REFLEXIVITY, ROLE CONFLICTS, AND THE MEANING OF ENGLISH SELF PRONOUNS^*

Abstract

This study offers an innovative, sign-based analysis of English self pronouns (myself, yourself, herself, etc.). While rejecting the traditional characterization of these forms as reflexive pronouns, the study borrows from the tradition by analyzing these forms as a kind of emphatic pronoun. The forms’ distribution can be explained by positing that they are semantic signals deployed by speakers to meet communicative goals. Speakers choose between self and simple pronouns when the additional meaning of self forms, INSISTENCE ON AN ENTITY(S), will steer hearers in particular interpretive directions. This approach has led to the discovery that reflexive uses of self pronouns are an instantiation of the general tendency to use these forms for unexpected messages, including those in which a single referent is playing more than one role at one time. The presence of such a role conflict accounts not only for reflexive uses, but also for the appearance of self pronouns in picture noun phrases, logophoric contexts, and other previously unexplained exceptions to the structural reflexivity account.

Keywords:
Reflexive Pronoun; English Grammar; Columbia School; Sign-Based Linguistics; Emphatic Pronoun

Duffley (2020) cogently illustrates the advantages of a sign-based approach to linguistic analysis that views linguistic forms as signals of invariant meanings that are deployed by speakers to meet their communicative goals. Duffley further argues that some facts that have been regarded as syntactic phenomena can better be accounted for in terms of meaning. One of the most striking examples of the strength of this meaning-based approach is its application to so-called reflexive pronouns in English (myself, yourself, herself, etc.). This paper aims to show that the distribution of these forms, which has generally been understood as a quintessential syntactic phenomenon, can be explained semantically. The analysis presented here was carried out in the Columbia School framework, a sign-based theory established by William Diver, Erica García and their students at Columbia University in the 1960s and actively pursued since.¹ After a brief critique of traditional syntactic approaches to self pronouns and an overview of an alternative meaning-based analysis, we turn specifically to reflexive uses of these forms.² The analysis will show that speakers choose between self and simple pronouns (me, you, etc.) depending on their communicative intent, and that the same explanatory construct - signals and their meanings, deployed for communicative purposes - can account for the distribution of self pronouns in all environments, including reflexives, in a principled and consistent way.

1. Traditional accounts

English self pronouns are traditionally divided into two groups: reflexives and emphatics. These groups differ in structural terms: reflexives, as in (1), are pronouns in argument position that are coreferent with another noun phrase in the same clause, while emphatics, or intensifiers, as in (2), are appositives (e.g., Quirk et al. 1985).³

• (1) a. Alyssa surprised herself

• b. Jessica believed in herself

• (2) a. Nicole herself was on the radio

• b. Kim couldn’t believe it herself

This classification is widely adopted by contemporary linguists, including syntacticians who identify self pronouns as anaphors and research the structural conditions for coreference between the forms and their antecedents (e.g., Chomsky 1981, 1982, 1995; Reinhart & Reuland 1993). However, scholars have long been aware of data that challenge this account, as in (3) in which the self form’s antecedent is not found within its clause; that is, the self form is neither reflexive nor emphatic.

• (3) a. The ‘Enigma Variations’ were first conceived humorously as a series of musical pictures of himself. (Cavendish, History Today 1999)

• b. It seemed that the case had happened to someone other than himself (McCann, Apeirogon 2020)

These counterexamples have been assigned labels: (3a) is a picture noun phrase, and (3b) is a logophoric use, in which the referent is the person whose “speech, thoughts, or feelings are reported or reflected” (Clements 1975, p. 141).

Another problem for the structural account is that simple pronouns also occur in syntactically reflexive environments, to refer to same-clause antecedents.

• (4) a. She takes the umbrella with her (COCA 1996)⁴

• b. He slowly looks around him (COCA 2004)

However, neither naming these uses, nor excluding them from the syntactic account (as suggested by Zribi-Hertz 1989, Reinhart & Reuland 1993, and others), alters the fact that they are counterevidence to that analysis. Nevertheless, researchers continue to base their investigations on this familiar approach (see Sperlich 2020 for an overview).

