love. It is in
an article written in collaboration with Iddo Tavory, however, that Swidler not only
acknowledges that collective representations shape social interactions, but she and
her coauthor fold signification into their theoretical conclusions, and in a powerful
way (Tavory and Swidler 2009).
Swidler’s (1986) argument is most powerful as a call for developing a conceptual
lexicon for cultural forms along multiple axes, but foremost in terms of their
degrees of being pervasive, broadly understood, and publicly available, on the one
hand, to being personalized, experience-proximate, and even highly stylized and idiosyncratic,
on the other: “Assumed here is a continuum from ideology to tradition to
common sense” (p. 279; italics in original, but for our purposes the emphasis is on
“continuum”). What remains unconvincing, however, is the postulate that a kind of
barrier exists between collective and personalized forms, or any assertion that these
forms are necessarily different not just in scale or degree but in kind. Difference
in degree—i.e., a continuum—keeps open the question of the extent to which an
individual’s personal ideas and intuitions may be shaped by collective representations
regardless of the actor’s awareness of this symbolic association. If the aim is
to produce a generalized theory of culture and action, then in this continuum formulation,
it is neither logical nor tenable to argue that the local and interactional
end of the continuum monopolizes causal significance while the other end, the one
consisting of discourses, narratives, and collective representations, does not share
the same capacity. Certainly, the contribution of each cultural dimension may vary
across empirical investigations, but a generalized theory ought not foreclose the possibility
of a cultural explanation that demonstrates connections between collective
forms and individual expressions.
It is possible to interpret Talk of Love as Swidler (2001) doubling down on pragmatism
in a defiant rejection of Geertzian structuralism. A more revealing and productive
reading, however, comes from focusing not on her theoretical conclusions
but on the summaries of her empirical analysis. In this reading, culture, as representations,
is always shaping action in some capacity, during settled and unsettled times
alike. For instance, in chapter six, Swidler states that this “chapter has shown how
images of romantic love continually resurface even among people who consciously
disavow them” (p. 128). Here Swidler describes mythic love as having the conceptual
characteristics of a collective representation, one that is pervasive, durable, and
powerful even for those who reject it with their utmost of intention. One could say
	J. L. Mast
it a collective representation that her subjects have “internalized.” Alongside mythic
love exists “prosaic-realism,” which is a second narrative of love that “is just as ‘cultural’
as the mythic view it claims to debunk” (p. 114), or, as one may be tempted to
rephrase it, “just as ‘mythic’ as the cultural view it claims to debunk.”
These binary repertoires (or we could call them narrative complexes) shape people’s
actions when they are making big, life-changing decisions, and when they are
navigating their routine, everyday lives. Swidler observes that they interpret love in
mythic terms when “thinking about the choice of whether to marry or stay married,”
on the one hand, and that they think about their intimate relations in prosaic-realism
terms when “thinking about their ongoing relationships,” on the other. As representationalism
in contemporary cultural sociology does not insist that there is a stable
overarching symbolic logic determining action, in this reading we can conclude
that Swidler’s findings are not a rejection of representational cultural sociology but a
demonstration of its purchase.
In a subsequent work, coauthored with Iddo Tavory, Swidler turns to signification
to demonstrate that semiotic cultural codes govern the meanings of intimate actions
and relationships, and to show how these symbolic representations are encoded
in material objects like the condom (2009). These meanings are not created in the
interactional context. They are fixed through convention, widely understood, and
powerfully constitutive. These meanings shape actions in the interactional context.
The authors describe their theoretical conclusions thus:
we show that culture constrains and shapes action not simply because all
actors, institutions, and actions instantiate cultural codes. Rather, within situational
contexts, individuals find actions to be semiotically charged a priori;
these a priori meanings shape all future actions…
Semiotic codes are powerful because they shape the ways we read the behavior
of others (and, reciprocally, the ways we know others will read our own behavior)
(p. 185).
As we move to discuss cognitive theory and its impact on cultural sociology, it
is critical to recognize the importance of Swidler’s receptiveness to the explanatory
power of representationalism’s tools. Tavory and Swidler conclude that cultural
codes shape the shared understandings and personal dispositions that actors’ draw
on when they negotiate and engage in intimate interactions. To put it in cognitive
culturalist terms, the authors are arguing that the shared cultural codes are operating
within each actor’s practical consciousness. To translate it one step further,
the authors are arguing that cultural codes shape the automatic cognitive schemata
embedded in the neural networks of individual brains within a vast network of social
actors.
Cognitive psychology’s contradictory signals to competing cultural sociology
programs
Paul DiMaggio (1997) opens his review of cognitive psychology’s potential for redirecting
sociology with exuberance. Cognitive theory offers the promise of exodus, or
Representationalism and cognitive culturalism: riders on…
passage from a land of “[i]nterpretive studies that offer great insight but fail to build
on one another,” to one governed by epistemological certainties that will enable the
sociological “study of lived culture [to become] a cumulative enterprise” (c.f. Lizardo
2014, p. 263). The map to the promise land: “clarify the cognitive presuppositions
behind” your theories, concepts, and units of analysis (1997, p. 263).
DiMaggio presents us with some basic cognitive concepts that will be useful for
navigating the remainder of this article. To the extent that order exists, DiMaggio
says that it is produced within the active human brain (p. 267). Cognitive schemas
perform this work. Schemas are “both representations of knowledge and information-processing
mechanisms” (p. 269). They are attributed a central role in structuring
action and order: “In schematic cognition we find the mechanisms by which
culture shapes and biases thought.” And they are the lynchpin of cognitive theory
and its claims to explanation: “It may be useful to treat the schema as a basic unit of
analysis for the study of culture” (p. 269). DiMaggio reports that cognitive psychology
“directs the search for sources of stability and consistency in our beliefs and
representations, first, to schematic organization, which makes some ideas or images
more accessible than others; and, second, to cues embedded in the physical and
social environment” (p. 267).
Just as schemata are attributed enormous epistemological power, so is the distinction
between automatic and deliberative cognition. Automatic cognition refers to
schemas shaping perception and recall in intuitive, unreflexive, and uncritical ways.
Schemas privilege inputs that resemble their pre-existing substantive orientation (or
disposition), both in terms of how they apprehend and attend to the external world,
and in how they recall information and experience in one’s memory. Automatic cognition
is undisturbed by an input’s or memory’s truth-value. To the contrary, it will
shape inputs to fit its pre-existing interpretive organization of the world. Likewise, it
is capable of inventing memories that confirm its ordering predisposition. Deliberative
cognition, by comparison, is effortful, challenging, and inefficient. Actors are
described as shifting from automatic to deliberative cognition when they are provoked
into self-awareness by the need to carefully consider information; when they
are motivated to respond to undesirable conditions; or when their schematic predispositions
prove strongly inconsistent with new information.
DiMaggio concludes from his review of the literature that cognitive psychology
“strongly supports the tool kit… view and suggests that the tool kit is very large
indeed” (p.  267). It soon becomes clear, however, that contradictions abound in
DiMaggio’s report, and in cognitive psychology more generally. For instance, as we
have seen, tool kit theory casts theories that attend to supra-individual culture as
missing the real sites of explanation. Yet DiMaggio reports that “[i]ncreasingly…
psychological research bolsters and clarifies the view of culture as supra-individual,
and even addresses supra-individual aspects of cognition directly” (266). Additionally,
he describes how cognitive “psychological research can help us appreciate”
that “relatively coherent cultural forms exist independently of persons in the
broader environment,” forms such as “diffuse myths, images, and idea systems,” and
other “relatively coherent representations [that] exist less formally as narratives or
stories repeatedly invoked in public discourse” (pp. 272–273). In a final example,
DiMaggio describes how cognitive research demonstrates that people do not acquire
	J. L. Mast
culture “by imbibing it… through socialization” (p. 267). Yet just a few sentences
later he notes that cognitive research “raises the possibility that socialization may be
less experientially based, and more dependent upon media images and hearsay, than
many” of sociology’s extant theories suggest (p. 268).
To complicate things further, the cognitive cultural works discussed below draw
on cognitive neuroscientific theory and research to substantiate conclusions that
contradict many of those that DiMaggio drew from his review of the literature. For
instance, Omar Lizardo (2004, 2007) draws on cognitive neuroscientific works to
argue against the second finding on socialization. John Martin (2010) draws on
works in cognitive neuroscience to argue against the thesis that “supra-individual
culture” is either part of cognition or has an impact on action. And in direct contrast
to DiMaggio’s main conclusion, Steve Vaisey (2009) builds on the moral foundations
theory, developed by social psychologist Jonathan Haidt (2001), and the works
of cognitive anthropologist Roy D’Andrade (1995), to argue that the tool kit theory
of action is in fact not supported by cognitive psychology at all. Below we will
consider in greater depth how Vaisey’s dual process model leans heavily on Haidt’s
moral foundations theory [however, we will not dive into the works of Kurt Gray
and Jonathan Keeney (2015a, b), who claim that their cognitive research disconfirms
Haidt’s “MFT (moral foundations theory) on its own terms”].
While DiMaggio concludes that cognitive psychology lends support to the tool kit
theory, some of the evidence he shares contradicts contemporary representationalism
in appearance only. For instance, DiMaggio reports that supra-individual culture
both exists and exercises force in social life while being at the same time in many
ways “fragmented across groups and inconsistent across its manifestations” (p. 264).
In fact, much contemporary cultural sociology is predicated on the understanding
that representational culture is a combination of patterns and structures unequally
distributed and varyingly constitutive across publics on the one hand, and a reservoir
of symbolic noise, on the other (Mast 2019). Its cultural universe is ordered and
messy, its rate of change variable. Representationalists’ research practices involve
identifying variably-coherent symbolic forms and demonstrating how they contribute
to shaping collective processes. For instance, in more concrete terms, a representationalist
might identify the resurrection and renewed constitutive power of dormant
narratives of work and self-worth (Berezin 2019). Another representationalist
might specify how religious codes morph to accommodate new social arrangements
(Gorski  2019). When these representationalists demonstrate how these symbolic
forms and processes contribute to determining an election’s outcome and shaping a
nation’s trajectory, their research practices represent a kind of scientific success that
is of greater explanatory value than the phrase “interpretive studies” connotes.
