Pejoratives as Social Kinds

Share Embed


Descrição do Produto

Pejoratives as Social Kinds: Objections to Miščević’s Account

Mirela Fuš

Abstract1 My aim in this paper is to present and discuss Miščević’s position on pejorative terms. Pejorative terms, for Miščević, are negative hybrid social kind terms that refer directly and pick out social kinds as their referents. Despite sharing some of Miščević’s intuitions on pejorative terms, I raise three main objections to his account. First, I argue that introducing pluralistic commitments about propositions is not helpful in any way for his account. On the contrary, I show that it brings about the falsity of meaning and the failure of reference. Furthermore, I point out that Miščević is mistaken in trying to avoid the problem of co-reference between terms such as “Boche” and “German”. I argue that there is no co-reference present because the referents of these two terms belong to two different social kinds. Finally, I raise a more general objection connected to the reference of social kind terms and, consequently, to pejorative terms as well. I bring up a worry that certain referential theories about pejorative terms, Miščević’s account included, might lack an explanation of what ties a social kind token to a particular social kind type.

Keywords: direct reference, natural kind terms, social kind terms, pejoratives

1

Early versions of this paper have been presented at the Language, Mind, and Knowledge conference (June 2014, Zagreb (Croatia)) and at the Mind, World and Action conference (August 2014, Dubrovnik (Croatia)). I would like to thank the audience for their useful questions and comments, especially to Timothy Williamson and Julija Perhat. Furthermore, I am grateful to Dunja Jutronić and Michael Devitt for their discussion and suggesting literature on social kinds and artifacts. Recent versions of this paper have been written while I have doing research at St. Andrews (UK) and Graz (Austria). I am also grateful for the feedback and many fruitful discussions I had, including: Aisling Crean, Patrick Greenough, Eliot Michaelson, Andrea Onfori, Walter Pedriali, and Timothy Williamson. I wish to warmly thank to my two temporary supervisors for the extensive feedback and helpful comments: Cristina Borgoni Goncalves, and Ephraim Glick. At last, many thanks to Nenad Miščević, for his support with my work in progress, and for challenging conversations on his theory.

Introduction Miščević’s (2014) idea is that one should seek help from the social ontology in order to solve the problem of reference for certain groups of pejoratives. His view is, thus, not in accordance with the views that say that the problematic phenomena of pejoratives we face, especially in terms of a negative material they contain, falls under the scope of language solely, and not under the scope of the ontology of social kinds. Instead, the literature on pejoratives has often treated pejoratives as linguistic phenomena under the linguistic domain. Pejorative terms such as “pussy”, “Nigger”, “Paddy”, “Boche”, etc. are sometimes taken to function as attributes or epithets. Usually, in the case of attributes or epithets (“a is F”, where “a” is a singular term, and “F” stands for a certain attribute or epithet) a predicative view is offered to explain their semantics. Consequently, pejorative terms are sometimes treated under the predicative view. The semantic function of a predicate typically thus takes to pick out a property or a set of things.2 On the other hand for those aiming at offering up a complete semantics for certain linguistic terms, they are interested both in theory of meaning as well as in theory of reference for the terms in question. In particular, those interested in a complete semantics of pejorative terms aim, apart from giving a theory of meaning for pejorative terms, to examine if there is and what would be the mechanism of reference for pejorative terms. In this paper, I focus on the possibility of a direct reference3 for pejoratives terms as social kind terms. Miščević (2014) explores such a possibility by taking a certain group of pejorative terms as negative hybrid social kind terms that can refer directly. He himself offers

2

However, when deciding on a possible theory of reference for pejorative terms, the predicate view seems not to be the best option. Such an account would require of one to give a theory of reference for properties or sets. This seems to be an utterly challenging project per se, and I will not pursue it here. 3

Direct reference puts one into a position to refer singularly. In order to have a singular thought, roughly speaking, it means to be in a position to think about a particular object, not about any object that might come to satisfy certain qualities or properties. Singularists have defended the view that there is a difference in content between singular and descriptive thoughts. The content of the former is a singular proposition (having properties and individuals for constituents), while the latter takes a general proposition as its content (consisting of concepts or properties). As for a cursory mention, there are three dominant theories of singular thought, namely Acquaintance Theory of singular thought (see Burge 1977, Donnellan 1979, Lewis 1979, Evans 1982, Boer and Lycan 1986, Bach 1987/94, Salmon 1988, Brewer 1999, Recanati 1993, Soames 2003, Pryor 2007), Semantic Instrumentalism (see Harman 1977, Kaplan 1989) and Cognitive Authority or Cognitivism (see Jeshion 2002, 2009, 2010). (see Fuš 2013, 201)

an account of “hybrid theory of reference” which he adopts from Devitt and Sterenly (1999)4 , thus extending it to social kinds5. However, there is an important restriction on the scope of the application of Miščević’s theory. A theory of reference for pejorative terms that Miščević offers applies only to a certain group of pejorative terms, namely to pejorative terms that are presupposed to have a social kind as their referent. This means that his theory of reference for pejoratives does not apply to all words that we consider to be pejorative terms. He is, rather, establishing it for the pejorative terms that can function as social kinds. Consider the terms “German” and “pussy”. The first one, according to Miščević, picks out a social kind whereas the second one, most likely, doesn’t6. Thus, according to Miščević, certain pejorative terms do not refer to a predicate—they rather pick out a social kind. For instance, if we say that a term “German” represents a social group or a kind in our ontology, then a pejorative term “Boche” also represents a social group or a kind. The idea that certain pejorative terms are referring to social kinds is a rather controversial one and, thus, it would take a bit more to explain its value. Nevertheless, it also seems to be an extremely relevant one7 . What, I believe, justifies Miščević to offer a theory of reference for pejorative terms as social kinds terms lies on the presupposition that a referent of a certain group of pejorative terms belongs to a social kind category in ontology8.