By contrast, the sign-based approach advocated by Duffley and adopted here treats the structural analysis as a testable empirical hypothesis that is open to falsification. We find that self pronouns occur not only in syntactically reflexive environments (i.e., with same-clause antecedents), but also in non-reflexive ones. Simple pronouns also appear in both contexts. To put it bluntly, the syntax-based reflexive pronoun hypothesis fails when confronted by the facts of English usage. A different approach is warranted, one in which the distribution of forms is viewed not through the lens of the a priori categories of emphasis and reflexivity, but rather, one in which the analyst asks: what motivates a speaker to choose a self form?

2. Overview of a meaning-based account of self pronouns

This sign-based analysis begins with the forms themselves and asks, what is the meaning of self pronouns? That is, what is the stable, linguistically encoded semantic contribution of these forms? As language users and analysts, we cannot observe the meaning of any form, but only its “capacity … to evoke certain messages in its uses in various contexts” (Duffley 2020, p. 87).

A crucial element of this analysis is one that Duffley examines at length, namely the distinction between linguistically encoded meaning on the one hand, and communicated messages on the other. Duffley (2020, p. 33) describes a “basic dividing line” “between semiologically-signified notional content (i.e. what is linguistically encoded) and non-semiologically-signified notional content (i.e. what is not encoded but still communicated).” In Columbia School terms, this is the distinction between meaning and message, where messages are the infinitely varied ongoing interpretive results of communication; the word ‘meaning’ is a technical term reserved exclusively for that which is semantically encoded by lexical items, grammatical formatives, and certain facts of word order.

This analysis rejects the traditional account of self forms as reflexive pronouns, or in modern parlance, as syntactic anaphors. But it does borrow from the tradition the notion that self forms are emphatic. Like simple pronouns, self forms signal meanings of Person, Number and, in some instances, Gender. Self forms also signal an additional semantic substance, INSISTENCE ON AN ENTITY(S), which can be understood as a forceful pointing, an energetic reference that draws additional attention to an entity or entities, as a way of saying to the hearer, “Yes, this entity. I do mean this one(s)!” (Stern 2006).⁵

There are many reasons a speaker might choose to INSIST on a referent. Self pronouns, as appositives and arguments, all contribute to the same range of messages, including contrast/comparison, exclusion of others, and prominence of referents (Stern 2004). These are not units of analysis, but are merely groups into which varying messages may be sorted. There is a great deal of overlap between these message categories, and the appearance of a form in a particular context is not necessarily motivated by, nor an instantiation of, a single one. Hearers must always infer speakers’ communicative intents, and their motivations for choosing to INSIST. Space limitations preclude a full discussion of these message effects, but some representative examples are shown in Figure 1, which reveals the semantic unity of self forms across all their uses.

Figure 1
Summary of message effects in appositive and argument uses (Stern 2004)

While Figure 1 shows a variety of message effects associated with self pronouns, in this paper I will focus on the use of the meaning INSISTENCE for another type, unexpected messages.

3. Unexpected messages

In addition to the message types shown in Figure 1, self forms also appear as appositives when there is something unexpected (Faltz 1985, 1995; Kemmer 1995; Kemmer and Barlow 1996; Stern 2004), as shown in (5).

• (5) a. It's been 14 years since The Holiday came out and one of the film's child stars is all grown up, completely unrecognizable, and a mom herself! (McRady 2020)

• b. The capitalist myth merges with Christmas as secular redemption allows Tiny Tim and his family to be saved from poverty (not from sin) and celebrate (a feast, not a sacrament) with Christmas plenty donated by none other than Scrooge himself. (COCA 1992)

In these examples, the referent of the pronoun has a role that is unexpected: it’s surprising that a child star is now a mom; and since Scrooge is known for stinginess, it’s quite unexpected for him to donate generously. The meaning INSISTENCE functions as an alert to hearers, putting them on notice that they should pay attention because there might be something about the referent that is not anticipated.

Reflexive messages are also unexpected, because a single referent is playing more than one role at one time.⁶ In a pragmatic account of pronominal coreference, Levinson (1991, p. 127) observes that “agents normally act upon entities other than themselves; the prototypical action - what is described by the prototypical transitive clause - is one agent acting upon some entity distinct from itself.”⁷ Reflexive uses then are a subset of unexpected messages, as it is unexpected for a single entity to act upon itself and thereby to have two roles in the same event. While reflexivity is not the meaning of self forms, it is one of the message types inferred in part from the meaning INSISTENCE.