As we turn to the cognitive cultural works in the next section, I offer a reminder
of the abstract nature of concepts, and present a question that permeates the remainder
of the argument. The abstract characteristic of concepts combines with our
claims to be representing empirical reality to produce a reifying effect on our interpretation
and use of the concepts themselves. This holds true for the concepts neuron,
schema, discourse, or collective representation, alike. A neuron is a cell in the
central nervous system. A cognitive schema, on the other hand, is a concept created
to represent a pattern of thought or behavior, which is akin to narrative being a
Representationalism and cognitive culturalism: riders on…
concept for representing patterns of representations. In the following works, schema
becomes our central concept of interest. We witness the reification occurring when
we examine the questions that we will be returning to, which include, what are the
forces and influences that shape a schema’s capacity to make substantive distinctions,
or that shape its dispositional tendencies? To what extent do signification and
symbolic representations influence the formation of schemas? DiMaggio arrives
at a sort of multidimensional assessment of the evidence, concluding that culture
“inheres not in the information, nor in the schemata, nor in the symbolic universe,
but in the interactions among them” (p. 274; emphasis added).
Part II: Contemporary cognitive culturalism: representationalism
and signification within
Omar Lizardo and the neurocognitive habitus
In his own writings and in collaboration with Michael Strand, Omar Lizardo has
worked from within Bourdieu’s theoretical framework to re-conceptualize the formation
and operation of that system of structured, structuring dispositions, the habitus.
The question I pursue in this section is what happens to signification and representationalism
through Lizardo’s efforts to conceptually reinforce habitus, first by
strengthening its cognitive foundations, and second by infusing it with neuroscientific
and cognitive linguistic theories.
Bourdieu was not subtle in his efforts to remove the late-Durkheimian and LeviStraussian
forms of representationalism from social theory’s models of social action
and order. He sought to exorcise Saussure’s “arbitrary nature of the sign,” or to displace
signification, with its meanings rooted in convention, and its mental concepts
determined by the internal play of signifiers, which ensure its relative autonomy
from social determinants. In its place, Bourdieu introduced a variety of cognitive
theory that conceptualized mental concepts as schemas as primarily shaped through
the body’s engagement with its material and social surroundings. Through these
interactions, the cognitive system internalizes the structural order of the external
world in “bodily schemas.” Shaped by the actor’s location in an arena of fields and
variable distributions of capitals, the dispositions or interpretive proclivities embedded
in an actor’s bodily schemas establish coordination between one’s mental activity,
action, and the exigencies and expectations generated by external structures.
Conceptually the habitus represents the cognitive internalization of the external
order. Meaning, while in a significantly reduced form, nonetheless continued
to serve a theoretical purpose in Bourdieu’s work. For instance, language would be
interpreted as serving the prevailing arrangements of domination: “the constitutive
power which is granted to ordinary language lies not in the language itself but in
the group which authorizes it and invests it with authority” (1977, p. 18). Symbols
would play a similar role: “symbolic power relations tend to reproduce and to reinforce
the power relations which constitute the structure of the social space” (1990a,
p. 135). These two quotations do not exhaust the fates of language and symbols in
	J. L. Mast
Bourdieu’s body of work, of course, but they do indicate a baseline against which
to interpret the theoretical movement signification and symbolic representations
undergo in the cognitive culturalist works discussed here.
Before turning to two of Lizardo’s works on the theme, here I briefly outline habitus’s
journey into neurocognitive terms, and give an indication of this journey’s consequences
for representationalism. (A) Through the concept bodily schemas, practice
is made the central component of cognitive development and activity, and the
semiotic variety of symbolic representations is decentered theoretically from meaning
and action. (B) Representations are then reintroduced to the theory as embodied
metaphors. In this theory of representation, grammar and semantic content, which
lend a sign system structure and enable it to constitute, are generated by a system
of motor schemes and motor operations, which are activated through the body’s
direct experience with its material and social surroundings. Through this process,
external structures become embedded in and made the constituent elements of bodily
schemas. (C) Activated in kind whether the actor is observing or performing an
action, mirror neurons ensure the transmission of the external structure’s social code
between persons with shared social relations and conditions. Put another way, during
social interactions, mirror neurons facilitate the consistent distribution of a cognitive
mapping that “socializes” actors in accordance with one’s station in the social
world’s distribution of institutions, field logics, and capitals.
In terms of the consequences for representationalism, through the introduction of
cognitive neuroscientific concepts, Lizardo alternately claims to have circumvented
the challenges posed by issues of signification and interpretation, on the one hand,
and to have simply moved representationalism further to the margins of theoretical
and empirical significance, on the other. Here we turn directly to Lizardo’s work on
habitus in order to unpack first how he bolstered the concept’s cognition-as-practice
dimensions, and second, to examine how he infused habitus with neuroscientific and
cognitive linguistic terms.
Though it is likely that he would not have anticipated the kinds of literatures
through which Lizardo reconstructs habitus, in his turn to Piaget, Bourdieu had
already taken the initial steps on this journey into cognitive psychology. Examining
Lizardo’s reconstruction of this development in Bourdieu’s conceptualization of
habitus will present a basic outline of how the concept is organized and indicate
the kind of work it does. With this established, we can consider the implications of
infusing it with cognitive neuroscientific theory, on the one hand, and the embodiment
theory of metaphors developed by George Lakoff and Mark Johnson (1999),
on the other.
Habitus is the intersection of two orders. One is external to the actor and consists
of the structures and institutions, and fields and capital distributions that pattern
the social. This order receives comparatively little attention in Lizardo and
Lizardo and Strand’s cognitive culturalism. The second is internal to the actor—
what others would call subjectivity—and it is this order, and the processes by
which it becomes ordered and gets transmitted, that Lizardo translates into neurocognitive
terms. This internal order is structured as well, and in a way that resembles
the structures outside. This resemblance contributes a sense of naturalness
and dimensions of flow to the practical experience of everyday life. Lizardo quotes
Representationalism and cognitive culturalism: riders on…
Bourdieu, and in a parenthetical draws our attention to the concept’s organization:
It is “the coincidence of the object structures and internalized structures, [notice
the reference to two types of structures] which produce the illusion of immediate
understanding” (Bourdieu 1990b, p. 26; quoted in Lizardo 2004, p. 379).
In his reconstruction of habitus, Lizardo explains how, by introducing into theories
of knowledge and mental concepts a robust dimension of practice, Piaget’s work
enabled Bourdieu to erode the theoretical significance of signification. Bourdieu
drew on Piaget’s developmental cognitive psychology to disrupt the prevailing models
of how mental concepts relate to practice, which had privileged the former and
emphasized their capacity to influence the latter. Piaget emphasized the interactive
relationship between bodily experience and mental concepts. He introduced the concept
bodily schemas to emphasize the notion that knowledge results less from passive
reflection on one’s surroundings than through experiencing one’s body acting
on and transforming reality.
The following quotation shows the theoretical steps taken to displace signification.
Noteworthy is the fact that Lizardo offers an interpretation of Piaget’s theory
of cognitive development that privileges practice but does not entirely dispense with
symbolic representations:
[C]ognitive structures are of primary importance in Piaget’s developmental
theory. However … Piaget’s primary emphasis was not on cognitive structures
as static symbolic representations, but on bodily schemas… and the operations
generated by way of these, through which the child is then able to transform
those representational structures into recognizable plans of action in the world,
and to acquire new cognitive structures from the feedback obtained from her
practical doings in the world. In this sense, Piaget considered knowledge to
be of a primarily operative nature, and of cognitive development as dictated
by the interplay of different structural systems, some bodily-motor, and some
symbolic representational (ibid., p. 384).
Cognitive schemas and the body become fused, literally, in the concept bodily
schemas. Symbolic representations are decentered from spaces in theoretical logic
dedicated to meaning and interpretation, foreshadowing theoretical maneuvers to
come. In these spaces, practice and cognitive schemas find new pride of place. Bodies
have boundaries, and limited ranges of motion and capacities for mobility. They
are only capable of physically interacting with the material and social worlds accessible
within these limitations (we must bracket for now how communication technologies
have extended these boundaries). While in this theory mental concepts are
derived through embodied interactions with one’s environment, what ends up getting
internalized nonetheless transcends one’s immediate surroundings.
As mentioned, habitus is the intersection of two structures. Incorporating Piaget’s
work in this fashion represents the internal structure as embodied schemas cultivated
through practice. If these specify form and process, then what kind of substantive
content gets cultivated into the schemas? (We return to this question in the next section,
when I discuss what shapes the elephant’s dispositions.) Lizardo argues that
Bourdieu “sociologizes the concept of internalized operations produced by reality.”
Bourdieu does this by
	J. L. Mast
giving Piaget’s skeleton of abstract reality the flesh of a sociological account
of the differential distribution of socially structured realities with which different
class fractions are faced. In this manner, he provides his conflict theory
with cognitive microfoundations that sidestep the problematic of order from
shared representations (pp. 387–388).
This adds, but does little to flesh out, the second dimension of habitus, the external
structure. Combined with the internal structure of bodily schemas, the concept habitus
becomes the site of “subjective harmony and objective coordination between the
internal and the external” (p. 389).
In the above descriptions, symbolic representations had retained a dimension of
causal capacity in the production of mental life. The theoretical fusion of external
and internal structures represented here, however, in conjunction with Lizardo’s version
of a “practical” Piaget, opens the possibility of dispensing with representational
structuralism altogether. Lizardo pursues the opportunity. He concludes that, “practical
action arises out of the operation of motor and operational schemes stored in
the socially produced cognitive [not Freud’s psychodynamic] unconscious, the true
repository of collective representations in the Durkheimian sense” (p. 389). What
has been revealed through this theoretical translation of cognitive theory into habitus,
Lizardo argues, is “the symbolic fallacy of interpreting cognitive structures in a
purely representational manner” (p. 388). Given the tenor of this theoretical turn to
practice, the adverb “purely” stands out as an interesting qualifier.