4

In “Against Direct Reference” Devitt (1989, 464) already advocated a version of direct reference he calls the Causal Theory: “A name designates an object solely in virtue of a certain sort of causal or historical chain connecting the name to the object.” 5

This paper attempts to fit within the picture of a wider project where I want to look at the possibility of a direct reference for the social and natural kinds. The general motivation to offer a theory of reference for social kinds comes from the analogy or rather the extension from the theory of reference that was given for natural kinds. 6

This said, it might be dubious whether words such as “Nigger” (that he actually uses as one of his main examples) fall under the social kind pejorative terms. Some might argue that such terms could, perhaps, be better adopted under the predicative view. 7

This, of course, involves certain ontological commitments the referent of a pejorative term bears. I hold that a shift from philosophy of language to the ontology of the referent appears as a one worth pursuing because it might give an answer to the problems other views struggle with. I will tackle some of these issues in the second part of this paper. 8

However, he does not stress this crucial point sufficiently—instead he somewhat tacitly introduces the notion in his paper and is often not clear on the fact that his theory is committed to the notion that at least some pejorative terms refer to a certain social kind. Though I believe that his theory is on the right track, this confusion, finally, results in the failure of his theory of reference, as I will try to show in the second part of this paper.

Furthermore, I believe that there are at least two big tasks that a theory of reference for pejorative terms as social kinds should answer: first, to justify the ontology of referents for certain pejorative terms as social kind terms; second, to offer a successful mechanism of reference for this ontology. The first task remains unanswered by Miščević and it will not be directly addressed in this paper, for Miščević does not offer the justification for the ontology he deals with. Rather, he presupposes it as a given. However, I will try to tackle it briefly in the second part of this paper while mentioning certain benefits that stem from it. Miščević, however, deals with the second task, namely he wants to offer a successful mechanism of reference for pejoratives. Thus, the aim of this paper is to evaluate the mechanism of reference for a hybrid social kind terms that he offers. I will proceed as follows. In part one, I summarize Miščević’s (2014) theory of pejorative terms as social kind concepts9, while also focusing briefly on presenting the quality dimensions in conceptual spaces he adopts from Gärdenfors (2000, 2014) in order to analyze the meaning of pejoratives. In part two, I raise three main objections that can be applied to Miščević’s account (and, perhaps, compatible accounts). First objection is about: Pluralistic commitments about propositions with regard to dimensions of pejorative bring about the falsity of the pejorative sentence; thus, on one hand, making the meaning false and, on the other, also do not contribute to the reference fixing. The second objection is about: Pejorative terms such as “Boche” and “German” when treated as social kind terms do not co-refer since their reference is to two different social kinds, namely social kind Boche and social kind German. The third objection is about: Miščević’s account of reference for pejoratives as social kinds cannot adequately explain what ties pejorative tokens to a particular pejorative social kind type.

9

In this paper I use terms ‘social kind concept’ and ‘social kind term’ interchangeably.

Part I: Miščević’s account of pejoratives 1.1. Background In his recent paper Miščević (2014) intends to offers a complete semantic account of pejoratives. In this section I explain Miščević’s theory of pejoratives, but first I contrast it with existent approaches. There have been two dominant traditions of theories of pejoratives, i.e. expressivism (offered by Frege (1979, 141), his example of “Köter”, (“cur”)), and semanticism (e.g. Dummett 1973). According to Frege, the negative aspect of pejorative terms does not belong to semantic content, but to the speaker’s attitude expression. In contrast, Dummett indicates that the negative aspect of pejorative term influences the truth value of a pejorative sentence. Miščević belongs to the semanticist tradition. We can, thus, map this semanticist’s architecture for pejorative terms with help of the scheme Miščević (2014) suggests.

Figure 1

Miščević (2012a, 2012b, 2014) takes that the “bad material” or a negative element attached to pejorative terms is a part of their meaning. Furthermore, he also takes Dummett’s truth-conditional semanticists’ side, as opposite to non-truth conditional (e.g., Potts (2005), Williamson (2009)10), since he believes that the negative element in pejorative terms can affect the truth

10

He also embraces Williamson’s (2009) rejection of Dummett's inferentionalism.

value of the pejorative sentence. In contrast to Dummett who was an inferentialist, Miščević believes that the negative part in pejorative terms is included in the concept of the term itself. However, Miščević’s position is to be distinguished from other possible positions such as Hom’s (2008, 2010). Hom offers a descriptive theory of pejoratives. Eve though he also believes that the negative element is a part of a pejorative term or a concept, Hom holds that pejorative terms do not refer. Finally, Miščević gives a truth-conditional pluralist alternative and offers a theory of pejoratives as negative hybrid social kind terms that I elaborate on further in the following parts of this section. To reiterate, Miščević aims at giving a model for semantics of pejoratives. He includes two desiderata within his semantics of pejoratives. First, he claims that pejoratives do refer directly. Second, negative-evaluative dimension he introduces is part of the meaning (and not of the reference).11 Let us place these desiderata into the proper context. There is usually an inconsistent triad that different accounts of pejoratives have to deal with (see Figure 2). (a) Pejorative terms do refer.

(b) The negative part is part of the reference.

(c) Pejorative proposition is false.

Figure 2

Under the semanticist view, the negative part of pejorative term, depending on whether included or not in securing the reference, has different effect on the reference and on the truth validity of a pejorative proposition. In other words, it seems that a pejorative term does not refer when its negative part determines the reference, for its negative part brings about the falsity of the pejorative proposition. Here are some of the proposals that try to avoid the falsity of a pejorative proposition by denying one of the propositions from the triad:

11

I discuss the second desiderata in great detail in the second part of this paper.