The meaning of many lexical items suggests that they involve two participants. For example, the meaning of find suggests someone who does the finding, and something (or someone) that is found. Typically when there are two participants they are distinct, as in the first token of found in (6). When just one participant fills both roles, the self form appears. The meaning INSISTENCE can serve as an alert that there is something unexpected, and it forestalls the unwanted inference that there are distinct entities performing distinct roles.

• (6) Three years ago, when her mother died unexpectedly of cancer, her coach found a new life in Canada, and she found herself alone (Specter 1994)

3.1 Unexpectedness in the number of roles: the Role Conflict Hypothesis

A discrepancy between the number of roles and the number of actors, which is referred to here as a role conflict, is one type of unexpected message. When there is a role conflict, the meaning INSISTENCE may be used to preempt the expectation that the number of roles matches the number of actors. I will refer to this claim regarding the use of self forms as the Role Conflict Hypothesis. I am thus distinguishing between a hypothesis about the form’s stable meaning (INSISTENCE), and a hypothesis about one of its modes of exploitation (Role Conflict) for the unexpected situation when a single entity (or group) has more than one role at one time.

The Role Conflict Hypothesis is not simply a restatement of reflexivity. It does not apply in every syntactically reflexive context, and it does apply in many non-reflexive environments. The role of a referent is not encoded by the self pronoun, nor by a construction. Instead, hearers must always infer speakers’ motivations for INSISTING, and a role conflict is but one possible inference. In addition, there is no structural or formal link between a pronoun and an antecedent that exists independently of the hearer’s efforts to identify the referent.

The domain of role conflict does not depend on any sentence-based structure. A sentential account of anaphora might posit deleted elements in (7) to fulfill a syntactic coreference condition in this book title.

• (7) Diana: In Search of Herself

However, in the semantic approach described here, it is sufficient to note that the interpretation of the utterance involves two roles (a searcher and a searchee), but one person filling both roles.

To recap, here is the chain of reasoning developed so far. The meaning INSISTENCE is often used for unexpected messages. One type of unexpected message is when an entity has more than one role at just one time (the Role Conflict Hypothesis). As we will see in Section 3.7, this hypothesis explains the appearance of self pronouns in reflexive contexts; it also explains the appearance of self forms in the two previously noted categories of counterexamples to the reflexive/anaphor account, picture noun phrases and logophoric uses, to which we now turn.

3.2 Picture noun phrases

It’s well known that in the syntactic environment of a noun phrase that refers to a picture, speakers can choose between self and simple pronouns. The nature of a picture - or any sort of likeness - is that there are always two roles involved, even if both are not signaled: the entity portrayed and the entity perceiving.⁸ In example (8), seen earlier as (3a), the referent of himself has two roles: the orchestral work (the musical picture) was conceived by him, and he is its subject.

• (8) The ‘Enigma Variations’ were first conceived humorously as a series of musical pictures of himself, his wife, and people they knew. Coming home after a grinding round of teaching one Friday evening in 1898, he sat down at the piano and idly played a tune. (Cavendish, History Today 1999)

As noted earlier, meanings are not motivated by just one message element. In this example, the meaning INSISTENCE is motivated by the fact that the referent has two roles, so there is a role conflict. In addition, the referent is important in the context; he is also named as part of a coordinate expression, where INSISTENCE often serves to differentiate the referent of the self form from others.

The Role Conflict Hypothesis predicts that self pronouns may appear in picture noun phrases when the referent of the form has more than one role in the conceptualization of the scene being described. However, if a speaker does not wish to draw attention to the multiple roles of the referent, then simple pronouns may be used, as in (9) to refer to the writer’s birth father.

• (9) After thirty-five years of no contact, she was able to locate him to let him know that I had reached out to her. I was totally blindsided, because I didn't ask her to locate him. I was petrified. This was beyond anything I had expected. Usually, the birth father is not part of the picture. Was he going to reject me, pretend that I didn't exist? I would soon find out. My birth father, David, reached out to me via email. His email was friendly and he included a picture of him with his wife, Thea, and their two sons. (COCA 2012)

Some factors that may have contributed to the choice to avoid the meaning INSISTENCE are that the perspective is not of the pronoun’s referent (a description of the tone of his email is clearly someone else’s perception), and the writer may wish to minimize the importance of the referent, because at this point in the story she is fearing his rejection. In these examples as elsewhere, the choice between self and simple pronouns depends not on the formal features of the context, but rather on speakers’ communicative intent. This is why any attempt to account for the distribution of self pronouns in syntactic terms will fail.