We have arrived at the cognitive habitus. Lizardo (2007) then turns to neuroscientific
theory to address the question of socialization, or the transmission of habitus
across the greater community of social actors, but with cognitive content substantively
patterned with the matrix of external structures appropriate to the actor’s
place in the distribution of capitals and embeddedness in fields (that is, in terms of
social space, the habitus must be both local and global). Lizardo also reintroduces
representation through a theory of metaphor that conceptualizes signs and systems
of signification as ultimately rooted in and derived from the body’s experience in the
material environment. Much as Piaget’s developmental psychology presented a way
of connecting cognition and practice, a theory of metaphors created by cognitive
linguist, George Lakoff, and philosopher, Mark Johnson, offers a pathway for fusing
meaning with embodied experience.
As we have seen, in this theory the cognitive habitus is not structured by and
therefore does not attend to signification of the relatively autonomous variety. The
habitus is a social phenomenon, however, that must be transmitted through interaction.
This means that its contents must be communicated to others via some
medium. Speech is not required, as the body itself, through unintentional movements
or patterned gestures, may serve as the medium. Indeed, as Lizardo notes,
“overt conduct,” even from the point of view of the cognitive habitus, “is fraught
with hidden ‘conceptual content’” (p. 332). That is, the body in practice signifies.
What remains, therefore, is the need for what amounts to a semiotic theory, or a
theory of the production, communication, and interpretation of signs.
In a move that enables him to avoid resorting to semiotic or representationalist
theories and methods, Lizardo turns to the work of Lakoff and Vittorio Gallese,
Representationalism and cognitive culturalism: riders on…
professor of physiology and cognitive neuroscience theorist, to argue that this content
“is easily graspable by our neurocognitive system” (p. 333). Yet even this does
not contradict the theoretical presuppositions of representationalism. This theoretical
step translates the signification process into a different lexicon, but the basic
process of a signifier being interpreted in some capacity by a receiver, remains the
same. The neurocognitive system is still “grasping” and distinguishing between
signs—a wink as opposed to an eye twitch (Geertz 1973b) that derive their meaning
from their positions within a system of representations.
Lodging this capacity in the neurocognitive system appears to circumvent the
subject or actor. What it does, rather, is describe the same phenomenon occurring
but on a different empirical level and in a different lexicon. That the neurocognitive
system is presented as the active agent does not detract from the overarching logic of
internalization, or eliminate the need for a theory of signification: Through interactions
with the social environment, the actor internalizes systems of representation
(variably and incompletely). In cognitive terms, this process shapes the cognitive
schemas lodged in the actor’s neurocognitive system, lending them particular dispositions
towards signs and their referents. In simpler terms, the process of transmission
still contains a signifying component, and some of what gets encoded in
cognitive schemas during the internalization process is dimensions of a relatively
autonomous sign system.
Therefore, to truly circumvent the tenets of reprsesentationalism, what is needed
is a theory of meaning that eliminates entirely any aspect of the arbitrary nature of
the sign. What is needed is a theory that grounds meaning not in convention but on
what Derrida (1967) called a transcendental signified, or a source that, by anchoring
all meanings—in semiotic terms: fusing signifier and signified, and thereby arresting
the deferral of meaning—enables the theorist to eliminate the issue of interpretation.
While in places Lakoff and Johnson sidestep this issue, Gallese and Lakoff
(2005) do not. As Lizardo puts it before quoting Gallese and Lakoff, the authors
“put the matter in strong terms”:
We … argue that conceptual knowledge is embodied, that is, it is mapped
within our sensory-motor system, [and] that the sensory-motor system not
only provides structure to conceptual content, but also characterises [sic] the
semantic content of concepts in terms of the way that we function with our
bodies in the world (Gallese and Lakoff 2005, p. 456, quoted in Lizardo 2007,
p. 333).
The sensory-motor system, in this formulation, is the transcendental signifier that
arrests the contingency of meaning.
Mirror neurons are what enable this theoretical move. When macaque monkeys
enter the text, we learn what mirror neurons are and what they do:
[I]t was discovered that neuronal activation increased in certain areas (i.e. F5)
of the ventral pre-motor cortex not only when the animal itself performed an
action on the object or was visually presented with the object, but when it perceived
another animal (or sometimes the experimenter) performing that action
on the object (Lizardo 2007, p. 329).
	J. L. Mast
Gallese builds on the mirror neuron phenomenon, which he argues demonstrates
that in human action “observation automatically triggers action simulation,” to
offer a theory of “embodied simulation.” He continues, asserting that this “process
of automatic simulation constitutes also a level of understanding, a level that
does not entail the explicit use of any theory of symbolic representation” (Gallese
2003, p. 523, quoted in Lizardo 2007, pp. 333–334). He calls this a “motor ontology.”
Embracing motor ontology would vanquish the arbitrary nature of the sign
from the model of social action. It would also problematize the transmission of
habitus, however. The theorist would face the choice of arguing that the external
social structures that are encoded into schemata via internalization are, in fact,
essentially natural and biological in nature, and consequently that external structures
are reflections of natural, biological imperatives. The other choice involves
acknowledging that in order for external social structures to be internalized, a
translation must take place in which they are rendered through and into a symbolic
code, which is then passed through words, actions, and materiality, to and
across social actors. This option, however, reintroduces the arbitrary nature of the
sign, and likewise moves the model back into the realm of a social, not a motor
ontology.
Our interest lies in how Lizardo incorporates this argument into his conceptualization
of the cognitive habitus and into cultural sociology, more broadly. At this
time, he interprets Gallese’s formulation in strong terms: “Conceptual content is
thus immanent to and irreducibly inseparable from—via the embodied simulation
mechanism—the overt actions of other social agents” (p. 334). This flirts with shifting
the research agenda from addressing processes and phenomena through the presuppositions
of a social ontology to those that undergird a motor ontology.
Lizardo’s interpretation of Gallese and Lakoff’s work indicates that he may
have been seeking to elide meaning and interpretation. Without embracing motor
ontology, however, signification and the tenets of representationalism sneak back
into the theory at the margins. Once this is acknowledged, the question becomes
what proportion of habitus formation and transmission, and of social action more
generally, will signification have to affect in order for strong cognitive practice
theory to incorporate it theoretically, and attend to it in empirical investigations.
One option is to treat it as an insignificant residual category, and when pressed,
resort to the theoretical logic of the base superstructure model. Lizardo equivocates
in the end. While he reiterates that Lakoff and Johnson’s concept of embodied
metaphors casts “the human body as the primary source domain for our tacit
understandings and explicit representations of most everyday life social situations
(and systematized bodies of knowledge)” (p. 346), he ultimately restores representationalism
to the theoretical playing field:
[M]ost recent neuroscientific evidence shows that the semantic/ practical
distinction, at the level of the neural structures that subserve the cognitive
representational capacities of the learning agent is not set in stone. In fact
“pragmatic” and conceptual representations of the world of objects and persons
interact in complex ways. In fact, the former sometimes “override” and
enrich purely semantic understandings of the object-environment (p. 345).
Representationalism and cognitive culturalism: riders on…
Steve Vaisey, the dual process model of culture in action,
and the representationalism within
In these two early articles, Lizardo drew on cognitive neuroscience to develop practice
theory in a more strictly anti-representationalist way. Per Bourdieu, the habitus
shapes social action, and Lizardo argues that cognitive neuroscientific theory and
empirical findings demonstrate that through the internalization of external structures
that are non-linguistic in nature, the habitus offers a model of the reproduction of
social order occurring through nonconscious mechanisms. Habitus, in this sense, is
similar to what Steve Vaisey will be calling practical consciousness (by way of Giddens)
in the article under consideration here. In his dual process model of culture in
action, Vaisey (2009) employs a similar theoretical logic, adds an empirical component,
and concludes that his findings substantiate the habitus model of social action.
To develop the dual process model, Vaisey turns to social psychologist Jonathan
Haidt, cognitive anthropologist Roy D’Andrade, and Robert Bellah and colleagues
work in Habits of the Heart.
Lizardo’s extension of practice theory through cognitive neuroscience was
impressively thorough in its attempts to corral and displace representationalism,
even if it sneaks back into the theory at its margins. Vaisey, too, argues against
representationalism, but in a more qualified way. While it ends up animating his
empirical investigation and findings and in a central and consequential way, Vaisey
nonetheless suppresses representationalist contributions to his study, and resists
integrating them into his theoretical conclusions.
Vaisey’s dual process model has been influential in legitimating the turn to cognitivist
models of cultural action. The theory represents the main reservoir of motives
and action as lodged in the automatic, non-deliberative domains of cognition. Cognitive
schemata located in neural networks are shaped through experience, and then
these “cultural-cognitive structures” act back on and shape an actor’s motives and
actions in a nonconscious way. Vaisey claims that the dual process model displaces
the rational pragmatism that animates Swidler’s approach because the tool kit theory
conceives of action as the product of deliberative, not automatic, cognition. Vaisey’s
claims regarding his findings’ implications for representationalism are more
complex. In this section, I analyze representationalism in Vaisey’s dual process
model, and conclude that while it “hides in plain sight” in his article, Vaisey nonetheless
fails to acknowledge its role or to pursue its significance in his theoretical
conclusions.
Vaisey’s argument is built atop four pillars. To argue that cultural action is not
shaped by deliberation or rationality, referred to as “discursive consciousness,” he
employs the theoretical logic that social psychologist Jonathan Haidt (2001, 2005)
used in assembling his “intuitionist” moral foundations theory. Vaisey’s sets out to
demonstrate that rational deliberation is “uninvolved in” (2009, p.  1690) cultural
action (Proposition 1, discussed below). Haidt theorizes that bio/psychoevolutionary
forces determine automatic cognition’s dispositions. While Vaisey employs Haidt’s
theoretical logic and method, he imports a different model of cognition, one called
“connectionist networks.” This is the argument’s second pillar. Vaisey gets the model
from D’Andrade (1995), who describes it as a mind-as-computer model designed to
	J. L. Mast
conceptualize the fundaments and dynamics of cognitive schemata. Vaisey argues
that the connectionist networks model explains how cognitive schemata are embedded
in neural networks in the brain and serve to shape action at the level of motives,
predilections, and moral dispositions. Giddens and Bourdieu represent the third pillar.
