Semiosis 37

o
FIGURE 4.1 Sign: the circle represents any swatch of something perceptible (e.g., cloud, plume of
smoke, sound of someone speaking).
4
reader, complete it by acting as an interpreter, imposing a process on it by taking it to
Semiosis stand for something. Part of what is missing here is the factor of time. Although the
omission of time from the problem of meaning may appear convenient for research
purposes,4 it results in a view whose flaws can be remedied only by the "addition" of
time back into the equation. s The reality is that nothing in the realm of meaning can be
examined independently of its position in the course of time. Think of what we do when
This chapter outlines a core set of neo-Peircean mechanisms for grounding the rela- we ask what a word means. Without access to time, we can't get a handle on it. If you
tional understanding of meaning that is central in the arguments of this book. l We want to know the meaning of a word, you have to consider the word in light of some-
will give substance to the idea, as Bateson put it, that "information is a difference thing that happens beforehand, simultaneously, or afterward. For example, you might
that makes a difference."2 The wording here is telling, as it refers to two distinct dif- make reference to a preceding sentence, a scene the speaker has just seen, a co-occurring
ferences. The first is a difference that can be perceived in the world-we refer to it gesture or facial expression, a translation into another language, or someone else's
as a sign-and the second is a similarly perceptible difference that results from, or response in a dialogue. The need for a temporal perspective is familiar in those branches
is a coherent reaction to, the first difference-we refer to it as an interpretant (not- of linguistics and related disciplines that recognize the distinct causal-temporal frames
ing that, as we shall see below, an interpret ant is usually also a sign). In the realm shown in Table 3.2 in Chapter 3. Our level of focus here is enchrony.
of meaning, this implies the existence of someone by whom the first difference is As an example of signs in a temporal context, imagine a moment at home. Bill
perceived and by whom the second difference is produced. And further, a meaning- looks out the window and sees dark, heavy clouds. He grabs an umbrella. Jane sees
ful response is not caused by brute force,' but makes sense insofar as it points to Bill grab the umbrella and asks, "Where are you going?"
some mediating object of interpretation. So, let me put it in the terms that will be
explicated in this chapter. Meaning is what we have when a sign gets someone to (8)
produce an interpretant, thus revealing an object of interpretation. The rest of this
I: Bill turns to see dark clouds outside
chapter explains what this means,
2: Bill grabs umbrella
3: Seeing Bill's behaviOl: Jane asks, "Where are you going?"

4.1 Anatomy of the Semiotic Process How do we analyze the relations between these three moves in sequence? If we
were to attribute only a surface interpretation, based on the identifiable swatches of
To understand meaning, we have to understand the semiotic process. Here we con- behavior, we would have simply a sequence of three signs (in the sense of Figure 4.1),
sider the basic anatomy of this process, with a special debt to Kockelman's recent one after the other (Figure 4.2).
work. 3 The semiotic process is fundamental to most of what is discussed in this This behaviorist representation can't be right, or at least it can't be the whole
book. Let us now explicate this complex process by building it up piece by piece. story, because we don't want to say that each move arises from the prior turn by
First, nothing can be said to mean anything unless it can firstly be perceived by
someone who attributes meaning to it. Whether it's a cloud in the sky, a plume of
smoke billowing from a forest grove, or a sentence whispered in your ear, it begins
as a swatch of something perceptible, something at a time and place. This swatch
o 'dark clouds' o A1grabs umbrella o A2 says 'Where are you
going?'

of something perceptible, something noticeable, is what we will refer to as a sign, FIGURE 4.2 "Behaviorist" representation of a sequence of three related observable signs, with two
abstractly diagrammed in Figure 4.1. agents (Al and A2) involved; each sign is an effect/response of the previous, but there are no
The diagram in the figure is inadequate because it does not explicitly include a mediating variables explicit in the diagram.
36 spatial-temporal context, or a perceiver. If the diagram works it is because you, the Note: on mediating variables and the interpretation of others' behavior, see Whiten (1997, 1996).

b
 r
38 Relationship Thinking Semiosis 39

natural causes in the same way that a billiard ball knocks another into motion. Agent
Instead, each of these moves has a meaning, something that is not directly per-
ceived but is signified, and the response to a move makes sense only in terms of
that signified or stood-for meaning. So, suppose we enrich this picture by explicitly
acknowledging that signs are not just perceptible swatches, but have signifieds-i.e.,
these signs stand for things. With a view of meaning by which a signifier maps onto
a signified concept (in Saussurean fashion; never mind that we are not yet talking
about words here), the sequence in (8) would look something like what is seen in Interpretant
Figure 4.3.
The version in the figure still lacks overt reference to time, and does not sup-
ply any explicit characterization of the relations between these signs in a sequence.
On a Saussurean account of meaning as static form-meaning representations, time
would have to be "added in" when signs occur in sequence like this. But in fact time
is already there. It is smuggled into the diagram-in Figure 4.3, by laying the signs
out in order on a left-to-right axis on the page. Are we to introduce these relations
of time and relevance by setting up an entirely distinct set of machinery alongside Object

the basic standing-for relation that underpins signs? Preferably not.