3.3 Logophoric uses

The Role Conflict Hypothesis can also account for logophoric uses, another well-known set of counterexamples to syntactic accounts in which the referent of the self form is the individual(s) whose point of view is represented even when there is no overt clause-internal antecedent (e.g., Cantrall 1974; Kuno 1987). To salvage the syntactic analysis of self pronouns as anaphors, analysts exclude these and other types of examples from the structural account (e.g., Zribi-Hertz 1989, Reinhart & Reuland 1993; Runner 2007). By contrast, the analysis described here applies equally to all uses of the forms.

In example (10), seen earlier as (3b), Bassam is thinking about his daughter Abir, after winning a court case following her murder.

• (10) Here she was again, Abir, multiple versions of her, yet always the same, his gone daughter. Someone touched his elbow. Congratulations, brother. A landmark. Can you believe it? He hung his head. It seemed that the case had happened to someone other than himself, someone out there hovering in a different world. (McCann, Apeirogon 2020)

The reader is witnessing Bassam’s thoughts, and Bassam has two roles: first, he is the cognizer (what Zribi-Hertz 1989 calls the Subject of Consciousness), and second, it seemed the case had happened to someone other than him.

3.4 First and second person, and the problem of reference

One might argue that with first and second person pronouns there is no need to signal that a referent is playing more than one role, because the identification of the referent is not at issue. However, language use pertains to conceptualization and construal, not merely to reference (see Langacker 1987, 1991; Diver 1975/2012). In (11), the meaning INSISTENCE is used to suggest that the forms’ referents are playing more than a single role at one time:

• (11) a. What did I do, Frank? Did I send myself a tape? Did I blow up my car? Did I set my apartment on fire? Did I fire bullets at myself? (COCA 2007)

• b. Go to FutureMe.org and send yourself an email, which you can schedule for delivery at a later date. (COCA 2015)

The meaning INSISTENCE can also be avoided, to suggest the presence of two distinct participants, even if there is but one entity on the scene. The objective characteristics of the scene do not determine the choice of forms; rather, it is speakers’ conceptions of a situation and what they wish to communicate that affect which meanings will be signaled. The next example shows a contrast between the use of myself and me. In both instances, the speaker has two roles on the scene; in the first, the meaning INSISTENCE is present, and in the second it is not.

• (12) Says Mitnick: “When I read about myself in the media, even I don’t recognize me” (Pennenberg, Forbes 1999)

In this passage, a computer hacker is commenting on unsubstantiated stories that have been written about him. He uses myself to INSIST on a dual role for himself: he is reading, and he is being read about. However, when he says I don’t recognize me, he is deliberately avoiding the meaning INSISTENCE to avoid the suggestion that he is playing two roles. Instead, he is describing the situation as if there were two different entities.

Even in third person, when speakers are confident that hearers will have no trouble identifying intended referents, they may avoid the meaning INSISTENCE for similar communicative effects. Example (13) shows a simple pronoun used to refer to an entity already mentioned within the same clause.

• (13) Doctor to 92-year-old patient: You’re in excellent health. You’ll live to be 150.

• Patient’s daughter: Sure she will. She doesn't have her to take care of! (Personal communication 2001)

Prosodic stress on the simple pronoun her in the daughter’s response helps the hearer interpret the utterance, as stress is an additional way that speakers can draw attention to specific words for communicative purposes (cf. Ahn 2015). However, the meaning INSISTENCE was avoided in this example because the speaker chose to describe the situation as if there were two different people involved, rather than one person in both roles.

A further illustration that reference-finding need not determine the choice between self and simple pronouns is found in (14) about the astronaut Alan Shepard. With either pronoun, the reference and the statement’s truth conditions would be the same, but the choice does result in a different interpretation regarding the number of roles played by Shepard.