Vaisey argues that, interpreted through this cognitive conceptual framework, his
empirical findings substantiate these theorists’ models of social action, that of practical
consciousness and habitus. This conclusion, which suppresses representationalism’s
readily apparent contributions to his empirical model and testing, suggests an
affinity with Lizardo’s early theoretical work on habitus, detailed above. Regarding
representationalism, the fourth and final pillar in Vaisey’s study is Robert Bellah
et al. (1985), and the research team’s empirical findings in Habits of the Heart. Bellah
et al. play a vital methodological role in Vaisey’s argument, and they deliver an
unacknowledged but consequential twist to the plot.
In terms of my main point, which is that cognitive cultural work is suffused with
unrecognized or suppressed representationalism, the Bellah et al. pillar is the main
site of action in this section. In the following, I illustrate how, in turning to Giddens
and Bourdieu, Vaisey brackets “meaning” theoretically, and replaces it with practical
experience. I argue that this theoretical conclusion does not flow logically from
his empirical model and findings. To the contrary, I demonstrate how, through incorporating
Bellah et al.’s (1985) findings into his empirical measurement tool, Vaisey
reintroduces signification to his study and in a central and consequential way, and
thereby ends up substantiating a representationalist theory of culture in action.
First, however, I discuss Jonathan Haidt’s (2001) work on moral intuitions. Intuition
is a central concept in what Haidt calls “moral foundations theory,” which
raised the profile of the “dual process model of cognition” considerably, due in no
small part to the social psychologist’s ventures into the crossover nonfiction market
and mainstream media platforms. Vaisey builds on this dual process model and
adds a cognitive twist. Given the extent to which Haidt’s theoretical logic animates
Vaisey’s argument (see Pugh 2013 for a similar observation), a close look at Haidt’s
presuppositions about the nature of social action and meaning are in order. (See
Gray and Keeney 2015a, b, in which these cognitive psychologists argue that cur-
rently their research is “disconfirming moral foundations theory on its own terms.”)
Haidt (2001) intervened in the literature on morality by arguing that its prevailing
consensus was predicated on a model of deliberation and rational action that his
research was proving untenable. Intuition and reasoning make up two important but
contrasting kinds of cognition, he points out (ibid., p. 818). Morality is experienced
as “moral intuitions,” Haidt argues, wherein intuition is rooted in the uncritical and
unreflexive arenas of automatic cognition, and not in those of deliberative cognition,
where judgment and reasoning take place. This is the basis of moral foundations
theory. Most cultural sociologists share at least one aspect of this theoretical orientation—in
terms of presuppositions about action, the cultural sociological consensus
conceives of morality in nonrational terms. Many will disagree with Haidt, however,
regarding the wellsprings and contents of moral categories.
In a later work, Haidt (2005, 2012) introduces the metaphor of a rider on an elephant
to illustrate his model. This rider on an animal metaphor has a long history.
Vaisey (2009, p. 1683) employs it, and agreeing that it is a useful heuristic, I use it
Representationalism and cognitive culturalism: riders on…
in this article. Now, Haidt’s cognitive version of the metaphor: A social actor experiences
the relationship between her mental and active physical life as if she were
a rider on an elephant. Social action feels like it consists of the rider reflecting on
one’s interests, desires, and moral boundaries, and then commanding the elephant to
transport oneself through social situations accordingly. Haidt tells us that this is not
how moral or deep-cultural action works. The rider represents deliberative cognition.
She has weak cognitive capacities, ones that are limited to controlled, reflective,
and “language-based reasoning” (2012, p. 54). The elephant, on the other hand,
is where emotion and intuition are stored and processed. Haidt describes it as a reservoir
of strong cognitive capacities and entrenched proclivities, ones that developed
over millions of years in accordance with evolutionary biological imperatives. Thus,
the elephant is both the engine and the true driver of this pair. “When human beings
evolved the capacity for language and reasoning at some point in the last million
years,” Haidt explains, “the brain did not rewire itself to hand over the reins to a new
and inexperienced charioteer. Rather, the rider (language-based reasoning) evolved
because it did something useful for the elephant” (ibid., pp. 53–54; parenthetical in
original). The elephant is the source of moral intuitions, and the content of moral
categories derive from the elephant’s needs and experiences.
One final note on moral foundations theory before turning to Vaisey’s version:
Haidt critiques the deliberationists on two grounds. First, they are misguided for
assuming that morality and moral action arise from reason and deliberation. Their
second mistake flows from the first. Since the deliberationists hold that moral experience
is rooted in judgment, they look to actors’ discursive justifications for morality’s
form (capacity) and content (dispositions). But since morality is guided not by
the rider (who deliberates) but by the elephant, or the non-discursive, non-reasoning,
intuitive part of the cognitive binary, the deliberationists’ method of looking at how
people talk about their moral choices is entirely wrong.
Vaisey puts Haidt’s model to good use. Haidt took aim at the rational deliberationists
in the morality literature. Noting similar assumptions in Swidler’s formulation
of culture in action, Vaisey applies Haidt’s logic of argumentation to tool
kit theory. Building on this logic, Vaisey (2009, p. 1680) argues that in eschewing
Weberian, Parsonian, and Geertzian theories of meaning and action, and in
emphasizing that culture consists of strategic use of tools and repertoires, Swidler
thereby commits herself to associating cultural action (note how cultural action
takes the place of moral experience4
) with the deliberative cognitive processes of
rational and reasoned reflection. Translated through Haidt’s metaphor, the tool kit
theory presumes that the rider is in control when in actuality the elephant is the
pair’s true wellspring of action. It is important to point out here that while Vaisey
4
 While Haidt discusses morality broadly, he remains focused on decisions and actions that explicitly
have a moral component, and for the most part he sticks to using terms such as “moral intuitions.” Vaisey
uses the terms moral-schemas and moral dispositions, but his study purports to speak to the broader concept
of culture in action, as both the role Swidler’s theory of cultural in action plays in his study, and the
title of his article, suggest. In this article, I treat the main subject of Vaisey’s study as the broader of these
two domains, or as culture in action.
	J. L. Mast
includes an elephant in his act, it is not of the same cognitivist breed as Haidt’s.
As mentioned, the latter’s is a combination of bio/psychoevolutionary theory and
embodied schemata, whereas Vaisey’s is born of the mind-as-computer metaphor,
represented by the concept connectionist networks. I discuss this in greater detail
below.
In an “empirical illustration of the model,” Vaisey outlines two propositions his
research model is designed to examine.
Proposition 1.—Because discursive consciousness is largely uninvolved in
routine moral decision making, interview respondents will either (a) tend to
explain their behavior in intuitive terms without a clear substantive referent
or (b) offer multiple “loosely coupled” logics of justification to support their
judgments
Proposition 2.—Because the practical consciousness or habitus will tend
(probabilistically) to generate a response that is consistent with its underlying
schematic organization, respondents’ forced choice of a cultural script will be
predictive of their future morally relevant behavior, even when other factors
are held constant (p. 1690; italics in original).
One thing to note is that many of the conclusions that Vaisey aims to draw about
theory and cognitive concepts are already central components in this “empirical”
illustration of the model.
Vaisey states that his “proposition 1” is designed to demonstrate that the rider,
discursive consciousness, is uninvolved in moral decision making (p.  1690). He
acknowledges this is difficult to demonstrate (p. 1690), but he draws on interview
data to indicate that the respondents’ answers were consistent with the proposition
1 (p. 1691). That is, the interviewees had a hard time explaining the moral logics
of their decision making. For instance, in an interview a young woman equivocates
over the morality of a terminally ill person using tobacco and cannabis, and she fails
to “provide any compelling reasons” for asserting that ultimately such actions would
be wrong (1695). Vaisey provides this as an example of discursive reasoning’s
absence in moral judgment. That the woman comes to a determination for which
she cannot vocalize a rationale is suggestive of how practical consciousness controls
moral reasoning. We will return to this example below. Our main interest, for now,
lies in proposition 2.
If the findings from proposition 1 indicate that the rider is not guiding moral
action but rather that the elephant is in control, then how does one go about unpacking
the elephant and determining what is shaping its dispositions? Taking cue from
Haidt (2005, 2012, p. 47), Vaisey proposes that interviews draw forth fragmented
narratives that interviewees create in their discursive consciousness. These are posthoc
rationalizations created in moments of intentional reflection, which is a cognitive
process that is “uninvolved” in moral or cultural action. Fixed response surveys,
on the other hand, ones that force respondents to choose answers even if they do
not feel strongly about the options, are better at ascertaining the workings of practical
consciousness (automatic cognition), or the moral intuitions represented by the
elephant, Vaisey argues.
Representationalism and cognitive culturalism: riders on…
Here we need to focus in on how Vaisey uses Bellah et al.’s (1985) empirical
findings, and pay particular attention to this research team’s methods and the role
language played in their study.
•	 Vaisey states that “interview methods engage with discursive consciousness
alone” (p. 1687).
•	 He states that his proposition 1, “discursive consciousness is largely uninvolved
in routine moral decision making” (p. 1690), has been demonstrated by existing
research, citing Bellah et al.’s (1985) Habits.
•	 Bellah et al.’s principal research method in Habits was the active interview.
•	 Therefore, per Vaisey’s definition of cognitive domains and the strict boundary
separating their functions, Bellah et al. gained their findings by accessing deliberative
cognition alone.