The solution is to begin with the neo-Peircean notion of a dynamic semiotic FIGURE 4.4 The basic semiotic process (by Kockelman 2011,2013, with adjustments): a sign is
process that incorporates four key, interrelated components. The essence of it, sensed by some agent, who takes the sign to stand for an object insofar as the perception of the
defined by Kockelman,6 is shown in Figure 4.4. sign results in the agent instigating an interpretant (some kind of reaction, including a thought or
Embedded within this figure is the Saussurean "standing for" relation: it can feeling), where this interpretant makes sense in terms of it being oriented to that same object. This
be seen in the sign-object relation, where" for example, a sound-image might be process, with its constituent elements and relations, is a primitive of the framework presented in
taken to stand for an idea. But this relation can exist only insofar as several other this book.
relations are involved with it. The present framework makes those relations explicit.
The relations of interest include, crucially, the relations that link the sign and the one hand by a sensing, instigating agent, and on the other hand by a relevant object,
interpretant in various ways. The link from sign to interpretant is mediated on the defined as whatever the sign is being taken by this agent to stand for (only insofar
as that can, and needs to be, ascertained). Think of it in terms of abduction. If you

'dark clouds' Ai grabs umbrella

o A2 says 'Where
are you going?'
want to know what object the sign is being taken to stand for, look for something
such that if the sign stood for this thing, then this interpretant would make the most
sense. A familiar application of this reasoning occurs when you hear unfamiliar
words in action. Suppose you go into a pub in New South Wales for the first time.
As you wait at the bar for your turn to order, you hear one customer ask A middy
of VB, thanks, and then a second customer A schooner of VB, thanks. The first
difference, in Bateson's terms, is the difference in sound between the words middy
[midi] and schooner [sku:nJ]. The resulting difference-i.e., the difference that is
made by the first difference-is in the bartender's distinct interpretants of the two
orders. In both cases, the bartender serves a glass of "Victoria Bitter" lager (hence
"VB"). No difference there. But in the first case, it is a 285 ml glass, in the second,
o = imminent rain 0= Ai is going out
a 375 ml glass. Everything runs off smoothly, and nobody complains, which allows
FIGURE 4.3 "Saussurean" representation of a sequence of related signs; each public, externally you to hypothesize by abduction that the sign [midi] stands for a 285 ml beer glass,
perceptible sign has an off-screen component (i.e., something that the sign stands for but that is and [sku:nJ] stands for a 375 ml glass. Such examples illustrate Zipf's idea that
not publicly or otherwise directly perceivable in the sign), though the status of each next sign as the meaning of a word can be defined as "a kind of response that is invoked by the
an effect of, or response to, the prior is not explicitly represented. word."? In this way, simple Saussurean psychological relations of standing-for are

b
 r
40 Relationship Thinking Semiosis 41

extrapolated from-indeed, created by-more complex social processes of inter- ~--------------------

pretative response.
A = someone
~
,'1>"",/
/
A2 "
\
I \
On this account, the three-sign sequence in (8) is readily analyzed as a chaining I \
of these atomic neo-Peircean processes. / I I
'?O>t". I I
Let's take it one frame at a time. What is labeled as an interpretant in Frame ~t". I I
1 (grabs umbrella) is an interpretant only because it has been framed as such here ~I I
I I
(Figure 4.5). In a subsequent frame (Frame 2, Figure 4.6), the behavior of grabbing I
I = 'Where are I
the umbrella is now framed as a sign, to be sensed by another agent, in turn making you going?'1
a further interpretant relevant. This new interpretant comes in the form of Line 3 I
I
("Where are you going?" spoken by Jane), which makes sense as a response to line I
2 insofar as line 2 has the meaning that it does (i.e., the object that it does, glossed I
I
here as "Bill is going out"). I
I
It is essential to understand what framing means. When we say that an I
agent, by producing an interpretant, is taking a sign to stand for some object, I
0= imminent rain , 0 = Ai is going out /
we are saying it only for a certain frame of current interest. It would be a misin- , -------------------- '
terpretation of this framework to think that signs are associated with necessary
FIGURE 4.6 A second frame of the sequence in (8), Line 2 (grabbing an umbrella) is sensed
~,--------------------------,
by an agent (Jane) and results in Jane instigating a controlled behavior as interpretant (asking
,~~,/ A1 = someone "" "Where are you going?"). This interpretant is both relevant to (or caused by) Line 2 and oriented to
~I \ something that Line 2 stands for (i.e., its object, glossed here as "Bill is going out").
I \
I \
I I
I I
I I and predefined interpretants insofar as those signs are supposed to stand for
.~¢o I
I q unique or correct objects. The whole point of the framework is to capture flex-
I 0<::- I
I Cj I ibility in semiotic processes. In the example given here, the behavior of grab-
I I
I I bing an umbrella happens to be responded to with the addressed utterance,
I ~~~ : "Where are you going?" To make sense of this utterance we can hypothesize
I I = grab umbrella I
I I
that a possible object of the sign "behavior of grabbing an umbrella" would
I I be "This person is going out." It is "the" object here only because it happens
I U't. I
I <»A to have been taken by this agent in this way on this occasion (and, let us sup-
I '/0.' I
~ I pose, it does not elicit surprise or correction from Bill, and so is effectively
I "0,.. I
I I ratified by being "allowed to go through"). Countless other interpretants could
I I
I I have been produced, pointing to countless other objects. This is what underlies
I I Noam Chomsky's critique of B. F. Skinner's notion that linguistic behavior is
I I
\ I
inflexibly caused by stimulus controJ8:
\ I
\. 0 = imminent rain / A typical example of "stimulus control" for Skinner would be the response to
" ,--------------------------- ,/
a piece of music with the utterance Mozart or to a painting with the response
Dutch. These responses are asserted to be "under the control of extremely sub-
FIGURE 4.5 A first frame of the sequence in (8). Line 1 (dark clouds) is sensed by an agent (Bill)
tle properties" of the physical object or event (l08). Suppose instead of saying
and results in Bill instigating a controlled behavior as interpretant (grabs umbrella). This behavior Dutch we had said Clashes with the wallpaper, I thought you liked abstract work,
is both relevant to (or caused by) Line 1 and oriented to something that Line 1 stands for (i.e., its Never saw it before, Tilted, Hanging too low, Beautifitl, Hideous, Remember our
object, glossed here as "imminent rain"; note that here we are not distinguishing the actual rain camping trip last summer?, or whatever else might come into our minds when
on that occasion from Bill's concept of it, though the distinction can be made). looking at a picture. .
42 Relationship Thinking Semiosis 43