(i) Hom’s (2008, 2010) account treats the negative part as a part of the reference that makes the pejorative proposition false. Thus, the pejorative sentence does not refer but rather includes a pejorative term as its descriptive concept: (b) & (c) & -(a).

(ii) Williamson (2009) argues that pejorative terms do refer. The negative part is not part of the reference and the reference is fixed via the conversational implicature: (a) & (c) & -(b).

(iii) Miščević’s (2014) position claims that pejorative terms do refer. He excludes the

negative part from the reference and thereby he supposedly avoids the falsity of a pejorative proposition with help of his pluralistic commitments: (a) & -(b) & modified (c).

1.2. Miščević’s dimensions of offensiveness: Pejoratives terms as social kind terms Miščević (2014) develops a representational view of the meaning of pejoratives based on Gärdenfors’ (2000, 2014) idea of geometry of representations. For Gärdenfors (2014, 21) “conceptual spaces are constructed out of quality dimensions.” The primary role of the dimensions is to represent various “qualities” of objects in different domains. Miščević (2014, 196) explains that the “notion of a dimension, [Gärdenfors] insists, should be understood literally.” However Miščević stresses that, for his purposes, he will “concentrate upon value; [he] would like to apply Gärdenfors’ picture of conceptual space to an area of language that has recently attracted a lot of interest, namely pejoratives or slurs.” Miščević takes words such as “Hun”, ”Ustasha”, “bitch” and “faggot” to correspond to pejorative concepts that “refer to moderately clearly identified groups of people, or to individuals under the guise of belonging to the group; either social (“Croat”), or psychological (“homosexual”). So, it is social kind terms, plus psychological kind terms that we should look at.” (Miščević 2014, 198). Furthermore, Miščević offers different dimensions of pejoratives or dimensions of offensiveness, namely (i) the neutral minimal descriptive dimension; two negative dimensions: (ii) descriptive-evaluative and (iii) prescriptive; and (iv) the expressive dimension. In Miščević’s words:

The first dimension has to do with neutral, non-evaluative sense: for a given class (group, social kind) K it delineates the basic causal-cum-descriptive components that determine the intended reference of the pejorative (say, the social kind “gays” for “faggot”). The second comprises the evaluative components ascribed to K, together with their associated descriptive bases. The third is a prescriptive one, suggesting how badly the target is to be treated. The fourth is expressive of speaker’s negative attitude towards members of K. The last three dimensions suggest that the concept associated with a pejorative is a thick concept, whose non-empty extension, is, however, determined by the first, neutral dimension. (2014,197)

The four dimensions are represented in the table below for the pejorative term “Nigger” Miščević offers.

Figure

3

According to Miščević (2011, 2014), the pejorative term “Nigger” is a thick concept that informs us that a word is about an Afro-American who is primitive, lazy and dangerous; furthermore that his individual should be avoided and discriminated and one should be disgusted by him or her. It is important to note that Miščević considers only the neutral causal-descriptive dimension to determine the reference. According to his account, the negative dimension does enter into the meaning, but it does not enter into fixing the reference. Miščević holds his theory for pejorative terms to be compatible with causal or causaldescriptive theories of reference12. He develops a theory of reference that treats pejoratives as negative social kind terms. He also considers it to have a hybrid nature since, for him, the reference includes both a causal-historical link and a descriptive part. He says: 12

See Devitt 1989; Devitt and Sterenly 1999.

So, mine is a social kind term hypothesis: ethnic, gender and similar pejoratives stand for presumed (and often real) social kinds to which they ascribe all sorts of negative properties. Since it combines causal account with the descriptive approach, it would be a hybrid theory, all in all a hybrid social kind theory. (Miščević 2014,198; italics mine)13 For my further analysis it is crucial to consider that Miščević treats pejoratives as social kinds rather than social kind terms. His use of the term “pejoratives” is ambiguous between standing for terms that are pejorative on one hand, and kinds that are social, namely what social kinds pejorative terms refer to. However, I will argue that he does not mean that the reference is to a social kind term, namely to a term “pejorative”, but rather to a social kind14 . Thus, in what follows, I will take his theory of pejoratives and the use of the word to be about social kinds, and not social kind terms.

1.3. Grounding the reference within Miščević’s account Miščević claims that the neutral (minimal) descriptive dimension reveals “the factual information about the target group, and contributes to securing the reference of the term” (Miščević 2014, 198). Furthermore, he suggests to adopt the hybrid theories of reference, such as the one offered by Devitt and Sterenly (1999)15. Their theory includes both a causal-historical account of naming as well as descriptive dimension of meaning: So, let me now restate the main point of my proposal: pejoratives, say “N”, are negative (derogatory) social kind terms, with hybrid nature. Their reference is partly determined by causal chain: the target group or social kind K has been called by somebody “N”, the name has been transmitted to the present users, and it refers to K and its members. (Miščević 2014, 200)16

13

See footnote 14.

14

Notice that in the quote that footnote 13 refers to Miščević claims that pejoratives are social kinds, whereas in the quote that footnote 16 refers to Miščević claims that pejoratives are social kind terms. 15

There have been some attempts to extend the notion of direct reference even further, namely to natural kind terms and, possibly even artifacts. Devitt and Sterenly (1999) proposed such a hybrid causal theory of reference, opening some questions for the cases of artifacts (e.g. pencil, pediatrician). 16

See footnote 14.