Vaisey’s goal is to demonstrate that deliberative cognition is uninvolved in shaping
cultural action. Nonetheless, he incorporates Bellah et al.’s findings, which were
derived through accessing deliberative cognition, into his survey tool, which is supposed
to be designed to access automatic cognition or practical consciousness.
As Bellah et al. (ibid., p. 304) explain in the Habits appendix:
“We conceived of our research from the beginning as a dialogue or conversation
with fellow citizens…,” and they add that “we sought to bring our preconceptions
and questions into the conversation and to understand the answers
we were receiving not only in terms of the language but also… in the lives of
those we were talking with.”
Bellah et al. describe their method in terms synonymous with deliberation: “The
interview as we employed it was active, Socratic.”
Regarding language, Bellah et al. explain that their findings about Americans’
moral dispositions were derived through methodically privileging signification:
“[W]hat we were interested in above all was the language people used to think about
their lives and the traditions from which that language comes.” The authors argue
that language contributes significantly to the construction of reality: “We believe…
that the mobile middle classes define reality for most of us in the United States.”
Bellah et  al. conclude that their results demonstrate that language and reasoning
are the strongest forces shaping moral dispositions: “We think our interviews
have allowed us to describe the most influential forms of middle-class language and
moral reasoning about private and public life in America today” (ibid., p. 306).
To sum up, Bellah et al. privilege signification and discursivity in their research
design, and they emphasize their explanatory power in their results. Despite this,
Vaisey not only argues that Habits substantiates his proposition 1—“discursive consciousness
is largely uninvolved in routine moral decision making.” He then uses
Bellah et al.’s results to assemble the central measurement tool for testing his proposition
2:
To illustrate proposition 2, I use a single variable to measure the respondents’
most accessible or salient moral schema… I rely on a question designed to
	J. L. Mast
mirror the moral typology developed in Habits of the Heart (Bellah et al., p.
1985). This typology includes four overarching moral logics….
His single-item measurement tool is “not ideal… as a measure of moral schemas,”
he concedes. Nonetheless,
As a single item, it is well matched to the four moral schemas outlined in Habits
of the Heart and was explicitly designed to measure them” (p. 1691; emphasis
added).
When Vaisey’s forced choice survey returns multiple significant coefficients, suggesting
that the tool is good at predicting moral behavior (pp. 1699–1702), Vaisey
concludes that the measurement tool has captured “something significant about the
respondents’ internalized moral-cultural schemas” (p. 1703). Likely it has. The conceptual
and theoretical conclusions Vaisey draws, however, do not flow logically
from these results.
While Vaisey’s (and Haidt’s) research helps unsettle one-dimensional rational
deliberationist explanations of cultural action, considerable interpretive work
remains to be done in order for Vaisey to turn his survey results into evidence demonstrating
that cultural action is most accurately depicted by theories of practical
consciousness and habitus. Vaisey turns to D’Andrade for tools to build the argument.
In the work Vaisey consults, D’Andrade (1995) gives a thorough overview
of the roots, branches, and (at the time) contemporary developments in cognitive
anthropology.
Vaisey finds that schema theory and “connectionist networks” (D’Andrade 1995,
pp. 138–149) best fit his needs: “In this view, cultural-cognitive structures are built
up out of experience and allow a person to respond to stimuli in ways that are automatically
generated by the weighted connections between the elements of the inputs
at hand” (p. 1686). To unpack this a bit, schemata are connected to one another, and
they are located in neural networks in the nervous system. They are described as
being shaped by interactions with forces and objects outside of the person. These
interactions shape schemata in the sense of giving them interpretive preferences and
dispositional proclivities. Once these are encoded in schemata, then when the social
actor is navigating the world and encountering “stimuli,” these schemata activate
within the actor’s brain and persuade it to interpret and interact with the environment
in a manner consistent with the schemata’s coded dispositions. This returns
us to the question of what forces shape the proclivities of the elephant? If we know
the answer to this question, then we have a theory of what causes social action. We
know Haidt’s answer.
Vasey’s description suggests that it is local physicality and direct interactivity
with objects and social relations that shape the elephant’s preferences. As we saw
above, in his early work, Lizardo argued that the local environment, both social and
material, (and I here I switch to my preferred terms) is coded with signs signifying
external social structures and relations. Lizardo moved representationalism,
or culture as a patterned, relatively autonomous sign system, to theory’s margins.
Does Vaisey conclude similarly? Are the meaning patterns embedded in language
and expressed through talk things that social actors experience in their everyday
Representationalism and cognitive culturalism: riders on…
lives (or in childhood development) and would thus be included in the category of
experiences that shape schemata? Vaisey’s answer is no: “schemas are nonlinguistic
cognitive constructs,” he asserts. With language and signification bracketed, Vaisey
sums up the work this conceptual model performs in his theory. He argues that his
empirical findings, embedded in the conceptual lexicon of cognitive connectionist
networks, demonstrate that “’the elephant’ provides a validated mechanism for
understanding an important way society can ‘get into’ human beings that is homologous
with Giddens’s practical consciousness and Bourdieu’s habitus” (p. 1684).
Let us briefly return to the example mentioned above regarding the young woman
who struggled to explain why she felt it is immoral for a terminally ill person to use
tobacco and cannabis. That question stems from the interview portion of Vaisey’s
study. Here we turn to the survey questionnaire. As established, Vaisey asserts that
the questionnaire is better able to access the moral reasoning that takes place in automatic
cognition or practical consciousness. Its dispositions are unperturbed by linguistic
or representational signs. The survey (pp. 1708–1709) asks respondents questions
about frequency of marijuana use and alcohol consumption (“not including at
religious services”), cheating at schoolwork, skipping school, and other behaviors in
this category of things. In order to arrive at his theoretical conclusions, the survey
questions must represent something about morality, and they must access dispositions
that reflect “nonlinguistic cognitive constructs.” Therefore, Vaisey is arguing that the
objects and actions specified in the questionnaire reflect the category of “moral,” and
he is asserting that these objects and actions bear no dimensions of representationalism,
no aspects of signification. These survey questions, however, are mini-narratives
replete with collective representations that are themselves embedded in moralizing
discourses, to which the question’s caveat, “not including at religious services,”
attests. The symbol “drug” itself is a coded boundary fixed by a variety of discourses.
Medical, legal, social justice, pop cultural, religious; each discursive domain asserts a
moral classification of “drug” substances, and specifies narratives representing how,
when, why, etc. such substances may be used appropriately, or in their own sense of
the term, “morally.”5
Much as Vaisey resists acknowledging the representationalism
that Bellah et al. bring to his study, so too does he fail to address these additional
dimensions of representationalism that pervade his survey questionnaire.
Furthermore, the survey evidence does not substantiate a connectionist network
model of cognition. It cannot. Discussing it in detail is fine to the extent that Vaisey
presents this part of his narrative as an act of theorizing. Additionally, because
Vaisey’s measurement of automatic cognition was based on a tool created through
5
  Currently, this drug classification system is in flux. Tobacco, deeply constituted by narratives of health,
class, and freedom [to versus freedom from], for instance, is becoming increasingly polluted. The scent
of “marijuana” on an urban, suburban, or almost any other street in the USA, on the other hand, would
no longer signify the “dirt,” or social categories of peoples deemed dangerous or “matter out place,” that
it would have even a decade ago. The signifier “marijuana” itself is also being interpreted as polluted
due to the role the sign played last century in the domination of racial and ethnic minorities, and is being
replaced by the term “cannabis.” To return to our main point: If we were to issue the same questionnaire
today and interpret its results faithfully through Vaisey’s theoretical representation of cognitive schemas,
then to the extent that the questionnaire detects shifts in dispositions, we would be led to conclude that
the changes were the product of an accumulation of experiences minus, or unperturbed by, any linguistic
or representationalist elements.
	J. L. Mast
thoroughly deliberative means, it is inaccurate to claim that his measurement device
tapped into and assessed automatic but not discursive cognition. Even if this could be
demonstrated (as if the tool could surgically distinguish between the two domains),
it is more realistic to conclude that automatic and deliberative cognition have a more
complex and interactive relationship than Vaisey’s presentation of the cognitive model
suggests [“interview methods engage with discursive consciousness alone” (p. 1687)].
Finally, the concept conveyed by “automatic cognition” plays an important role
in the theories of action that animate contemporary representationalism. Signs and
collective representations are internalized during cognitive development, however
unevenly and variably, and thus shape the dispositions embedded in cognitive schemata.
Nonetheless, here as elsewhere in cognitive cultural work, Geertz is cast to
play the foil, and his work is caricatured as insisting that culture is “perfectly shared
or perfectly internalized by all members of a given society” (p. 1686). Needless to
say, Vaisey concludes that his findings do not support such a version of culture in
action. Bellah and Geertz were both followers of Talcott Parsons, and both of them
distinguished themselves from Parsons, and at around the same time, by advocating
for conceptualizing culture in more autonomous terms (Bellah 1970; Geertz 1966).
In fact, Vaisey does acknowledge that representationalist concepts contribute to
shaping automatic schemata, but he does so only in a cursory way. For instance, he
quotes D’Andrade, the same cognitive anthropologist from whom he borrowed the
connectionist networks model of cognition, as asserting that the cultural “shaping of
emotions gives certain cultural representations emotional force, in that individuals
experience the truth and rightness of certain ideas as emotions within themselves”
(Vaisey 2009, p.  1686, citing D’Andrade 1995, p.  229, emphasis in original).6
6
  Vaisey is accurate when he asserts that D’Andrade, in his presentation of the connectionist networks
model, emphasizes that experience is a constituent element of culture and downplays the role of signification.
Yet it is also clear that D’Andrade emphasized experience because he believed that the state of
theory in anthropology at the time was in danger of succumbing to “extreme idealism” (p. 149) and of
embracing “epistemological relativism” (p. 148; italics in original), a development he believed would
make “objectivity and science impossible” (p. 148).