A:l'" someone Chomsky's point here is that nothing can predict a "correct" interpretant. This is
why we do not want to say that linguistic responses are governed by natural causes
in the sense of stimulus control, in the same way that a collision causes a change of
direction of motion. But crucially, none of the possibilities that Chomsky gives-
Clashes with the wallpapet; etc.-will be taken as merely "whatever might come
/ into our minds." They are more than this. They will all be taken to be relevant, in
/
/
/
I ;::: pass salt the key sense of relevance shown in Figure 4.4. All of them are interpret ants that
I
\ make sense as responses to a sign (in this case, the visual perception of a painting)
'\
'\ insofar as they are oriented to some object that the painting is being taken to stand
'\
'\ for. More generally, this illustrates the key point that a sign has an object when, and
'\
'\ only insofar as, it is taken by someone to stand for one. This means that any sign
'\
'\
'\
has potentially infinite objects.
'\ It is true that conventional linguistic signs such as words have objects that are
'\
'\ 0 =that white stuff. / /
'\ NaCI / effectively coded or fixed, but this is only the cumulative effect of normative ori-
\ ;<
'\ / entation to words in usage in a community. The best we can say is that words are
'\
......... / conventional signs because they are taken by a lot of people on a lot of occasions to
FIGURE 4.7 Tile English word salt is depicted on the right in Saussurean terms, as a sign stand for effectively the same object.9 This is the essence of Peirce's definition of the
consisting of a pairing of a sound image [saltl/saltj with a concept "that white stuff, NaCI" andl
or some actual white stuff; here we see that this is an idealized abstraction from a Peircean
semiotic process, e.g., as embodied in a sequence like 1. A says Pass the salt, 2. B passes that AGENT
white stuff, NaC!.

Agent

Interpretant

Interpretant

Object
OBJECT

FIGURE 4.9 What psychologists and phoneticians (among others who tend to adopt a
FIGURE 4.8 What semanticists and other students of structures of linguistic meaning who adopt a microgenetic frame) often focus on: agents as perceivers of inputs and as instigators or producers
synchronic frame tend to focus on: sign-object relations, the forms of signs and the concepts and of outputs. A key research goal from these perspectives is to find correlations between these
tl1ings that signs stand for. (Note that by "object" here we do not commit to whether it is a concept inputs and outputs and, in the case of psychology, to attribute mediating states to the agent that
in the mind or a thing in the world; both are possible.) best account for the data.

b
44 Relationship Thinking Semiosis 45

symbolic relation of sign to object: this sign stands for this object because of a com- A = someone
munity convention by which people produce interpretants that happen to orient to
just this object. So, we can now show more clearly how the classic Saussurean sign,
as pairing of signifier-signified, is actually an idealized substructure of the larger
semiotic process. See Figure 4.7.
The natural embedding of the Saussurean sign within a Peircean semiotic pro-
cess is illustrated in Figure 4.8.
Compare now research disciplines that focus on the psychological agent rather
than on a system of signs. The underlying semiotic process is the same, and the dis- Ib = bring in washing
ciplinary differences in interest and emphasis are captured by highlighting different
elements of the diagram, as shown in Figure 4.9. 10
And finally, we can capture the concerns of those interested primarily in
sequences of interaction in an enchronic frame, again by highlighting sub elements
. of the common, underlying semiotic process. This is shown in Figure 4.10.
A semiotic process framework allows us to show that seemingly disparate
approaches to meaning-structuralism, cognitivism, interactionalism-are focused
on differing but complementary parts of a single, basic type of process. None of the
0= imminent rain
elements are dispensable, though each of the various disciplines tends to proceed as
if they were, by rendering some elements invisible.
FIGURE 4.11 Different interpretants can point to the same objects; for example, as responses to
seeing dark clouds, the behaviors of grabbing an umbrella (Ia) and bringing in the washing (Ib)
Agent both orient to the object "imminent rain;' in that they both make sense as relevant to dark clouds
insofar as dark clouds may be taken to stand for imminent rain.