For example, being in a direct or indirect17 epistemic contact with a homosexual provides the speaker with a causal-epistemic link needed when referring via pejorative term “faggot”. And the rest (of the meaning) she gets from the descriptive-evaluative dimensions of a term in question. Let us move one step back to explain how Miščević arrives at the formulation above. He refers to the idea developed by Devitt and Sterenly (1999) about the hybrid theories of reference. They argued for the possibility of a theory of reference that includes elements from both descriptive and causal theories of reference. For them, in case of kind terms, or at least natural kind terms, the causal part comes from the samples or tokens of the particular kind, which grounds the reference. On the other hand, the descriptive part gives a description of the function of a term. In the end, the reference is to anything that has the same function as the sample, thus enabling the reference to be borrowed. Miščević identifies three components that Devitt and Sterenly (1999) claim to “determine the reference: first, the original naming, second, the transmission of the term up to the present use, and third the description, appealing to the intended use and to the fact that the object has” (2014, 198-199). Miščević also stresses what Devitt mentions as “cases where religious or social beliefs mystify the function of certain artifacts” (Devitt 1999, 99) in order to show the similarities between “mystification of the function” and a “negative dimension” of a pejorative. Furthermore, Miščević utilizes this as motivation to accommodate the negative dimension of pejorative into its meaning. Based on the use of Devitt and Sterenly’s ideas Miščević further explores an example from astrology, i.e. “Scorpio”, which he takes to be a hybrid name. He believes that term “Scorpio”18 functions as what he calls a problematic “psychological kind”; this reference is, nevertheless, possible, for it has a causal-historical link as well as a descriptive part. He claims:

17

The causal theory of reference for natural kind terms given initially by Kripke and extended by Putnam includes perceptual or descriptive dubbing in the presence of samples of a natural kind in question. Borrowing the reference via communicative exchange is, thus, also possible no matter the ignorance of the borrowers to the properties of the kind in question. He develops the same point using examples such as “teacher” (see 2014, 199), “medicine man” (see 2014, 199). 18

Similarly, with astrological terms like “Scorpio”. The meaning of the term involves the date of birth, and assumption about character traits (dangerous, intelligent, etc.). So, “Nenad is a Scorpio” is true of myself, the author of the paper, born on Nov. 1, but I doubt that I have the requisite ‘Scorpio’ traits, and even if I have them, I am sure that this is not the case in virtue of me being born at the right interval. (Miščević 2014, 200)

The meaning of an astrological term “Scorpio” involves various descriptive elements of content, some of it obviously false. What secures the reference to a kind “Scorpio” in a sentence “Nenad is a Scorpio” is the neutral descriptive part of the term “Scorpio” pulled out of the whole descriptive content or the meaning that term consists of and put together with the causal-epistemic contact the speaker has with Nenad. Finally, after admitting a possible application of such a hybrid theory to the social and psychological kind terms, he utilizes it with pejoratives: The parallel between “Boche” and “medicine man”/“Scorpio” is clear. “Boche” would fail to refer, in virtue of its derogatory content alone, like “medicine man”, strictly taken in its superficial meaning, would. But the causal and minimal descriptive part do secure the reference, and we, hearers, reconstruct the reference to Germans, from the minimal neutral descriptive part, the way we do with other sentences involving dubious kind terms. (Miščević 2014, 200)

Miščević argues that a pejorative term “Boche” shares its content structure with terms such as “medicine man” or “Scorpio”, meaning that they all have a false component in their content attached to the meaning that would cause the reference to fail. However, he believes what secures the reference are not all descriptive components of content in the structure of the meaning (or dimensions) he introduces, but rather only a causal-historical link to the sample of the kind plus the neutral minimal descriptive part. I will raise my doubts regarding this move taken by Miščević in the second part of this paper.

1.4. Truth conditions for dimensions of meaning and pluralism about propositions I focus now on briefly summarizing the dimensions of meaning with regard to their contribution to the truth conditions within Miščević’s account. On Miščević’s (2014) account, (i) the

causal-historical and (ii) the negative descriptive evaluative dimensions (or layers) of meaning are strictly semantic. He presupposes each dimension to function as a proposition, and, thus, the former (i) is neutral and true, whereas the later (ii) is usually some prejudice and is, thus, false. Furthermore, (iii) the expressive meaning could be treated as actual speech act, i.e. belonging to pragmatics, or could be, less strictly, adopted within the semantics. Miščević is emphatic on how to treat the last dimension, i.e. (iv) the prescriptive dimension—he takes it to be semantic so far as “it can be gauged from the fixed semantic material. Is it truth conditional, or does it, at least, show some CI-features? I shall leave this question open.” (Miščević 2014, 207) In connection to the truth conditions of different dimensions of meaning, Miščević agrees with the pluralists about propositions (e.g. Potts and Bach)19. They claim that certain sentence can express different proposition, and which respective proposition is being stressed depends on the context. However, Miščević does not accept that the true proposition is the basic one. He offers the well know example from the literature:

“Lessing was a Boche” expresses at least two propositions: first, L. was a German. the true and decent one, and then L. was cruel because German. the false (and indecent) one. But if nobody is cruel just because German (or dangerous because gay, etc.) then “Boche”/“Hun” (and “faggott”) do not refer. Some authors, like C. Hom (2008, 2010), are ready to swallow this counterintuitive consequence. (Miščević 2014, 202)

Miščević believes that the hybrid social kind theory can answer the problem of “false components in their senses: one can refer to sofa even if one has false beliefs about it, as Devitt has noted (1999: 98)” (Miščević 2014, 203). According to this, a kind term could refer even when associated with “partly false concepts”.20 Moreover, he is convinced that pejoratives: 19

“I accept the idea of the plurality of propositions expressed by sentence figuring a pejorative, proposed by Potts (2005) and Bach (web), who claim that a typical pejorative sentence expresses a plurality of propositions. In contrast to Potts, I will not assume that they are organized by relations of conventional implicature.” (Miščević 2014, 202) 20

“The same can happen with natural kind terms: for example, at the beginning of the study of electrons, they were defined in a manner involving false hypotheses, e.g. that they are the smallest particles out of which all matter is made. What won was the causal connection, linking the use of the term with the phenomena observed in primitive cathode ray tubes.” (Miščević 2014, 203)

… refer thanks to their first dimension, the neutral one (involving probably neutral descriptive plus causal elements). This kind of determination of reference, where sense contains false elements, both predicts the extant trouble with the truth-status, and offers the means to solving it. (Miščević 2014, 203) 21 Let me conclude this part of the paper by reminding that Miščević’s account of pejoratives depends on securing the reference by relying on the causal-descriptive theory of reference and plurality of propositions. Some of the expressed propositions happen to be true (the neutral ones) and some false (usually the derogatory ones). He holds that their heterogeneous content is clearly something that pejorative terms share with other social kind terms.