  To explain his concern, D’Andrade (1995, p. 148) shares a quote from Marshall Sahlins (who was quoting
Leslie White) in which Sahlins advocated in too strong of terms for what D’Andrade interprets as an
absolutist form of representationalism:
Thus [with symbols] man built a new world in which to live…. Between man and nature hung the
veil of culture, and he could see nothing save through this medium … permeating everything was
the essence of words.
  The veil argument, D’Andrade explains, “comes from the idea that language and other symbol systems
determine what we experience” (p.  148–149; emphasis in original). While he clearly chafes at what
he interprets as epistemological over-determination, he nonetheless makes it explicit that rejecting the
importance of language and signification altogether is not a tenable stance, either. Instead of simply dismissing
representations, and in order to establish a causal hierarchy, D’Andrade performs a nifty theoretical
maneuver that subsumes language to experience, arguing that
[in] the connectionist model ‘words’ do not ‘encode’ experience. Rather, words signify schemas,
which means that the units activated by a particular speech sound also activate a larger pattern of
connections which are the active schema for a particular experience. The sounds of words are like
‘pointers’ to patterns of experience—indices to internal mental structures, not ‘veils’ between reality
and experience (p. 149).
Representationalism and cognitive culturalism: riders on…
Despite this, and despite what I have demonstrated above, namely that representationalism
pervades his empirical analysis and findings, in his conclusions, Vaisey
excludes signification and representationalism from his theory of dual process cultural
action.
John L. Martin’s interpretation of cognition’s limits
The conclusions Vaisey (2009) drew from his dual process intervention were
promptly taken up by cognitive culturalists and referenced as representing not theoretical
interpretations but empirical facts. Anti-representationalism in these subsequent
works remained a central animating ethos. Geertz, and those building on his
semiotic and hermeneutic framework, such as Alexander and the strong program,
remained the central foil.
John Martin (2010) mounted the most forceful attack on representationalism.
Martin argues that discoveries in the cognitive neurosciences demonstrate that the
theory of social action predicated on “culture as a web of signification” is untenable.
Human cognition cannot internalize a complex system of representations. It does not
have the capacity to store the system. And if it could, it could not draw upon it to
inform everyday experience in a manner efficient enough to facilitate action. Given
the “evidence pointing to our cognitive limitations,” Martin (2010, p. 230) states,
the theses that “culture as a complex web of meaning and culture as inside the minds
of actors—cannot both be correct, for the simple reason that our minds are not good
at holding lots of connected things in them” (p. 229).
If this were true, it would indeed be unwelcome news for representational cultural sociologists.
Martin’s is a misrepresentation of some of his most significant cognitive sources,
however, and his thesis—that limited cognitive capacities mean social action cannot be
shaped by signification—is contradicted by some of the very same sources he cites.
For instance, he turns to George Miller’s (1994 [1955]) “justly famous piece” in
which the early pioneer of cognitive psychology reviewed experiments that used
tone, loudness, taste, and visual stimuli to examine the limits of short term memory
information processing.7
Results showed that people became confused when trying
  So, while he rejects the absolutist-representational assertion suggested by the veil metaphor, D’Andrade
nonetheless maintains that symbols remain an important dimension in cognitive anthropology’s future theories:
“By changing the model of human cognition and meaning from a system containing only symbolic
serial processing to a system containing both symbolic serial processing and connectionist parallel distributed
processing, a number of things about human culture look different” (ibid., emphasis in original).
Footnote 6 (continued)
7
 Given the theoretical and methodological stakes of Martin’s argument, cultural sociologists will be
motivated to familiarize themselves with the kinds of methods and findings on which he is basing his
assertions. Miller’s argument is based on a meta-analysis of experiments in information processing. The
following is a sample of some of the experiments Miller drew on to arrive at his conclusions.
  Miller describes Irwin Pollack’s (1952) uni-dimensional experiment this way: “Pollack… asked listeners
to identify tones by assigning numerals to them. The tones were different with respect to frequency,
and covered the range from 100 to 8000 cps in equal logarithmic steps. A tone was sounded and the
listener responded by giving a numeral. After the listener had made his response, he was told the correct
identification of the tone” (Miller 344).
	J. L. Mast
to distinguish between and hold a certain number of bits of information in their
minds, namely “the magic number seven, plus or minus two.” Presented out of context,
the number seems quite small. Read Miller’s article through to the end, however,
and it becomes clear that humans have developed methods (p. 348) and tools,
like systems of representation (349–350), that enable them to store and use highly
complex structures of information.
Miller breaks information processing into two categories, “absolute judgment”
and “immediate memory,” and he divides stimuli into “bits of information” and
“chunks of information” (p. 349). While the processing of either is limited to around
seven units, humans extend their cognitive capabilities far beyond those insinuated
by Martin by combining bits into chunks, and by building ever more complex combinations
of bits into assemblies of chunks. In Miller’s words,
Since the memory span is a fixed number of chunks, we can increase the number
of bits of information that it contains simply by building larger and larger
chunks, each chunk containing more information than before (p. 349).
This process is called recoding. Miller explains:
The point is that recoding is an extremely powerful weapon for increasing the
amount of information we can deal with. In one form or another we use recoding
constantly in our daily behavior… [T]he most customary kind of recoding
that we do all the time is to translate into a verbal code (p. 350).
Our language is tremendously useful for repackaging material into a few
chucks rich in information (p. 351).
Miller summarizes his argument thus:
First, the span of absolute judgment and the span of immediate memory
impose severe limitations on the amount of information that we are able to
receive, process, and remember. By organizing the stimulus input simultaneously
into several dimensions and successively into a sequence of chunks, we
manage to break (or at least stretch) this informational bottleneck (p. 351).
Footnote 7 (continued)
  Miller describes an auditory study done by Pollack and Picks (1954). “They managed to get six different
acoustic variables that they could change: frequency, intensity, rate of interruption, on-time fraction,
total duration, and spatial location. Each one of these six variables could assume any one of five different
values, so altogether there were 56, or 15,625 different tones that they could present. The listeners made
a separate rating for each one of these six dimensions. Under these conditions the transmitted information
was 7.2 bits, which corresponds to about 150 different categories that could be absolutely identified
without error. Now we are beginning to get up into the range that ordinary experience would lead us to
expect” (Miller p. 347).
 Miller describes Hayes’s (1952) multi-dimensional stimuli experiment this way. Hayes tried out “five
different kinds of test materials: binary digits, decimal digits, letters of the alphabet, letters plus decimal
digits, and with 1000 monosyllabic words. The lists were read aloud at the rate of one item per second
and the subjects had as much time as they needed to give their responses. A procedure described by [R.
S.] Woodworth was used to score the responses” [Miller 1994 (1955), p. 349].
Representationalism and cognitive culturalism: riders on…
That is, we break the cognitive limitations that Martin presents as immutable. Miller
continues,
Second, the process of recoding is a very important one in human psychology
and deserves much more explicit attention than it has received. In particular,
the kind of linguistic recoding that people do seems to me to be the very lifeblood
of the thought processes (p. 351).
Martin (2010) cites Miller’s justly famous piece as evidence that social action is
not influenced, indeed cannot be influenced, by signification or systems of representation.
Language is the quintessential form of signification. Miller concludes that
linguistic recoding is the most important aspect of cognition.
Martin (2010, p. 231) then turns to psychologist Nelson Cowan (2000) to support
his claim about cognition’s limits. Forty-five years after Miller’s publication, Cowan
examined the latest evidence and revised the limit from seven down to four. This
number is tremendously misleading if presented out of context, however.
Cowan’s aim in this piece is to establish the boundary conditions necessary to
accurately observe the limits on information processing capacity. That is, Cowan
pursues the question of what rules must be followed in order for experiments on
chunk processing to be valid. One condition that he specifies reiterates Miller’s
(1994 [1955]) conclusion about language’s centrality to cognition. It also refutes
Martin’s (2010, p.  232) characterization of long-term memory. After presenting
musings on long term memory’s tendency to produce “networks of concepts and
ideas,” which oddly reads like a semiotic theory of memory encoding and recall,
Martin argues that a “webbiness” aspect of culture cannot influence action because,
due to the long term memory component of cognition, “we can’t keep things unconnected”
(ibid.).
Cowan states that, in addition to three other conditions, accurate readings of
chunk processing capacities may occur when “steps are taken specifically to block
the recoding of stimulus items into larger chunks” (p.  87). Cowan is stating that
cognition seeks to recode stimuli into chunks, and that if cognition recognizes a
familiar chunk in the measurement tool, it will automatically intuit a distinction and
separate the chunk from the other signs in the presentation or flow of stimuli. That
is, signification shapes cognitive schemata; the sign system makes cognition “see”
distinctions.
Whereas Martin (2010, p. 232) argues that “a great deal of the problem we have
with our parsimonious memory system seems to be in stopping the flow of associations,”
Cowan argues to the contrary that familiar signs force distinctions between
stimuli and thus, because such signs move from memory to the foreground of awareness
as if instantly, they corrupt the measurement of information processing. As
Cowan (2000, pp. 89–90) puts it,
It would be assumed that the number of chunks can be estimated only when
inter-chunk associations are of no use in retrieval in the assigned task. To use
a well-worn example inspired by Miller (1956), suppose one tries to recall the
series of letters, “fbicbsibmirs.” Letter triads within this sequence (FBI, CBS,
IBM, and IRS) are well-known acronyms, and someone who notices that can
	J. L. Mast
use the information to assist recall. For someone who does notice, there are
pre-existing associations between letters in a triad that can be used to assist
recall of the 12-letter sequence.
Cowan continues:
If we further assume that there are no pre-existing associations between the
acronyms, then the four of them have to occupy limited capacity storage separately
to assist in recall. If that is the case, and if no other optional mnemonic
strategies are involved, then successful recall of the 12-item sequence indicates
that the pure capacity limit for the trial was at least four chunks. (In practice,
within the above example there are likely to be associations between the acronyms.
For example, FBI and IRS represent two U.S. government agencies,
and CBS and IBM represent two large U.S. corporations. Such associations
could assist recall. For the most accurate pure capacity-based limit, materials
would have to be selected so as to eliminate such special associations between
chunks) (p. 90; italics added).