4.2 Flexibility in Semiotic Processes

Different interpretants can point to the same object within the frame of a single
sign event. Suppose the same dark clouds cause one person to grab an umbrella and
another to bring in the washing. See Figure 4.11.
INTERPRETANT And interpretants can point to different objects of the same sign within the
frame of a single sign event. Recall the example of the betel nut and basket, from
Chapter 2 (Figure 4.12).
Objects are not given by signs. They are oriented to, and thus suggested by,
interpretants.

4.3 Inference as a Semiotic Process

Object
We can use this semiotic process framework to depict chains of inference that are
not all directly observable but that, we have good reason to suspect, are involved in
FIGURE 4.10 What interaction researchers who adopt an enchronic frame tend to focus psychological processes of understanding. One of these is the process of implica-
on: sign-interpretant relations (though actually with rich reference to objects, phrased in terms of ture, described by the philosopher Paul Grice. 11 Suppose I ask How was the movie?
"orientation"). and you reply Good. Good entails "good" because that is the meaning the word
 r
f

46 Relationship Thinking Semiosis 47

A= someone
~0~ ~ -
x,'IfI
- - - - - -
A = someone
- - - - - - -
/'
--= :: ::::'"\_____________ _
;:'l- A ..... '\
I 1<,,(0' \ \
I I" I ,
I I I
I I I
: ~I,\-:>: I
""'I I
I I
I I
I relevance relevance I:::: 'not I
I excelient' I
Ib =say 'You'll chew?' I I I
I I I
I I I
I I I
I I I
I I I
\ \ I
\ \ /
'\ ..... _ _ _ _ _ ..:: .:..·~o~ _ _ _ _ _" ' - _ _ _ ~ /' 0 = 'not excelient' / /

------------------~

FIGURE 4.13 Pragmatic inference depicted as a two-frame chain of semiotic processes (cf.
Ob = wants to chew betel Kockelman 2011). The interpretant in Frame 1 happens to be the same as the object that it is
oriented to. Another way of phrasing what is going on here is that the agent in Frame 1 takes the
object of Frame 1 as a sign (in Frame 2). More generally, the interpretant of one sign (in particular,
FIGURE 4.12 Different interpretants can point to distinct objects of a single sign; as a response
the idea of its object) can bring to attention another sign to interpret.
to seeing the woman reach for the basket, the behavior of passing the basket (Ia) orients to the
object "she wants the basket" in that it makes sense as relevant to the woman's reaching insofar
as her reaching stands for her desire to get hold of the basket; at the same time, as a response
the analysis of Gricean pragmatics. Qualitative differences between the two steps
to seeing the woman reach for the basket, the behavior of saying the words "You'll chew?" (Ib)
are likely to be attributable partly to differences in ground, that is, in the basis for
orients to the object "she wants to chew betel" (a reason for her wanting the basket), in that it
makes sense as relevant to the woman's reaching insofar as her reaching stands for her desire to
the relation of standing-for that holds between sign and object at each step (see
ch ew betel nut. below). Note also that this analysis is applicable not only to conventionalized impli-
cature of the kind illustrated in Figure 4.13 but also to particularized implicatures. 15
So, suppose I ask whether you could come to dinner on Tuesday, and you say [
effectively encodes, but since there are many stronger ways to appraise something have family visiting this week. If I understand your response to convey a rejection
positively, to say good implies that it was not more than good; it was not fantastic, of my invitation, I will have taken your first interpretant (a statement meaning "I
not excellent, and not brilliant. On this account, as Kockelman points out,12 this
A A A A
basic two-step process of interpretation (first we attribute a type-level meaning to
the word; second we calculate from context any relevant implicature) is a chain
of sign-interpretant relations. Beginning with the sign Good (i.e., a phonetic string
[gud)), the interpreter takes it as a sign that stands for its type-level, coded lexical I/S_...:.::.:.:==-.... 1/5
meaning "good." Let the interpretant in this first frame be the attribution of this
meaning, which in this case has the same content as the sign's object. In a subse-
quent frame, this type-level object of the lexical item good then becomes a sign that
is now itself to be interpreted, in the context of a relevant semiotic system, namely o o o o
the language with its full set of expressive options. 13 In light of this context, a likely
interpretation of good will be "not excellent."14 FIGURE 4.14 Unbounded chain of semiotic process as basis for cultural and cognitive
This common type of interpretation is captured here as a two-step process, but epidemiology.
not one that requires two completely different kinds of step, as is often implied in Source: Sperber (1985, 2006); Enfield (in press c).