• (14) His career moved along typically: flight training in Texas and Florida and service on aircraft carriers in the Mediterranean. Then he entered test-pilot training at Patuxent River, Md., elevating himself into the elite of military aviation. When NASA asked 110 test pilots to volunteer to be astronauts, Shepard made the list and was one of the seven chosen ones. (New York Times 1998)

Because the self pronoun appears, the hearer can infer that Shepard plays two roles in the event of elevating - he is also the agent. The same passage with a simple pronoun, (15), shows that the hearer would still identify the form’s referent, but might not infer that it was Shepard’s own actions that led to the change.

• (15) Then he entered test-pilot training at Patuxent River, Md., elevating him into the elite of military aviation (Unattested)

The context in (14) reveals that the writer wanted to distinguish between circumstances that simply unfolded (his career moved along), from what follows the word then, which indicates a new phase in which Shepard was personally responsible for his career path.

3.5 Other types of role conflicts

Many counterexamples to the syntactic account do not fall into identified categories of exceptions. However, the meaning INSISTENCE and the Role Conflict Hypothesis do account for them. For example, there is no evidence that either of the next two examples is logophoric; the reader is not witnessing the referent’s thoughts, but there is in each instance a role conflict.

• (16) By a nearly unanimous vote, Oregon House Rep. Mike Nearman (R) has been expelled from Congress after he assisted protesters in breaching the Capitol and provided them directions on how to do it. The 59-1 vote is the first time in Oregon history a member has been expelled. The 1 “no” vote was himself. (@travisakers 2021)

• (17) Mr. Peña Nieto delivered a 30-minute, wide-ranging televised address meant to pull himself out of the lowest point of his two-year tenure and respond to mass demonstrations expressing frustration with the political class and demanding action. (New York Times 2014)

In (16), the representative has more than one role: the vote was about whether he should be expelled, and the “no” vote was his. Similarly, in (17), Peña Nieto has two roles: he gave the address that was meant to increase support for him. The meaning INSISTENCE is useful to help communicate that entities unexpectedly play more than one role in a single event.

3.6 Simple pronouns in syntactically reflexive environments

The appearance of simple pronouns in syntactically reflexive environments is a major problem for any purely formal analysis, and for any analysis that omits the speaker/writer as an active participant in the communicative event. By contrast, this meaning-based analysis does account for these distributional facts. When the referent’s role is expected, as in (18a) - (c), speakers opt for simple pronouns.

• (18) a. John pulled the blanket toward him (Levinson 1991 p. 116)

• b. She takes the umbrella with her (COCA 1996)

• c. He slowly looks around him (COCA 2004)

• d. John pointed the missile toward himself.

There is no need to INSIST in (18a) since, as Levinson points out, it’s hard to imagine that John could be pulling a blanket toward someone else, and the most likely scenario is that John and him are coreferential. Similarly, when a person takes an umbrella, they are unlikely to take it with someone else. And given the nature of sight, when one looks around it is expected that one looks around oneself. By contrast, Levinson observes that it would be quite unexpected for someone to point a missile toward himself in (18d).

Reflexive environments, then, make room for both simple and self pronouns depending on the details of the context and the intended message. The appearance of simple pronouns in syntactically reflexive environments is anomalous only from the point of view of sentence grammar and the reflexivity account.

3.7 Canonical reflexives and the status of subjects and objects

We have seen that the hypothesized meaning INSISTENCE accounts for the appearance of self pronouns when they contribute to the inference that a referent is unexpectedly playing more than one role at one time. However, mere unexpectedness is not sufficient to account for the use of the forms in what has been seen as the prototypical reflexive construction, such as (19). Although the referent has two roles (he sees and is seen), there is nothing unexpected: when people look in a mirror, the expectation is that they will see themselves.

• (19) When he went to the bathroom, he saw himself in the mirror over the sink (COCA 2018)

Similarly, the meanings of words like behave and perjure also lead to the expectation that the second named participant will be the same as the first, because one cannot behave or perjure someone else.

• (20) a. She behaved herself

• b. He perjured himself

Why, then, don’t speakers use simple pronouns in (19) and (20)? The answer takes the form of a sign-based account of English word order. The English System of Degree of Control (hereafter, the Control System) posits meaningful positional signals that indicate relative degrees of Control exercised by what are traditionally called subjects and objects (Diver 1984; Reid 1991, 2010; Huffman 2009). These grammatical meanings, described below, create the expectation that there will be distinct entities in different roles.

