Language And Social Inferences Research Paper

Thematic features of an event refer to a variety of social inferences that are tacitly indicated to a reader or listener of such a communication. The choice of interpersonal terms (verbs and adjectives) in representing an event has been shown to lead to a variety of inferences about a social event. These are about who the implied causal agent for the event is, how long the event is implied to have lasted, inductive and deductive inferences about the qualities of the persons in the event, inter alia.

Interest in this field emerged in the 1960s and 1970s when social psychologists (e.g., Abelson and Kanouse 1966) and linguists (e.g., Garvey and Caramazza 1974) discovered that interpersonal verbs influence the inferences people make in an orderly fashion. It was in the 1980s that this field was more systematically charted. This started with the discovery that interpersonal verbs referring to visible actions—and consequently classified as verbs of action (e.g., to help, to buy, to cheat)—systematically imply that the subject in a sentence is the agent of the event or the person who caused the event (as in ‘John cheats Edward’). In contrast, verbs of state (e.g., to love, to respect, to abhor)—which refer to unobservable interpersonal states—systematically convey the impression that the causal origin for the event is to be found in the sentence subject (e.g., ‘John respects Edward’). Thus, the verb used in such sentence constructions can systematically direct the thematic focus of an event to either the sentence subject or object as the causal origin of the event (e.g., Brown and Fish 1983). Another illustration of this can be seen in the use of action and state verbs in question formulation. These verb types tacitly issue instructions to a respondent by conveying information about how the thematic agency of an answer should be structured. Thus given the very same event, an interviewer’s choice of ‘Why did John trust Edward?’ (state verb) instead of ‘Why did John confide in Edward?’ (action verb) imposes a specific focus on the theme of the interviewee’s answer. The question requests an answer that is primarily about the qualities that make Edward trustworthy, rather than John’s motives and reasons for confiding in Edward (Semin 2000).

Implied causation and thematic focus are not the only systematic inferences that are mediated by interpersonal language. We communicate about a social event in order to pass on what we saw, heard, or believed. In talking about an event we make implicit choices about how to represent it in our communication. By means of these choices a communicator structures the linguistic representation of an event implicitly, albeit strategically, in order to convey a diversity of inferences to the recipient of the message. Generally, we do not consciously reflect upon our particular choice of verbs and adjectives when we talk about a particular event. Thus, we might refer to the same event by saying ‘John pushed David,’ ‘John challenged David,’ ‘John hates David,’ or ‘John is aggressive,’ to name just a few options. These choices vary from a term that is very concrete, and refers to a directly perceivable and contextually specific act (pushing), to a term that is most abstract, that maintains no references to any specific context, relationship, or situation and expresses a nonvisible quality in the form of an adjective (aggressive). ‘Challenge’ and ‘hate’ occupy intermediate positions. The former verb still refers to a concrete act without preserving the direct reference to the act. The latter refers to an unobservable psychological state that characterizes a specific relationship between two people. The concrete–abstract dimension on which interpersonal language can be systematically classified has been shown to carry a number of tacit, systematic, and general inferences (Semin and Fiedler 2001). It is important to note that while any of the four speech acts about the ‘hostile’ nature of an event is intended to impart information about what has occurred, the particular choices of interpersonal terms in formulating the speech act conveys additional tacit information that is not consciously accessed. This information can be about who caused an event, who was responsible for the event, whether or not the behavior was under the voluntary control of the agent, whether it was determined by the personality make-up of the person, or situationally caused, and so on. Thus, different properties of interpersonal language such as verb type or abstraction–concreteness lead to a variety of tacit social inferences. So far we have discussed the types of social inferences mediated by interpersonal language. Two issues arise. The first is about the context within which language and social inferences are relevant. The second issue concerns the psychological processes that lead to specific patterns of language use in a message and those processes that are induced by the tacit properties of a message.

The context issue becomes a comprehensible one when one inserts ‘language and social inferences’ into a communication framework. The simplest exemplification of a communication framework requires (a) a transmitter who produces a message about an event (e.g., a hostile act), (b) a medium upon which the message is mapped—language, and (c) a recipient. The transmitter of a message has plans about an intended action in the form of cognitions that are mapped upon language. The behavioral execution of cognition as intended action constitutes a speech act. The particular message that is structured by an intended action can thematically focus an event in such a way that A (A hit B—action verb) rather than B (A hates B—state verb) is the causal locus for the event. Such a message is designed to structure the representation of the event such that the recipient of the message makes the appropriate social inferences (e.g., A engaged in behavior under voluntary control, A intended the act, A caused B to be harmed, etc.) Let us consider another example, to illustrate the point in a different context. A police investigator interviews a rape victim who is expected to be trustworthy or untrustworthy. In the former case, the interviewer is likely to pose questions that imply the agency of the perpetrator to the events that led to the incident (e.g., ‘Did he dance with you?—implying ‘Did he initiate the dancing?’). In the latter case, where the investigator expects the victim to be untrustworthy, the question may actually be like ‘Did you dance with him,’ implying that the victim was the initiator for the ensuing event.