b
 r
r
Semiosis 49
48 Relationship Thinking

Possible signs. What is a possible sign? Peirce's technical term qualisign is meant
to answer this question. To know which things could possibly be signs, one needs
to know the nature of the perceiving entity, and so to ask about possible signs is
really to ask about possible relations between signs and perceiving agents. It is a
1/5
matter of affordances, to use psychologist 1. 1. Gibson's term.!9 The question about
possible signs is really a question about ways of sensing, given a certain kind of
sensing agent. Part of the answer to the question of possible signs, as relevant to
this book, is delivered by what we know about human perception, by the various
o o o
sensory modalities.
The signifying resources that we use for constructing moves in social interac-
FIGURE 4.15 Unbounded chain of semiotic process as basis for cultural and cognitive tion may be analyzed into distinct semiotic dimensions (see Chapter 6).20 In the
epidemiology; this version of the diagram highlights those parts of the chain that are made explicit aural modality, we can independently modulate meaning by varying the pitch,
in anthropologist Dan Sperber's "cognitive causal chain" depiction. loudness, rhythm, and timing of our utterances. In the domain of human vocal sig-
nals, we can vary phonation (voiced, nonvoiced, breathy), vowel quality, segmental
have family visiting this week") to be a sign that stands in turn for something more articulatory distinctions (place and manner of articulation), identity qualities (who
directly relevant (i.e., so I can't take up your invitation). The process is the same as the speaker is, whether it is a man, woman, child), transient state qualities (whether
illustrated in Figure 4.13 for conventionalized implicature. the speaker is drunk, tired, excited), and more. 2! In the visual modality, for example
with hand gestures, we can independently modulate meaning through values such
as moving versus stationary, punctuated versus persistent, acceleration in motion,
4.4 Cultural Epidemiology as a Semiotic Process direction of motion, path of motion, manner of motion, shape, symmetry, posi-
tion along spatial dimensions, pitch, roll, yaw, and more. 22 For each of these many
The basic machinery of the semiotic process is the exact same one needed for distinct dimensions, we may assess its relative status on a number of perceptible
modeling cultural epidemiology, the causal model of cultural transmission long properties:
advocated by cognitive anthropologist Dan Sperber.!6 Cultural epidemiology is the
geographical and historical distribution of culture and cognition in human popula- Persistence (imagine a cline from less persistent to more persistent, running
tions. The present model of the semiotic process captures this by showing how the from stop consonants to vowels to hand gestures to sand drawings to print
basic unit of semiosis can generate unending chains!? of meaningful interaction on paper to stone inscriptions)
involving arbitrary numbers of agents, as both producers and interpreters of signs, Controllability (e.g., linguistic phonetic structures more controlled, facial
and allowing that chains may be entirely private or may also involve public signs. expression less controlled)
Affordancesfor different types of ground (visual media are more readily
This is illustrated in Figures 4.l4 and 4.15.
Without the mediating structure here termed the object, Sperber's model is at taken to be iconic by a vision-foregrounding species such as humans)
risk of being wrongly taken for a behaviorist model in which the causal links are These are the kinds of dimensions that could yield a fine-grained anatomy of the
not mediated by meaning. logical possibilities for the formulation of composite utterances, together with com-
binatoric principles by which multiple signs in combination can come to have richer
meanings, whether by simple association or by complex syntax.
4.5 Elements of the Semiotic Process and Their Possibilities Possible objects. The object of a sign-what we take a sign to stand for-can be
many things, and indeed many quite distinct types of things. As we have long known
To describe a full anatomy of the semiotic process would entail explicating a from classic work in semantics from Odgen and Richards to Carnap to Lyons,23 a
typology, or at least a taxonomy, for each individual element of the process. What semiotic object may have intension, that is, it may consist of some set of defining
are the possible types of sign, agent, object, and interpretant? What are the pos- properties; it may correspond to concepts in the mind. And an object may have
sible types of sensing, instigating, standing-for, and orienting-to? What kinds of extension, that is, correspondence with things in the world. Like other elements of
relevance are there? Some of these questions have been raised, and to an extent the semiotic process, an object can be a token (actual instance) or a type (abstract
answered, by Kockelman among others,!8 and I will say only a few words about class).24 When I say She's a Manx, pointing to an individual cat, my composite
them here.