The meanings signaled by the Control System depend on whether there are two participants in an event (traditionally called transitive clauses) or three participants (ditransitives). Control may be volitional and/or it may refer to the extent of involvement or level of participation that entities have in events. In all instances, hearers must infer specific roles for participants, based on the positionally signaled meanings of the Control system as shown in Figure 2 and illustrated in Figure 3.

Figure 2
The English System of Degree of Control

Figure 3
Illustration of Control Meanings

In two-participant events the entity named before the verb is a signal of HIGHER Control, and the entity named after the verb is a signal of LOWER Control. In three-participant events, the entity named before the verb is a signal of HIGH Control; the first entity named after the verb is a signal of MID Control, and the second entity named after the verb is a signal of LOW Control.

By signaling different Degrees of Control for each participant, the meanings of the Control System create the expectation that there will be distinct entities. Once again, the meaning INSISTENCE alerts the hearer that there is something unexpected: the same entity has more than one role.

Space limitations preclude a more thorough description of the Control System (see Reid 1991, 2011; Stern 2016, 2018 for details), but the key point is that a role conflict is created when the same entity is grammatically signaled to have differing degrees of Control in the same event. That unexpected role conflict leads to the appearance of self pronouns.

4. Summary and conclusions

Like Duffley’s work, this investigation has provided further evidence of “what a natural-language semantic approach based on the semiological principle can bring to the analysis” (p. 137). The research described here began not with a universal message category, but by looking instead for the stable semantic contribution of a set of linguistic forms. The analysis has shown that speakers choose between self and simple pronouns when the additional meaning of self forms, INSISTENCE ON AN ENTITY(S), will steer hearers in particular interpretive directions. This approach has led to the discovery that reflexive uses of self pronouns are simply an instantiation of the general tendency to use these forms for unexpected messages, including those in which a single referent is playing more than one role at one time. The presence of such a role conflict accounts not only for reflexive uses, but also for the appearance of self pronouns in picture noun phrases, logophoric contexts, and other previously unexplained exceptions to the structural reflexivity account.

One clarification regarding Columbia School theory is warranted. This analysis shows that, contrary to Duffley’s (2020) criticism, not all Columbia School meanings “divide up semantic domains in an exhaustive way” (p. 189). The meaning INSISTENCE is not divided (exhaustively or otherwise), and is signaled only by self pronouns; it is relational only to the extent that forms with the meaning INSISTENCE contrast with forms that do not carry this meaning. Whatever the merits or flaws of this analysis, it has not been constrained by a theoretical requirement related to exhaustiveness.

For researchers interested in classifying messages, reflexivity may be a useful way to do so. Similarities across languages in the messages that are expressed and the means used to do so are naturally of interest to linguists. However, we have seen that reflexivity is not a linguistic category of English, as it is encoded in neither self pronouns nor in the utterances that contain them. A construct that finds no empirical support in language use is not a solid foundation with which to understand areas like language processing or acquisition, or to learn about other languages. Taking reflexivity, or another a priori grammatical construct, as a starting point “has the unfortunate consequence of giving the analyst the impression that he or she knows all about the semantics of the form in question, when in actual fact it is the semantic content expressed by the linguistic sign which the analytical efforts of the linguist should be endeavoring to uncover” (Duffley 2020, p. 55).