1.5. The positive remarks Before turning towards objections to Miščević’s theory of pejoratives I wish to point out the positive aspect of his account. Miščević’s theory of pejoratives might still function as a valid theory when it comes to a theory of meaning for pejorative terms, even if it turns out that his offer fails to secure the reference. It is also beneficial to keep in mind that, perhaps, not all pejorative terms refer since not all of them have social kinds as their referents. In addition to what is mentioned in the introduction to this paper, namely even if it is true that some pejorative terms are referring terms, there are going to be other pejorative terms which are clearly predicates (or epithets) yet that might still have a lot of characteristics they would share with the ones which might refer. Let us suppose “gay” is a neutral term and, say, “faggot” is a referring term. Then a word such as “faggy” when used as an adjective can still be insulting and differ from “gay” or “gayish”—therefore, one wants to be able to theorize semantically about those kinds of terms as well. In the end, a neutral theory of meaning for pejorative terms might still be welcome and, perhaps, Miščević’s dimensions could facilitate in explaining some of the interesting characteristics of the meaning of pejorative terms no matter whether they are or are not referring terms.

21

See footnote 25.

Part II: Objections to Miščević’s account 2.1. First objection: Pluralistic commitments about propositions bring about the falsity of the P-sentence22 ; thus, on one hand, making the meaning false23 and, on the other, they also do not contribute to the reference fixing. In Part 1 of this paper, at the beginning of a section 1.4., I supplied a summary of Miščević’s dimensions of meaning with regard to their contribution to the truth conditions. Let me utilize the aforementioned example of a pejorative term “Nigger” in order to look at a truth value of the sub-propositions, which, on his account, are different dimensions:

EXAMPLE

DIMENSIONS

MEANING

TC or CI

The neutral causal-

someone called

true

descriptive dimen-

them Afro-Amer-

sion

icans

“Nigger”

Semantics

The negative dimension: descriptive-evaluative

The negative diSemantics or Pragmatics

mension: prescriptive The expressive di-

primitive, lazy,

false

dangerous

to be avoided

,

false/CI?

discriminated! Yuck! and more

true/CI?

mension

Figure 4

22

Miščević uses the P-sentence to stand for a pejorative sentence that is complex and expresses a plurality of propositions, some of which are true and some false. I explain its complexity in this section. 23

For our purposes, the meaning of proposition can be defined as a function that picks out possible worlds in which the proposition is true.

Miščević believes that implementing pluralism about propositions enables him to offer different dimensions of meaning; no matter what their truth value is. Thus, according to him, dimensions of a pejorative give meaning to a pejorative, and these dimensions are taken to be different sub-propositions that P-sentence expresses. Furthermore, when it comes to a reference-fixing of a pejorative or a P-sentence, Miščević insists that the neutral causal-descriptive dimension alone is enough to secure the reference. At this point one might wonder why Miščević does not include all the layers in order to secure the reference? Especially since he is committed to a hybrid theory of reference which, apart from the causal, also includes a descriptive part. This would further mean that, potentially, all of dimensions that are part of the meaning (which is descriptive) could be included in securing the reference, along with the causal link. However, there is a very simple answer why not to include them all. The reason is that the negative part included in the meaning is false and would bring about a failure of reference. Clearly, Miščević wants to rely on the only necessarily24 true sub-proposition at hand in order to avoid the reference failure. He must be aware that if he were to include the negative descriptive-evaluative dimension as the one that contributes to the reference-fixing too, the term would not refer anymore. Let me explain further how the pluralistic commitments contribute to this result or, rather, prevent him from including the negative dimension of the meaning to secure the reference of the pejorative term. The truth validity and the negative-descriptive dimension are connected via pluralistic commitments since different dimensions are treated as different propositions to which, then, one can ascribe truth values. As quoted above25, Miščević rightly anticipates the trouble with the truth-status of Psentence though he believes he can circumvent it by his commitment to the pluralism of propositions. I argue that, on the contrary, Miščević’s theoretical choice brings him closer to a further trouble with truth-status of P-sentence, namely to the necessary falsity of a complex pejorative proposition. The pluralistic commitments combined with the inevitable fact that in a case of pejoratives a “sense contains false elements” (Miščević 2014, 203) make, thus, the whole P-sentence necessarily false. This is simple to see. The meaning of a P-sentence can be 24

To press even further, one could think of pejorative terms where even the neutral causal-descriptive dimension is sometimes not true or where the negative descriptive-evaluative dimension is not false. For example, all niggers are not necessary black can, under some interpretations, cover both versions. 25

See the quote that footnote 21 refers to.