b
50 Relationship Thinking Semiosis 51

action incorporates a word Manx and an entity, the cat. The use of the word Manx Note that to really speak of types of interpretants, it is hard not to also make refer-
is itself also complex in structure. It incorporates both a signifier (token [m<elJkhs] ence to the relation of relevance, as well as the objects that are crucially implied in
of type Im<eIJksl) and a signified (the actual cat as token of the type "Manx cat"). that relation, since not all instigatedlcontrolled behaviors are necessarily interpre-
Turning from simple referential expressions (word mapping onto thing), we tants. To be an interpretant, a controlled behavior must be oriented to some object
find further complexities in the kinds of object that a full proposition may have. For of a sign.
one, a proposition, as possible object of an utterance, may be internally complex in Possible ways of standing-fm:' types of ground. Peirce famously specified three
terms of topic-comment structure. A proposition will topicalize or thematize some- kinds of reason that a sign can be taken to stand for an object. First, the relation
thing, and at the same time say something about it. 25 In She's a Manx, the cat is is iconic to the extent that a sign is taken to stand for an object because it has
thematized-"she," pointing at the cat-and then the cat is characterized by being perceptible qualities in common with the object. Second, the relation is indexical
described as a particular variety of domestic cat. to the extent that a sign is taken to stand for an object because it has a relation of
A second type of internal complexity that the object of an utterance may have contiguity (spatial, temporal, or causal) with the object. Third, the relation is sym-
arises from the feature of "displacement" inherent to language, namely the articula- bolic to the extent that a sign is taken to stand for an object because of a norm by
tion of relations between speech event and narrated event, giving rise to a complex which this sign shall be taken to stand for this object by people in this group. We
set of deictic specifications such as tense, spatial reference, and social indexicality, commonly hear of icons, indexes, and symbols, as if these relations are exclusive
and subsequent possibilities for transposition of perspectives. 26 Fleshing out the from each other. But they are not. These three types of ground co-occur all the time.
possibilities of types of object at the level of utterances is the subject matter of Take the example of a telltale shoe print at a crime scene. The print is both iconic
propositional semantics. and indexical. It is iconic in that it resembles the shape of a shoe sole, and because
Possible interpretants. What are the ways we react to a sign, and in so doing of this resemblance it stands for a shoe sole. It is indexical in that (a) it was directly
give it meaning? We can view the answer to this question as a kind of mirror of caused by a shoe sole at that place (thus standing for an event of stepping andlor
the case of signs, discussed above, in that when one talks about interpretants, one standing), and that (b) the sole is in contiguity with the rest of the shoe, which is in
is really talking about the relation between interpretants and the instigating agents contiguity with someone's foot, and thus with the whole person, and thus ultimately
that produce them. Just as the question of possible signs was a question of the standing for the person.
sign-agent relation of sensing, so the question of possible interpretants is-at least
in part-a question of the agent-sign relation of instigating. A great amount of the
answer, then, is provided by what we know about human action involving the body 4.6 Payoffs of This Framework
(motor abilities, etc.), human emotions, and productivity in language. Kockelman
describes four types of interpretant, elaborating on Peirce: 27 The representation in Figure 4.6 of a simple interactional sequence, shown above
in (8), is grounded in the basic four-node semiotic process shown in Figure 4.4.
An ultimate interpretant is an entirely private cognitive response that does
This process is nothing more than is already needed for describing meaning in the
not necessarily beget a further interpretant (hence "ultimate").
most basic sense. It captures the general capacity of people to attribute meaning to
An affective interpretant is in the form of some uncontrolled feeling or
things that they perceive, such as heavy rain clouds, ink marks, or tracks in snow.
sensation (e.g., I blush in response to your flattering words).
Using this primitive semiotic process as a framework for linguistic meaning and
An energetic interpretant is in the form of a controlled behavior (e.g., you
related aspects of human communication has three significant payoffs: generality,
hand me the salt in response to my request).
inclusiveness, and learnability.
A representational interpretant is in the form of some symbolic response
(e.g., I say He's a Manx in response to your question What sort of cat is
that?). 4.6.1 GENERALITY

In other words, we can create meaningful responses through anything we think, The semiotic process, shown in Figure 4.4, gives us the general underlying
feel, do, or say. The former two are relatively inaccessible. Fleshing out the pos- structure of any instance of "someoneAGENT takingINTERPRETANT somethingsIGN to
sibilities for types of interpretant at the level of moves is the business of research stand for somethingoBJEcr" It is analytically parsimonious to use the schema to
on talk in interaction,28 being primarily concerned with ways in which moves form cover what might otherwise be thought of as disparate phenomena. It allows
links in enchronic interactional sequences. From this view, a move has a double us to capture semiotic processes as responses to stimuli, where these responses
identity. It is both a response to a prior move and a prior to a responsive move. make sense not in terms of brute natural causes but in terms of going beyond