References

• Ahn, B. T. 2015. Giving Reflexivity a Voice: Twin Reflexives in English. University of California, Los Angeles.
• Biber, D., Conrad, S., Reppen, R. 1998. Corpus Linguistics: Investigating Language Structure and Use. Cambridge University Press.
• Cantrall, W. R. 1974. Viewpoint, Reflexives, and the Nature of Noun Phrases. The Hague: Mouton.
• Chomsky, N. 1981. Lectures on Government and Binding. Dordrecht, The Netherlands: Foris Publications.
• _____. 1982. Some Concepts and Consequences of the Theory of Government and Binding. Cambridge, Massachusetts: MIT Press.
• _____. 1995. “Language and nature”. Mind 104 (413), pp. 1-61.
• Clements, G. N. 1975. “The logophoric pronoun in Ewe: Its role in discourse”. Journal of West African Languages, 10, pp. 141-177.
• Contini-Morava, E., Goldberg, B. S. (eds.) 1995. Meaning as Explanation: Advances in Linguistic Sign Theory. Berlin: Mouton de Gruyter.
• Davies, M. 2008- . The Corpus of Contemporary American English (COCA). Available online at https://www.english-corpora.org/coca/
  » https://www.english-corpora.org/coca/
• Davis, J., Gorup, R. J., Stern, N. (eds.) 2006. Advances in Functional Linguistics: Columbia School beyond its origins. Amsterdam: John Benjamins.
• Dieter, S., Wright, S. 1995. Subjectivity and Subjectivisation: Linguistic Perspectives. Cambridge: Cambridge University Press.
• Diver, W. 1984. The grammar of modern English. Unpublished manuscript.
• _____. “Theory”. In E. Contini-Morava and B. S. Goldberg (eds.) (1995), pp. 43-114.
• _____. 2012/1975. “The nature of linguistic meaning”. In A. Huffman and J. Davis (eds.), pp. 47-63.
• Duffley, P. 2020. Linguistic Meaning Meets Linguistic Form. Oxford: Oxford University Press.
• Erker, D., Shin, N. L. (eds.) 2018. First Names: How theoretical primitives shape the search for linguistic structure (Papers in honor of Ricardo Otheguy). Philadelphia: John Benjamins.
• Faltz, L. 1985. Reflexivization: A study in universal syntax. New York: Garland Publishing Inc.
• Givón, T. 1993. English Grammar: A Function-Based Introduction. Vol. 2. Amsterdam: John Benjamins.
• Goldberg, A. E. (ed.) 1995. Conceptual Structure, Discourse and Language. Stanford, California: Center for the Study of Language and Information.
• Haiman, J. 1983. Iconic and economic motivation. Language, 59(4), pp. 781-8193.
• _____. 1995. Discourse, Grammar and Typology: Papers in honor of John W.M. Verhaar (Studies in Language Companion Series), Vol. 27. T. Givón, W. Abraham and S. A. Thompson (eds.), Amsterdam: John Benjamins.
• Huang, Y. 2000. Anaphora: A cross-linguistic study. New York: Oxford University Press.
• Huang, Y. 1991. A neo-Gricean pragmatic theory of anaphora. Journal of Linguistics, 27(1), pp. 301-335.
• Huffman, A. 2001. “The Linguistics of William Diver and the Columbia School”. Word, 52 (1), pp. 29-68.
• _____. 2009. Modern English: A Columbia School grammar. Unpublished ms.
• _____., Davis, J. (eds.) 2012. Language: Communication and human behavior. The linguistic essays of William Diver. Leiden/Boston: Brill.
• Kemmer, S. 1995. “Emphatic and Reflexive-self: Expectations, Viewpoint, and Subjectivity”. In S. Dieter and S. Wright (eds.), pp. 55-82.
• _____., Barlow, M. 1996. “Emphatic-self in Discourse”. In Adele E. Goldberg (ed.), pp. 231-248.
• König, E., Gast, V. 2002. “Reflexive Pronouns and Other Uses of Self-Forms in English”. Zeitschrift Für Anglistik Und Amerikanistik, 50 (3) pp. 1-14.
• Kuno, S. 1987. Functional syntax: Anaphora, discourse and empathy. Chicago, IL: University of Chicago Press.
• Langacker, R. W. 1987. Foundations of cognitive grammar: Theoretical prerequisites. Vol. 1. Stanford University Press.
• _____. 1991. Concept, image, and symbol: The cognitive basis of grammar. R. Dirven and R. W. Langacker (eds.), Berlin: Mouton de Gruyter.
• Levinson, S. C. 1991. “Pragmatic Reduction of the Binding Conditions Revisited”. Journal of Linguistics, 27, pp. 107-161.
• Quirk, R., Greenbaum, S., Leech, G., Svartvik, J. 1985. A Comprehensive Grammar of the English Language, New York: Longman.
• Reinhart, T., Reuland, E. 1993. “Reflexivity”, Linguistic Inquiry, 24, pp. 657-720.
• Reid, W. 2011. “The Communicative Function of English Verb Number.” Natural Language and Linguistic Theory, 29, pp. 1087-1146.
• _____. 1991. Verb and Noun Number in English: A Functional Explanation. New York: Longman.
• Rolston, L. (ed.) 2016. University of Washington Working Papers in Linguistics. https://depts.washington.edu/uwwpl/editions/vol34.html
  » https://depts.washington.edu/uwwpl/editions/vol34.html
• Runner, J. T. 2007. “Freeing Possessed NPs from Binding Theory.” In L. Wolter and J. Thorson (eds.) University of Rochester Working Papers in the Language Sciences, 3(1), pp. 57-90.
• Sperlich, D. 2020. Reflexive Pronouns: A Theoretical and Experimental Synthesis (Language, Cognition, and Mind, Vol. 8). Springer. https://doi.org/10.1007/978-3-030-63875-7
  » https://doi.org/10.1007/978-3-030-63875-7
• Stern, N. 2004. “The semantic unity of reflexive, emphatic, and other-self pronouns”, American Speech 79(3), pp. 270-280.
• _____. 2006. “Tell me about yourself”. In J. Davis, R. J. Gorup and N. Stern (eds.), pp. 177-194.
• _____. 2016. “Word order as a signal of meaning: English reflexive pronouns and why we behave ourselves”. In L. Rolston (ed.), pp. 111-119.
• _____. 2018. “Ditransitives and the English System of Degree of Control: A Columbia School analysis”. In D. Erker and N. L. Shin (eds.), pp.157-188.
• Van Hoek, K. 1997. Anaphora and Conceptual Structure. Chicago, IL: University of Chicago Press.
• Zribi-Hertz, A. 1989. “Anaphor Binding and Narrative Point of View: English Reflexive Pronouns in Sentence and Discourse”. Language, 65, pp. 695-727.