either true or false. The P-sentence is a complex proposition consisting of its sub-propositions, which are different dimensions of the pejorative he offers. The conjunction of these dimensions determines the meaning of a P-sentence. In other words, some of the dimensions Miščević introduces can turn out to be true or false. These are at least the neutral causal-descriptive dimension and the negative descriptiveevaluative one. As mentioned earlier, Miščević stays undetermined on the status of treatment the negative prescriptive and the expressive dimension as a truth conditional or as a conventional implicature. However, if there is a P-sentence consisting of certain sub-propositions of which at least one is false, the complex proposition, namely P-sentence, cannot be true. It has to be false. In fact, even if there were only two semantic dimensions in his theory, if one were false, the P-sentence would turn out false. As it often turns out, in the case of pejoratives, that the negative descriptive-evaluative dimension is a false sub-proposition. The subsequent problem is that if the complex proposition is false how it can, then, refer? Can one isolate only the causal link together with the neutral and true dimension alone to fix the reference and neglect other sub-propositions that are offered? It seems the answer is that one cannot. Thus, my further objection to Miščević is why, while giving a sense or dimensions of a pejorative, would one want to appeal to the idea of plurality of propositions and their truth validity if the other sub-propositions do not contribute to fixing the reference after all? Furthermore, when it comes to the meaning of a pejorative, pluralistic commitments about propositions turn out to be in no way useful since they leave us with a false P-sentence or a false sense. On the other hand this choice also does not prove itself to be necessary at all since when it comes to fixing the reference since Miščević only takes the neutral, the true subproposition into consideration—not all of them. Therefore, I believe that introducing pluralistic commitments to treat these dimensions does not contribute to Miščević’s theory: it does not contribute to the meaning of a term since it makes the meaning false, as well as it does not determine the reference since Miščević does not want to include the negative proposition in fixing the reference for it would bring about the failure of reference. One can perhaps say that there is another motivation to advocate the pluralistic commitments, namely to recognize that a pejorative term “Boche” has more layers than an ethnic term “German” and thus avoid the problem of co-reference. However, in the next section I

will show that there is no need for pluralism because there is no co-referring between the two terms in question.

2.2. Second objection: A “delayed qua problem”; “Boche” and “German” do not co-refer since their reference is to two different social kinds. I believe that the crux of the Miščević’s ambiguous treatment of pejorative terms both linguistically and ontologically26 can be seen when he tries to solve a pseudo-problem for his theory, i.e. the alleged problem of co-referring. On one hand, Miščević does want to avoid a qua problem by adding more layers or dimensions into the meaning of pejorative. We saw in the previous section that he introduced pluralistic commitments in order to avoid the failure of reference, yet in doing so he also intended to solve the co-referring problem, the one he takes to appear in kind terms such as “Boche” and “German”. However, he seem not to be aware that his theory cannot suffer from the problem of co-referring because his theory presupposes that the referents of these two terms are two different entities, i.e. two different social kinds. In this section I want to further question Miščević’s hypothesis that taking into consideration only the neutral causal-descriptive dimension is enough to secure the reference. I want to argue that if we allow for the aforementioned neutral causal-descriptive dimension by itself to suffice for a pejorative to refer, then we run into a further problem which I call a “delayed qua problem”27. I argue that the problem for Miščević arises when one discovers that what looks as a co-reference of two different kind terms (because of their sharing the same neutral causal-descriptive dimension and causal link to the same object, e.g. “German” and “Boche”) is not a co-reference at all but it is a reference to two different social kinds. Thus, if referencefixing is a causal and minimal-descriptive, it fails to distinguish one kind from another (e.g.

26

Linguistically here means treating them as predicates, whereas ontologically means presupposing that some pejorative terms do refer to social kinds. 27

I call it a “delayed qua problem” because by introducing dimensions of meaning Miščević tries to circumvent the qua problem introduced by Devitt (1981) by appealing to a hybrid causal-descriptive theory. For Devitt “the namer must know what kind of thing the nominatum-to-be is (…) Now, to know what kind of object one is naming is to conceptualize that object, to think of it as an object of a certain sort, as (in other words), satisfying a certain predicate. It is thus to think of it qua such-and-such.” (Reimer and Michaelson 2014) I explain the aspect of “delayed” more clearly in what follows.

kind Boche from kind German) since one should not think about two different kinds qua same kind when one does not take them as one kind in the first place. I claim that, under the assumption that reference for social kinds is possible28, conditions of reference that Miščević offers for social kinds such as German and Boche would be indistinguishable since there is obviously a full overlap between the neutral causal-descriptive dimension of a social kind German and the one of Boche. As a consequence, whenever referring to German, one would be in danger of referring to Boche as well, and the other way around. This result is unwanted since it would be impossible to distinguish reference conditions that would have to be fulfilled for the successful reference to social kinds German and Boche. After all, Miščević presupposes that “German” and “Boche” are two different social kind terms for two different social kinds in question. Miščević offered the dimensions of meaning to help us explain what kind of thing the nominatum-to-be-is, and now we see that his dimensions do not provide us with means for a proper distinction between two different social kinds when they share certain dimensions. Moreover, according to Miščević’s theory, such terms would have the same reference conditions whereas different meanings since, for instance, term “Boche” has some other dimensions that “German” does not have. Though this might not seem problematic at first glance, here are some additional reasons why this is not the case.

2.2.1. Why “Boche” and “German” are not co-refering terms and some further considerations Why the underlying idea that if one refers via using the social kind term “Boche” one is referring to the same referent as if one referred via social kind term “German” does not work for Miščević’s account? One reason is that Miščević wants to offer a theory of reference for social kinds that works for pejoratives too, and, as noted above, he claims that there are two different social kinds involved, namely Boche and German. Even though these two kinds might share some dimensions, the terms that refer to these two different social kinds, i.e. have different referents. 28

Miščević’s proposal is supposed to fix the reference for social kind terms in general because “it combines causal account with the descriptive approach, it would be a hybrid theory, all in all a hybrid social kind theory.” (Miščević 2014, 198)