Thus, one function of language is to give overt expression to cognitions as intended actions, by providing the transmitter with the means to formulate a verbal act. The second related function is to structure the specific way in which the recipient of the message represents the event. This representation involves also the social inferences that the transmitter intended when structuring the message. Thus, a communication framework helps locate the functions that interpersonal language and associated social inferences have in a transmitter and recipient context.

So far we have covered the following:

(a) the inferential properties of language and

(b) a communication framework that specifies

(i) how a transmitter uses language to give public shape to intended actions, and

(ii) how language in the form of a message shapes the representation (and inferences) of a recipient.

The second issue involves the nature of the psycho- logical processes. This issue has two facets. First, what are the psychological processes by means of which the transmitter produces messages with a specific set of inference inducing patterns rather than others? Second, what types of psychological processes are induced in the recipient by the specific linguistic properties of a message?

We begin by illustrating the types of psychological processes involved in message production and how a closer examination of the linguistic properties of a message facilitate highlighting cognitive and motivational processes underlying message production. One specific example is to be found in the research on the transmission and maintenance of stereotypes. This is best represented by the research on the ‘linguistic intergroup bias’ (Maass 1999). This bias refers to the tendency to describe positive behaviors by members of the ingroup and negative outgroup behaviors by the use of abstract language. In contrast, negative ingroup

behaviors and positive outgroup behaviors are communicated by the use of concrete language. By describing positive ingroup and negative outgroup behaviors in abstract terms (e.g., adjectives / traits such as ‘friendly,’ ‘hostile’), the communicator implies that the behaviors in question stem from enduring qualities that are unlikely to change and are likely to be repeated in the future. In contrast, using concrete language (e.g., verbs of action such as ‘to help,’ ‘to hinder’) suggests situationally specific behaviors that are not likely to recur at a future point in time. The bias has been shown to be driven by motivational and cognitive factors. Thus, the differential abstract versus concrete language use is motivated in order to preserve a positive ingroup image and derogate the outgroup and has been shown to be much stronger when the ingroup identity is threatened. Moreover, it has also been shown that stereotype consistent behaviors are communicated at a more abstract level than stereotype inconsistent behaviors. This expectancy driven bias constitutes the more general case in that it suggests that all other things being equal (i.e., motivational factors) the abstractness–concreteness of a message is driven by expectancies. Basically, the ‘linguistic intergroup bias’ is responsible for the transmission and maintenance of stereotypes and prejudice (Maass 1999). The ‘linguistic intergroup bias’ is one illustration of how transmitter-specific cognitive and motivational processes shape the linguistic properties of a message. Of course, by the reverse logic it is also possible to infer what specific psychological transmitter processes are at hand by analyzing the linguistic properties of a message.

We now turn to recipient-specific processes. Here the focus is on controlling the linguistic properties of a message and examining its psychological implications. First, messages controlling for verb type or abstractness concreteness have systematically been shown to induce the social inferences described earlier. Thus, recipients of messages that arise through the linguistic intergroup bias do interpret the message as intended— confirming that the linguistic properties of the message are responsible for the transmission of stereotypes. Further research suggests that if an event is represented by the use of concrete language (e.g., verbs of action) rather than abstract terms (e.g., verbs of state or adjectives), then it is more vividly represented, processed more systematically, and recalled better and with fewer intrusions. In contrast, representing an event abstractly leads to more heuristic processing, poorer recall, and more intrusions in memory. Moreover, if people are asked to recall an event by using a concrete term (‘When was the last occasion that you helped somebody?’) as a retrieval cue, then they recall events that are significantly more recent than when they are asked to recall an event with an abstract retrieval cue (‘When was the last occasion when you were helpful?’). Thus, properties of interpersonal language also have recipient-specific cognitive processing implications in terms of recall, or representation, aside from the specific social inferences that they mediate.

Bibliography:

1. Abelson R P, Kanouse E E 1966 Subjective acceptance of verbal generalizations. In: Feldman S (ed.) Cognitive Consistency: Motivational Antecedents and Behavioral Consequences. Academic Press, New York, pp. 171–97
2. Brown R, Fish D 1983 The psychological causality implicit in language. Cognition 14: 237–73
3. Garvey C, Caramazza A 1974 Implicit causality in verbs. Linguistic Inquiry 5: 459–64
4. Maass A 1999 Linguistic intergroup bias: Stereotype perpetuation through language. Advances in Experimental Social Psychology 31: 79–121
5. Semin G R 2000 Language as a cognitive and behavioral structuring resource: Question–answer exchanges. In: Stroebe W, Hewstone M (eds.) European Review of Social Psychology. Wiley, Chichester, UK, pp. 75–104
6. Semin G R, Fiedler K 2001 The Linguistic Category Model. Psychology Press, Philadelphia, PA