• 1
  For full theoretical statements, see Diver 1995; Huffman 2001; for a bibliography of Columbia School work, see www.csling.org.
• 2
  The term ‘self pronoun’ is used to avoid any analytical claim that is implied by the label ‘reflexive’.
• 3
  Following Kemmer and Barlow (1996), the term ‘argument’ is applied here not only to participants selected by the verb, but to any entity with its own role, including objects of prepositions (e.g., Bruce was looking at a picture of Sue in the newspaper), where Bruce, picture, Sue, and newspaper are all described as arguments.
• 4
  Data for this study are drawn from actual instances of language use, including tokens from my own collection as well as many from the electronic Corpus of Contemporary American English, also known as COCA (Davies, 2008-). In contrast to studies based on grammaticality judgments or introspected data, attested examples include context that provides evidence of speakers’ communicative intentions, and allow analysts to “investigate how speakers and writers exploit the resources of their language” (Biber, Conrad & Reppen, 1998, p. 1).
• 5
  Other linguists (e.g., König & Gast 2002; Ahn 2015) have observed that some argument self forms function emphatically, but they do not apply that analysis to syntactic reflexives.
• 6
  Givón (1993, p. 89) suggests that part of the definition of the reflexive construction is that “The same referent participates in the clause in two different roles”.
• 7
  Huang (1991, 2000) and Sperlich (2020) also propose a role for a syntactic-pragmatic continuum with respect to anaphora, but conclude that “English reflexive pronouns appear to be syntactically constrained....” (Sperlich 2020, p. 2).
• 8
  Kuno (1987) and Van Hoek (1997) observe that the use of self pronouns corresponds to point of view/awareness effects, implicitly recognizing the additional role of the form’s referent.

Article info

• 10
  CDD: 412

• Erratum

  In the article “REFLEXIVITY, ROLE CONFLICTS, AND THE MEANING OF ENGLISH SELF PRONOUNS”, with DOI: https://doi.org/10.1590/0100-6045.2022.V45N1.NS, published in the journal MANUSCRITO: REVISTA INTERNACIONAL DE FILOSOFIA, Vol. 45, n. 1; pp. 90-116, on page 90:
  Where it reads:
  “The City College of New York and Graduate Center, CUNY
  Department of Philosophy
  New York, N.Y.
  U.S.A.”
  It should read:
  “The City College of New York and Graduate Center, CUNY
  New York, N.Y.
  U.S.A.”

Publication Dates

• Publication in this collection
  21 Mar 2022
• Date of issue
  Jan-Mar 2022

History

• Received
  12 Aug 2021
• Reviewed
  28 Sept 2021
• Accepted
  24 Oct 2021