Cowan is asserting, in effect, that systems of signification are deeply embedded
in memory and intertwined with recall; in Miller’s words, we can think of “memory
as the handmaiden of discrimination” (p. 348). Miller’s and Cowan’s examinations
show that signs combine to create increasingly complex chunks of information—
we could call these complex chunks “codes,” “icons,” “collective representations,”
“narratives,”—and this kind of recoding occurs in human cognition as if automatically.
Signification shapes the interpretation of stimuli by specifying distinctions
and transmitting them into (variable degrees of) awareness as if through uninhibited
reflection.
Martin interprets works such as Miller’s and Cowan’s otherwise. He concludes
that they substantiate the assertion that “[c]ulture may be complex, but that means it
is not in our heads” (p. 240). He combines this with a reference to Vaisey (2009) to
offer the methodological conclusion that “the last thing we should do is to conduct
in-depth interview with a selection of informants, any more than we would expect
to strike gold by asking them for whatever change is in their pockets” (ibid.). In
representationalism, symbol systems are modeled on language. Gallese describes
language as “the cognitive tool of abstraction par excellence” (2003, p. 518). As
we have seen, Miller argues for the centrality of language in cognition, and Cowan
asserts that signification shapes cognition.
In one final example of how cognitive psychology’s research findings can be
interpreted in radically different ways, I turn to another reference in Martin’s case
against representationalism. Martin (p. 231) cites the work of Intraub et al. (1996)
to further his argument that memory is incapable of internalizing a complex, structure-like
system of collective representations. In this article, however, Intraub et al.
demonstrate not only that human memory is composed of “abstract representations”
but that cognitive processing depends on the “top-down” incorporation of such representations
in order for it to connect partial views during visual scanning.
Citing three studies, the authors state that “there is growing evidence that an
abstract, nonsensory spatial representation (a memory structure) plays a role in
Representationalism and cognitive culturalism: riders on…
the integration of eye fixations” (p. 120, emphasis added). The authors explain that
“perceptual schema” create continuity during visual scanning (119). A perceptual
schema “is a mental representation of the likely structure of the scene that is understood
to ‘exist’ just beyond the edges” of a picture’s boundaries. In wording that
suggests cognition operates in ways very much like the hermeneutic circle, the scientists
explain that such schema allow “the viewer to understand the partial view
within a larger context.” Visual apprehension of a picture, they continue, “reflects
not only the actual bottom-up information that has been presented, but highly probable
top-down information from the schema” (119, emphasis added). The authors
conclude that their work speaks not just to issues of memory but to “the type of
processes that take place during scene perception” (p. 132). Their work indicates
that memory involves storing abstract representations, and that cognitive processing
involves bringing forth these “abstract, nonsensory” representations to make sense
of the “bottom-up information” that is visually present. In their words, “an abstract
representation of the expected layout of a scene aids in comprehension and integration
of… partial views, thus giving rise to the viewer’s experience of a continuous
visual world” (p. 132).
In sharp contrast to Martin’s characterization, we conclude this section with a theory
of how signification and systems of representation helped to give rise to human
cognition.  After he (1995) published his influential volume, The Development of
Cognitive Anthropology, Roy D’Andrade (2002) radically resituated language in his
theory of cognition and culture by arguing that, as a product of human evolution,
cognitive schemas are inextricably intertwined with signs systems and practices of
signification.
D’Andrade (2002) argues that culture is a “selective factor in human evolution,”
by which he means “that our bodies and our psyches have been affected by a past
history of living a cultural way of life.” He continues:
Having a body and a certain kind of psyche, with certain inbuilt emotions,
desires, and cognitive skills, has been selected for because we have been living
in a cultural world—a cultural niche—for millions of years. In rather different
sense of the term than is usually used, it can be said that humans are, via evolution,
‘culturally constituted’” (p. 223).
D’Andrade explains that, just as the initial capacity to make and use a tool “selected
for the specialized human ‘precision grip’ – the way one holds a screwdriver” (i.e.,
tools contributed to shaping the evolution of the human body), so too did language
and symbol systems select for specialized humans with “precision grips” of representational
capacities for constituting, signifying, and interpreting.
Grammatically and semantically complex language is a cognitively demanding
and costly capacity to cultivate and participate in, he points out. Why then did
humans and humans alone develop it? D’Andrade reasons that
[A] cultural way of life increased the fitness of individuals who were more
effective at producing and understanding representatives because hominid
band life became organized in such a way that information outside the immediate
world of the here and now came to have important value. According to this
	J. L. Mast
argument, a strongly cultural way of life, involving learned age and sex roles
with complex learned routines, performed at different times and in different
places, fragmented [emphasis in original] the unitary here-and-now togetherness
of the primate band, at the same time requiring ever more coordination by
means of environmental and social information concerning things not present.
In such a world, displaced linguistic representations became the glue that held
together the intersubjective world of the band (p. 226, emphasis added in final
sentence).
Language and symbol systems proved extraordinarily valuable, he argues. They
allowed knowledge to be shared between persons, and across and over generations.
As its value increased, so did the advantage of having a brain capable of processing
and storing language’s increasingly complex rules and contents.
It was not increases in brain size that enabled the emergence of the human capacity
to create and store language, D’Andrade explains. Rather, it was language that made
“possible an advantage in having brain structures large enough to store hundreds of
thousands of items because it makes learning from other brains effective and efficient”
(p. 227). Language was a causal factor in not only producing the comparatively large
human brain (p. 227), it also “significantly increased specializations of cortical and
subcortical areas” (p. 228). D’Andrade, we should note before moving to the next section,
asserts that cognition is capable of storing an enormous amount of information.
Signs systems caused the development and elaboration of complex cognitive
and embodied schemata in humans, D’Andrade argues. Their content, not just their
form, shaped brain and body evolution. Signs and complex symbol systems (the
“glue” of the intersubjective world), cognitive schemas, and bodies, are inextricably
intertwined.
Conclusion
Summing up, D’Andrade argues that cognition and representationalism are deeply
interconnected. Swidler’s empirical findings, particularly in her more recent work,
demonstrate a social world woven through with symbol systems and collective representations.
Despite Lizardo’s impressive corralling of representationalism in his
early theoretical work, signification still sneaks in at the margins. Although he failed to
acknowledge it, representationalism contributed mightily to Vaisey’s efforts to introduce
the dual process model of culture in action into sociology. His findings substantiate
representationalism inasmuch as they refute it. Some of the sources upon which
Martin builds his argument against representationalism dispute his conclusions in
explicit terms. In fact, D’Andrade’s argument above, about representational forms and
signification processes contributing to the very production of contemporary humanities’
neurocognitive infrastructure and processing capacities, offers the strongest counterargument.
Representationalism runs through cognition and its sciences.
More than any others, Lizardo and his periodic coauthor Michael Strand have
incorporated neurocognitive research in effort to translate representationalist concepts
like belief into practice-theoretical terms like habit (Strand and Lizardo 2015;
Representationalism and cognitive culturalism: riders on…
see also Lizardo and Strand 2010). Through this impressive body of work, the
authors have nonetheless acknowledged that representationalism remains a significant
dimension of both theory and lived experience. “Our intention in developing
the notion of practical belief is not to completely eliminate the representationalist
understanding” of action, the authors note, but to sketch its limits and identify “substantive
problems that it is not equipped to handle and new areas of research that it
cannot address.” The authors conclude that, in “this respect, the notion of practical
belief supplements representationalism, it doesn’t ‘eliminate’ it” (Strand and Lizardo
2017, p. 187). Consistent with Strand and Lizardo’s conclusion, I am advocating
similarly, even as I am arguing that representationalism should play a more
robust role in general theories of action.
My analysis also suggests the need to reexamine how we conceptualize relations
between automatic and deliberative cognitive processes, and consequently how we
conceptualize action as the product of interactions between these domains. Jeanette
Kennett and Cordelia Fine’s (2009) analysis of the cognitive literature suggests
as much. Kennett and Fine argue that “the empirical literature indicates a complex
interplay between automatic and deliberative mental processes in moral judgment
formation, with the latter [the rider in the “rider on an elephant” metaphor] constraining
the expression and influence of moral intuitions” (p. 77).
Cognitive culturalist interventions are helpful in prompting representationalists
to revisit questions of socialization and internalization. Figures like Lev Vygotsky
and Jacques Lacan are resources. Cognitive culturalism should also spark representationalists
to take steps to better specify culture’s diversity of forms, and to examine
how these forms produce variability in actions within and across segments of a
shared cultural community. Representationalists need not immediately resist neurocognitive
interventions. As presented here, avenues exist for integrating such theories,
concepts, and findings in ways that substantiate the presuppositions that buttress
representationalism and signification.
Finally, it is worth reiterating that cultural sociology’s varieties of representationalism
have distanced themselves from the characteristics and practices cognitive culturalists
tend to attribute to them. The varieties of post-structuralism that grew out
of the linguistic turn do not posit that all persons interpret objects, actions, and ideas
in the “exact” same way, of course. Since at least the reception of Jacques Derrida’s
critique of Levi-Strauss’s structuralism, cultural sociologists have been orienting
toward their research sites by looking for emergent or durable symbolic forms in a
fragmented arena of cultural signs and social conflict. Even Geertz himself abandoned
the web metaphor.8
In a discussion on the “overarching points about cultural
coherence and incoherence, [and] whether everybody has everything,”  Geertz, as
quoted by anthropologist Richard Shweder (2007, p. 199), explains:
As I’ve said before, the elements of a culture are not like a pile of sand and not
like a spider’s web. It’s more like an octopus, a rather badly integrated crea-
8
  See Smith (2005, pp. 44–45), for a discussion of the steps representationalist cultural sociology should
take in order to move beyond Geertz’s “webs of significance” metaphor.
	J. L. Mast
ture—what passes for a brain keeps it together, more or less, in one ungainly
whole. But we must, as anthropologists, search for as much coherency as we
can find, try to find connections, and where we can’t find them simply say that
we can’t find them.