Furthermore, I believe that he tries to deal with a “fake” co-reference. In the previous paragraph I stressed sharing the same dimension between two different referents, i.e. social kinds German and Boche. Since the reference is not to a person (who might only happen to be an instantiation of these two different kinds, i.e. to be both a German and Boche), but rather to two social kinds, thus, there is no co-referring in question. We have learnt from the original qua problem that a causal link cannot alone fix the reference. Here we can see a similar bad result arising if we take a neutral causal-descriptive dimension alone to provide the reference for pejorative term that refers to a social kind. According to the ingredients Miščević offers to fix the reference, it would turn out that the same ingredient that would fix the reference of “Boche” would fix the reference for “German” too. From this one might be tempted to further arrive at the identity statement such as “All Germans are Boches” and “All Boches are Germans”. Though we might agree with the later, we could not say the same for the former. One way to respond to my objection is to say that in case of “German” and “Boche” we have two terms that are different from the linguistic point, however they can still be picking out the same kind. Why is, thus, a bad result that terms “German” and “Boche” are coextensive for it seems natural that some terms, in fact, do co-refer? Moreover, if the extension of one term falls into the extension of another, the fact that extensions of “German” and “Boche” are overlapping should not be bad per se, namely sharing certain predicate properties should not be a bad result. However, I believe this is false. Terms “Boche” and “German” are not coreferring since they have different referents, i.e. they refer to two different social kinds: a social kind German and a social kind Boche. Another potential response to my objection could go as follows. Suppose a certain theory that implies that “Boche” and “German” co-refer to the same social kind, why would this still be a inconvenient result? The commonly accepted underlying assumption when treating pejoratives is that, as a matter of fact, it is true that all Boches are Germans and the theorists have been trying to explain the negative aspects of uttering this sentence. They claim that one cannot take as a datum that it is not true that Boches are not Germans. The common ground for disagreement is the following: if it were true that not all Boches are Germans, that would be a reason to say that Boche and German pick out different kinds because their mem-

bership would be different29. I again reply to this question with the assistance of Miščević’s thesis that pejoratives refer to social kinds, namely that a term “Boche” refers to one kind, and a term “German” to the another. More importantly, with the help of this thesis, Miščević’s theory deals better with the problem of co-referring and gives us the desired result: Not all Germans are Boches but all Boches are Germans. Furthermore, let us take for granted Miščević’s dimensions of pejoratives, i.e. the content of a pejorative term. For instance, Boche would be a German person who has some negative part associated with her. She is also inherently lazy and bad, and when one uses that term one might also suggest what our attitude towards her should be, such as to be avoided and discriminated. Yet, if we take the two terms to co-refer, then we need to explain how there might also exist the Germans who are not lazy and we do not associate all the dimensions we do with the Boches. On the other hand, when German and Boche taken as two different kinds, we do not have the same problem. For there would be two different terms for two different referents, namely terms “Boche” and “German” for kinds Boche and German. The idea of “Boche” and “German” referring to two different social kinds is the one I find relevant and it comes directly from Miščević’s work. I believe that it solves problems that other theories, focused on treating pejorative terms as a linguistic phenomenon solely, cannot answer. For treating “German” and “Boche” as two social kind terms that belong to the two different social kinds avoids the problem of co-referring from the beginning. And such treatment, thus, should prevent Miščević from the mistake he is doing, namely the attempt of trying to solve the problem of co-referring, since there is not problem to be solved.

2.3. Third objection: Miščević’s account does not explain what ties a kind token to a particular kind type. I wish to address briefly another problem that haunts kind theories of reference such as the Miščević’s account belongs to. Rigidity is a matter of taking an object as one’s extension. If we commit ourselves to a type/token distinction when it comes to kind terms, it seems that a referent of a certain kind is a type rather than token of a particular kind, i.e. a kind term is a 29

This is what some strongly oppose whereas Miščević’s account takes this for granted and does not offer a solution to this problem. I mentioned earlier that I agree with Miščević here even though I cannot fully defend it here. In short, reference to a social kind is not to an object or the individual as in the case of proper names or some other referring tokens, but is rather to a social kind type. Though the tokens are members of the kind, the reference is to the kind. Social kind German and Boche are two different kind types.

rigid designator for a kind type, not a kind token. Token is connected to its type in a constitutive way, i.e. token is an instantiation of a certain type. In case of causal and hybrid theories of reference there is causal relation between a linguistic token (e.g. natural or social kind terms) and a kind token, yet the causal relation is not to the referent, which is a type. Instead, a particular kind token would have to do a job of leading to its kind type, and putting a linguistic term in contact with the referent, namely a kind type. And this is a problem which all theories for kind terms would have to deal with, namely they would have to be able to explain how a rigid designator works at a level of a kind type, and not a kind token. For example, when giving a theory of reference for kind terms, one is trying to link a term “crocodile” to a kind type crocodile. However, a causal link would go from the speaker to the particular kind token of a natural kind crocodile. We need a further step to reach the referent, namely a natural kind of crocodile30. Or in case of the natural kind term “water” the referent is not a token, but a kind water.31 Now, let us look at a pejorative term “Boche” for which Miščević argues that refers to a social kind of Boche, and not to a social kind German. According to his theory, there are two different ontological entities, i.e. a social kind Boche and a social kind German. Furthermore, a causal hybrid theory of reference is supposed to do the job of securing the reference. However, it seems that, according to this causal hybrid theory, two different causal links should lead to each of the referents, in order to distinguish the two social kinds uniquely. This is at least what happens when the referent is a concrete object or an individual. However, in case of social kinds German and Boche we are presupposing two different entities, yet the only available causal link is from the speaker to the kind token of German and Boche. In this case, it also turns out that the two causal links are to the same token entity, which is not the referent, but is as close to the referent the causal link can go. Consider the following cases:

30

There is an unsolved problem that most theories of reference for kind terms share called nominalism. In other words, it seems that such theories are committed to abstract objects as their referents since kinds (types) are abstract objects. 31

Some think that natural kind terms do not refer—when you say “This is water”, the term water does not refer at all. It expresses a property or a set of something. For the purposes of this paper I focus on the theories that embrace natural kinds, e.g. Putnamian theories such as the one from Devitt and Sterenly (1999).