b
52 Relationship Thinking Semiosis 53

the data to something that is inferred from these stimuli. This covers every- built up separately and added in; rather, the Saussurean sign is abstracted from
thing from taking dark clouds to stand for rain, to taking a noise like [salt] to such sequences that are already inherent to semiosis in its most basic form. Recall
stand for that white stuff people add to food to make it taste better. By "taking that Peirce defined a symbolic relation as a case in which a sign stands for its object
something to stand for something" I mean displaying an interpretation through because people in a community conventionally take the sign to stand for the object,
some response, which in the case of the dark clouds as a sign might range from and not because of any qualities in common with, or connection to, the object
remarking It's going to rain to merely thinking it to grabbing one's umbrella on (though there may well be some). Or, to be precise, in a symbolic relation a sign
the way out the door. stands for its object because the range of normative interpretants people produce
A possible complaint about this semiotic-process framework is that it is too that are relevant to the sign all point to the region of the object. In other words, the
broad, too all-encompassing. If it covers everything from thinking something to attested range of interpretants would be natural if the object were in this region.
saying something to carrying out some controlled behavior, how are we to distin- Doing lexical semantics, for example, is abducting an object on the basis of data
guish between these clearly different kinds of response? Well, for one thing, gener- from an appropriately broad sample of interpretanti.
ality is a virtue. A framework or theory is better if it is able to explain a broader How does one learn such sign-interpretant mappings? When we say people in
range of phenomena. Second, this framework does not claim that the ways in which a community "take that sign to stand for that object," we are going by the only evi-
we react to clouds and to spoken utterances are exactly the same. What it claims is dence we have, namely the interpretants they produce. The expression "A takes X to
that these interpretive reactions have at their core the same fundamental organizing stand for Y" is a way of saying "A produces an interpretant of X that makes sense
logic. 29 insofar as it orients to y''' So, for me to pass the salt when you make a noise like
[pas 0<) suIt] is for me to treat "salt" as standing for that white powder. I produce an
interpretant of [suit] that makes sense insofar as it orients to the idea of that white
4.6.2 INCLUSIVENESS
powder, and a real-world instance of it. Getting a theory of learning for free is a
We have argued that a sequence of communicative moves is a chaining of semiotic significant reason to favor this approach.
processes in which a sign gives rise to an agent's interpretant (which is oriented
to an object of the sign) and the interpretant then becomes a sign to be further
interpreted in a new frame. If this is true, then we can capture the insights of mul- 4.7 The Saussurean Sign: A Convenient Untruth
tiple approaches that are typically held to be quite disparate, if not incompatible.
The framework captures the insights of the Saussurean notion of a static, repre- The idea that a Peircean notion of meaning as a process could be at the center of
sentational view of meaning as a form-concept mapping, an insight behind use- an entire system of communication is sometimes said to be incompatible with a
ful activities such as writing dictionaries and grammars. It captures the insights Saussurean sense of language as a system of static or synchronic representations.
of historically grounded approaches to language and culture that generalize at the But it isn't. You don't have to choose between Saussure and Peirce; nor do you have
population levepo And it captures the insights of behaviorally, sequentially ori- to posit two distinct machineries. I have explained how symbols are extrapolated
ented approaches such as conversation analysis. Some approaches often don't seek from usage data that we all accumulate. We observe a large range of interpretants
to explicitly incorporate the insights of others, assuming that the others can be of these stable sound-image signs, and we form an abstract object or word defini-
presupposed in some way, or dealt with by means of some separate analytic or tion that has to be consistent with the full range of interpretants. This is what good
descriptive framework. With the present semiotic-process account, however, they lexical semanticists do. 32 It is meaning as hypothesis building, compatible with the
are inherently connected. And note that the approach advocated here does not take model of infant and child word learning outlined by the pioneering psychologist
those other approaches and "add them together"; it shows instead that they are of language Roger Brown. 33 A Saussurean language system is a kind of description
already interconnected aspects of a single phenomenon. Any claim that they are that we could imagine actually corresponds to the kind of mental entity we call an
not would be doing subtraction. idiolect. But we do not directly observe such a system, in an abstracted, detempo-
ralized form. The best we can do is infer its existence. 34 Think of the idealized syn-
chronic language system as a convenient untruth. All we really have are contrived
4.6.3 LEARNABILITY
fragments of it, where the primary unit, by the way, is not the word ("tree," etc.) but
Lastly, this account supplies within its very structure an account of how con- the utterance. The utterance gives us the frame for the words, from which we get
ventional signs are learnedY It allows us to see how the Saussurean sign is not data on their distribution. It is now known from stochastic models of grammar that
"dropped into" sequences of interaction, where a theory of sequence has to be this is all you need to derive a system. 35
54 Relationship Thinking
r Semiosis 55