Post-Geertzian cultural sociology does continue to assert that meaning forms
through convention over time, that this sets in motion the perpetuation of a relatively
arbitrary symbol system, and that this system mediates how people interpret
and understand their worlds. Searching for patterns or cultural structures does not
mean that one presumes that the universe of symbolic forms is ‘coherent’ and stable,
however. To the contrary, this research is predicated on investigating a symbolic
arena that is characterized by flux, ambiguity, and amorphousness, inasmuch as it is
defined by structures.
Acknowledgements The reviewers’ comments and those of the editor helped to sharpen this article’s
arguments considerably. The author thanks them for their significant feedback. The author also thanks
the participants of the Yale Center for Cultural Sociology Workshop, Alexis Galán and Agustin Casagrande
at Excellence Cluster Normative Orders, and Erik Ringmar for their insightful comments.
References
Alexander, J.C. 1987. The Centrality of the Classics. In Social Theory Today, ed. A. Giddens and J.
Turner, 11–57. Stanford, CA: Stanford University Press.
Alexander, J.C. (1988 [1987]). Action and Its Environments. In Action and Its Environments: Toward a
New Synthesis, ed. Alexander, J.C., 301–333. New York: Columbia University Press.
Alexander, J.C. 2006. The Civil Sphere. Oxford: Oxford University Press.
Alexander, J.C., and P. Smith. 1993. The Discourse of American Civil Society: A New Proposal for Cultural
Studies. Theory and Society 22 (2): 151–207.
Bellah, R. 1970. Between Religion and Science. In Beyond Belief, ed. R. Bellah, 237–257. New York:
Harper and Row.
Bellah, R., et al. 1985. Habits of the Heart. Berkeley, CA: UC Press.
Berezin, M. 2019. On the Construction Sites of History: Where did Donald Trump Come From? In
Politics of Meaning/Meaning of Politics, ed. J.L. Mast and J.C. Alexander. New York: Palgrave
Macmillan.
Bourdieu, P. 1977. Outline of a Theory of Practice. Cambridge, MA: Cambridge University Press.
Bourdieu, P. 1990a. In Other Words: Essays Towards a Reflexive Sociology. Cambridge, MA: Polity
Press.
Bourdieu, P. 1990b. The Logic of Practice. Cambridge, MA: Polity Press.
Brekhus, W., and G. Ignatow. 2019. The Oxford Handbook of Cognitive Sociology. New York: Oxford
University Press.
Collins, R. 1998. The Sociology of Philosophies: A Global Theory of Intellectual Change. Cambridge,
MA: Harvard University Press.
Cowan, N. 2000. The Magical Number 4 in Short-Term Memory: A Reconsideration of Mental Storage
Capacity. Behavior and Brain Sciences 24: 87–185.
D’Andrade, R. 1995. The Development of Cognitive Anthropology. New York: Cambridge University
Press.
D’Andrade, Roy. 2002. Cultural Darwinism and Language. American Anthropologist 104 (1): 223–232.
Derrida, J. 1967. Of Grammatology. London: Johns Hopkins University Press.
DiMaggio, Paul. 1997. Culture and Cognition. Annual Review of Sociology 23: 263–287.
Gallese, V., and G. Lakoff. 2005. The Brain’s Concepts: The Role of the Sensory-Motor System in Conceptual
Knowledge. Cognitive Neuropsychology 22: 455–479.
Representationalism and cognitive culturalism: riders on…
Gallese, V. 2003. The Manifold Nature of Interpersonal Relations: The Quest for a Common Mechanism.
Philosophical Transactions of the Royal Society of London B 358: 517–528.
Geertz, C. 1966. Religion as a Cultural System. In The Religious Situation, ed. D. Cutler, 639–688. Boston:
Beacon Press.
Geertz, C. 1973a. Ideology as a Cultural System. In The Interpretation of Cultures, ed. C. Geertz, 193–
233. New York: Basic.
Geertz, C. 1973b. Thick Description: Toward an Interpretive Theory of Culture. In The Interpretation of
Cultures, ed. C. Geertz, 3–30. New York: Basic.
Geertz, C. 1973c. Deep Play: Notes on the Balinese Cockfight. In The Interpretation of Cultures, ed. C.
Geertz, 412–453. New York: Basic.
Gorski, P. 2019. Why Evangelicals Voted for Trump: A Critical Cultural Sociology. In Politics of Meaning/Meaning
of Politics, ed. J.L. Mast and J.C. Alexander, 164–184. New York: Palgrave Macmillan.
Gray, K., and J.E. Keeney. 2015a. Impure, or Just Weird? Scenarios Sampling Bias Raises Questions
About the Foundation of Morality. Social Psychology and Personality Science 6: 859–868.
Gray, K., and J. Keeney. 2015b. Disconfirming Moral Foundations Theory on its Own Terms: Reply to
Graham. Social Psychological and Personality Science 6 (8): 874–877.
Haidt, J. 2001. The Emotional Dog and Its Rational Tail: A Social Intuitionist Approach to Moral Judgment.
Psychological Review 108 (4): 814–834.
Haidt, J. 2005. The Happiness Hypothesis. New York: Basic.
Haidt, J. 2012. The Righteous Mind. New York: Penguin.
Hayes, J.R.M. 1952. Memory Span for Several Vocabularies as a Function of Vocabulary Size, in
Quarterly Progress Report. Cambridge, MA: Acoustics Laboratory, Massachusetts Institute of
Technology.
Intraub, H., C. Gottesman, E. Willey, and I. Zuk. 1996. Boundary Extension for Briefly Glimpsed Photographs:
Do common perceptual processes result in unexpected memory distortions? Journal of
Memory and Language 35: 118–134.
Kennett, J., and C. Fine. 2009. Will the Real Moral Judgment Please Stand Up? The Implications of
Social Intuitionist Models of Cognition for Meta-Ethics and Moral Philosophy, Ethical Theory and
Moral Practice 12 (1): 77–96.
Lakoff, G., and M. Johnson. 1999. Philosophy in the Flesh: The Embodied Mind and Its Challenge to
Western Thought. New York: Basic Books.
Lizardo, O. 2004. The Cognitive Origins of Bourdieu’s Habitus. Journal for the Theory of Social Behaviour
34: 375–401.
Lizardo, O. 2007. ‘Mirror Neurons’, Collective Objects and the Problem of Transmission: Reconsidering
Stephen Turner’s Critique of Practice Theory. Journal for the Theory of Social Behaviour 37 (3):
319–350.
Lizardo, O. 2014. Beyond the Comtean Schema: The Sociology of Culture and Cognition versus Cognitive
Social Science. Sociological Forum 24 (4): 983–989.
Lizardo, Omar, and M. Strand. 2010. Skills, Tool Kits, Contexts and institutions: Clarifying the Relationship
Between Different Approaches to Cognition in Cultural Sociology. Poetics 38 (2): 205–228.
Martin, J.L. 2010. Life’s a Beach But You’re an Ant, and Other Unwelcome News for the Sociology of
Culture. Poetics (Special Issue on: Brain, Mind and Cultural Sociology) 38 (2): 229–244.
Mast, J.L. 2019. Introduction: Fragments, Ruptures, and Resurgent Structures: The Civil Sphere and the
Fate of ”Civilship” in the Era of Trumpism. In Politics of Meaning/Meaning of Politics, ed. J.L.
Mast and J.C. Alexander, 1–16. New York: Palgrave Macmillan.
Moore, R. 2017. Fast or Slow: Sociological Implications of Measuring Dual-Process Cognition. Sociological
Science 4: 196–223.
Miller, G.A. (1994 [1955]). The magical number seven, plus or minus two: some limits on our capacity
for processing information. The Psychological Review 101 (2): 343–352.
Patterson, O. 2014. Making Sense of Culture. Annual Review of Sociology 40: 1–30.
Pollack, I. 1952. The Information of Elementary Auditory Displays. Journal of the Acoustic Society of
America 24: 745–749.
Pollack, I., and L. Picks. 1954. Information of elementary multi-dimensional auditory displays. Journal
of the Acoustic Society of America 26: 155–158.
Pugh, A.J. 2013. What Good are Interviews for Thinking About Culture? Demystifying Interpretive Analysis.
American Journal of Cultural Sociology 1 (1): 42–68.
Shepherd, H. 2011. The Cultural Context of Cognition: What the Implicit Association Test Tells US
about How Culture Works. Sociological Forum 26 (1): 121–143.
	J. L. Mast
Shweder, R. 2007. The Resolute Irresolution of Clifford Geertz. Common Knowledge 13 (2–3): 191–205.
Smith, P. 2005. Why War?. Chicago: University of Chicago Press.
Strand, M., and O. Lizardo. 2015. Beyond World Images. Sociological Theory 33 (1): 44–70.
Strand, M., and O. Lizardo. 2017. The Hysteresis Effect: Theorizing Mismatch in Action. Journal for the
Theory of Social Behaviour 47 (2): 164–194.
Swidler, A. 1986. Culture in Action: Symbols and Strategies. American Sociological Review 51 (2):
273–286.
Swidler, A. 2001. Talk of Love: How Culture Matters. Chicago: University of Chicago Press.
Tavory, I., and A. Swidler. 2009. Condom Semiotics: Meaning and Condom Use in Rural Malawi. American
Sociological Review 74 (2): 171–189.
Vaisey, S. 2009. Motivation and Justification: A Dual-Process Model of Culture in Action. American
Journal of Sociology 114 (6): 1675–1715.
Publisher’s Note  Springer Nature remains neutral with regard to jurisdictional claims in published
maps and institutional affiliations.
Jason L. Mast  is a Research Fellow at the Excellence Cluster Normative Orders at Goethe University
Frankfurt, Germany. His recent work includes the volume, Politics of Meaning/Meaning of Politics: Cultural
Sociology of the 2016 U.S. Presidential Election (Palgrave Macmillan, 2019; co-edited with Jeffrey
Alexander).