4.8 A Frame·Content Dynamic Frames in

sequence

If the utterance-and not the word-is the primary unit in the causal ontology of
language, this suggests a frame-content dynamic. 36 I adopt this terminology from
Frame
linguist Peter MacNeilage's work on syllables in spoken language phonology.
The idea is as follows. The basic-level structural unit is called a frame, and this
unit has internal organization, called the content. This is a statement not merely \!I
[Jf~~lC~JCJJ
about the possibly hierarchical or nested nature of behavioral units, but rather Frame with
content
about how the human conception of semiotic system structure develops (see also
Chapters 11 and 12). Think about how we learn the syllables of a language. We FIGURE 4.16 Frames as a privileged unit in learning; they can be fleshed out with internal content
might first learn them as whole units. Once we have those units in place, they may (pointing down and in), and they can be concatenated into larger structures (pointing up and out).
then serve as a frame whose internal content can be analyzed into constituent
elements with ordered interrelations. From these ordered interrelations we can
derive combinatoric principles or rules for combining the elements that flesh out interpreters, derive understandings of our rich, multimodal surroundings.
the content of frames. Once in place, these combinatoric principles can be used Through meaning-imbued residence in the world, we read our environment, as
generatively in deriving or recognizing new arrangements of content in frames. In revealed in how we respond to it. 39 When we see people heed an affordance, such
this way, the frame is a privileged beginning point. It functions as an exoskeleton as when they stride on flat ground, we see that they have read their environment
for the content that fleshes it out at a lower level, building down and in. And it in a certain way. When we see people use technology, for example hammering
functions as a building block for higher-order structures that are built from mul- nails or going up flights of stairs, we conclude they have read their environ-
tiple frames, building up and out. For syllables, this level of frame concatenation ment in a certain way. When we see people respond to spoken utterances, say,
would yield polysyllabic words. by answering questions, again we see how they have read their environment. The
Exactly the same idea can be applied to the basic unit of linguistic utterances sense of the term environment in these three examples moves from natural to
(also known as the clause, the turn-constructional unit, the intonation unit, the technological to social. Although a natural affordance (e.g., of the terrain) may
move). An important part of the process of first-language acquisition in infants determine what is possible, a function (e.g., of a tool) is a product of a designer's
is the learning of ostensibly complex utterance units as chunks, which are only intention, not just what can be done with it but what is meant to be done with
later analyzed into their internal components, isolating constituent morphemes it. So, I subprehend that you will not be surprised when I use a hammer to bang
and yielding morphosyntactic rules that then become productive (see Chapter 6 in a nail. On the other hand, I can anticipate that you will be surprised if I use
for further discussion and elaboration).37 Figure 4.16 illustrates the general the hammer to stir a casserole, regardless of its effectiveness for the purpose. In
frame-content idea. this sense, spoken utterances have the same norm-regulated functionality that
What ultimately follows from the frame-content dynamic is a structured notion everyday artifacts do. 40
of grammar in the fullest sense: a phonological system, a lexicon, sets of mor- The public nature of meaning in the sense just described enables us to sub-
phosyntactic rules and paradigms, grammatical relations, animacy hierarchies, prehend others' interpretants any time we are in the presence of other people. 41
information structure, morphological dependencies, routines of usage, everything We have a tacit sense of how our behavior will be taken by others, and this
(see also Chapter 12).38 From an items-in-utterances model, where utterances cor- subprehension is what guides us in shaping our communicative actions. The
respond to frames as illustrated in Figure 4.16, language systems can be inferred point is crucial for understanding multimodality in the sense that includes phys-
and constructed in their full richness. ical, spatial surroundings as potentially meaningful. 42 Mutually aware of a rich
environment with affordances and functions, people anticipate or subprehend
one another's interpretations of that environment, thereby maintaining rela-
4.9 Meaning as a Public Process tively symmetric attitudes, or degrees of common ground. 43 With this common
ground in view, our partners in interaction draw our attention to momentarily
To conclude this survey of semiosis, consider what it is that a sign producer relevant features of context, and then exploit our likely habits of assessment
presupposes of an interpreter. Consider the semiotic means by which we, as and interpretation in order to bring about the social outcomes they are aiming
 r
r

56 Relationship Thinking

for, whether it be getting someone to pass the salt or defending the accused in
a court of law.

* * *
The fundamental mechanisms of semiosis drive people to see meaning in just about 5
anything. So when we find ourselves in a social context, our behavior is unavoidably
subject to the interpretations of others, as sociologists from Goffman to Garfinkel
to Heritage have insisted. Who we are is determined not simply by what we say and Status
do but also by how others orient to what we say and do. And this determines our
status, the topic of the next chapter.

The word status is used here in a technical sense, derived from the early 20th century
anthropologist Ralph Linton, and fleshed out in Paul Kockelman's recent work. I
Note that this notion of status does not imply that a social category is immutable
or in some other sense not contestable or negotiable. The term status will never be
used in this book to mean prestige or high status. Status here is akin to the notion
of membership category used in ethnomethodology and conversation analysis, or
to the notion of identity as the word is sometimes used. 2 A person's status is defined
here as a collection of her rights, duties, and dispositions. at a given moment, rela-
tive to other members of the social group. A person has innumerable statuses at any
one time, and any of them may be invoked for one or another reason. We can char-
acterize the relevant entitlements and responsibilities directly in terms of enchrony
and its semiotic elements, as introduced in the preceding chapters, without need-
ing to introduce new analytical machinery. The concepts of right/entitlement and
duty/responsibility are defined straightforwardly through the relations of appro-
priateness, effectiveness, and social accountability that hold for any communicative
actions in enchronic sequence (see Chapters 3 and 4).

5.1 Status Predicts and Explains Behavior

A person's status should fit with the person's behavior. Status both predicts and
explains behavior, just as behavior both predicts and explains status. In Kockelman's
terms, if behavior is a sign, status is its object. The behavior will have a degree of
appropriateness or effectiveness, as measured for instance by the degree of norma-
tively justified surprise or sanction in response to certain behavior, given a certain
status. Let us consider an example.
If Jane is a university professor, we can say that this is a status because it may
be characterized as a set of entitlements and responsibilities, regimented both nor-
matively and legally, regarding her social relations with certain other people in cer-
tain communicative contexts. While inhabiting or enacting this status, she will have
rights and duties in relation to her behavior toward her students. Were she to invite 57