(1)

Hans is a German.

(2)

Hans is a Boche.

In both cases, the referent is not an individual Hans, but a social kind, namely a social kind German in (1), and a social kind Boche in (2). In other words, the causal link goes as far as to an individual Hans that in both cases happens to be a kind token. The causal link thus ends with Hans, both in the case of (1) and (2). The problem of securing the reference arises since the causal link does not reach the social kind type. Therefore, how to secure the reference from a token to its type is a common problem for most of the theories of kind terms. The causal link goes from the speaker to the token or a member of a kind only, e.g. to Hans, but this causal link alone cannot determine the intended kind type, e.g. German or Boche. On the other hand, a given kind token can be a token of many kind types, and a causal link to a kind token solely does not by itself determine a kind type. We need an explanation how a word or a term gets linked to a kind type via kind token, since often we have many candidates for kind types all of which are tokened by the same thing, e.g. a token Hans is both a member of a kind German and Boche. Miščević’s theory, too, falls short in explaining which of the many kinds types this kind token belongs to. 


Conclusion Is it possible to give a theory of direct reference for pejorative terms? In his paper, Miščević (2014) utilizes Devitt and Sterenly’s causal hybrid theory of reference in order to give an account of pejoratives treating them as hybrid social kind terms. In other words, Miščević takes the referents of pejorative terms to be social kinds. While criticizing Miščević’s account, my attempt was to show that the most relevant part of Miščević’s paper is his treatment of certain groups of pejorative terms as social kind terms that refer to a certain social kind. I am convinced that treating certain group of pejoratives as social kind terms with their referents being social kinds is the most valid premise Miščević introduces, even though he does not stress this idea enough neither does he utilize it properly. In the first part of this paper, I have given a summary of Miščević’s (2014) theory of pejorative terms as social kind concepts and criticized his account of reference for pejorative terms. In the second part, I have identified three objections for his theory. I have argued that introducing pluralistic commitments about propositions is not helpful in any way for his account. I have further argued that it, on the contrary, brings about the failure of reference and falsity of meaning. Furthermore, I have pointed out that Miščević is mistaken in trying to avoid the problem of co-reference between terms such as “Boche” and “German” since there is no co-reference because their referents are two different social kinds that his theory presupposes from the start. Finally, I have raised a more general objection connected to the reference of social kind terms and pejorative terms as well, namely the need for a further explanation of what ties social kind tokens to a particular social kind type.


References Bach, K. 1987/94. Thought and Reference. Oxford: Oxford University Press. Burge, T. 1977. “De Re Belief”. Journal of Philosophy 74: 338–62. Boer, S. and W. Lycan. 1986. Knowing Who. Cambridge, MA: MIT Press. Brewer, B. 1999. Perception and Reason. Oxford: Calderon Press. Devitt, M. 1989. “Against Direct Reference”. Midwest Studies in Philosophy 14(1): 206-240. Devitt, M. 1999. “A Naturalistic Defense of Realism”. In Metaphysics: Contemporary Readings, ed. S. D. Hales, 90-103. Belmont, CA: Wadsworth. Devitt, M. and Sterenly, K. 1999. Language and Reality. Oxford: Blackwell. Donnellan, K. 1979. “The Contingent A Priori and Rigid Designators”. In Contemporary Perspectives in the Philosophy of Language, ed. P. French, T. Uehling, and H. Wettstein, 12-27. Minneapolis: University of Minnesota Press. Dummett, M. 1973. Frege: Philosophy of Language. London: Duckworth. Evans, G. 1982. The Varieties of Reference. Oxford: Oxford University Press. Frege, G. 1979. “Logic”. In Posthumous Writings, ed. H. Hermes, F. Kambartel and F. Kaulbach, 126-151. F. Oxford: Basil Blackwell. Fuš, M. 2013. “Bach’s Constraint on Extending Acquaintance: Some Questions and a Modest Proposal”. Croatian Journal of Philosophy 13 (2): 201-216. Gärdenfors, P. 2000. Conceptual Spaces. The Geometry of Thought. Cambridge: The MIT Press. Gärdenfors, P. 2014. The Geometry of Meaning. Semantics Based on Conceptual Spaces. Cambridge: The MIT Press. Harman, G. 1977. “How to Use Propositions”. American Philosophical Quarterly 14: 173-176. Hom, Ch. 2010. “Pejoratives”. Philosophy Compass 5 (2): 164–185.

Hom, Ch. 2008. “The Semantics of Racial Epithets”. The Journal of Philosophy 105: 416-40. Jeshion, R. 2010. “Singular Thought: Acquaintance, Semantic Instrumentalism, and Cognitivsm”. In New Essays on Singular Thought, ed. R. Jeshion, 105-140. Oxford: Oxford University Press. Jeshion, R. 2009. “The Significance of Names”. Mind and Language 24 (4): 370–403. Jeshion, R. 2002. “Acquaintanceless De Re Belief”. In Meaning and Truth: Investigations in Philosophical Semantics, ed. J. K. Campbell, M. O’Rourke, and D. Shier, 53-78. New York: Seven Bridges Press. Kaplan, D. 1989. “Demonstratives”. In Themes from Kaplan, ed. J. Almong, H. Wettstein, and J. Perry, 481–563. New York: Oxford University Press. Lewis, D. 1979. “Attitudes De Dicto and De Se”. Philosophical Review 88: 513–543. Miščević, N. 2012a. “Offensive Communication. The case of pejoratives”. Balkan Journal of Philosophy 4 (1): 45–60. Miščević, N. 2012b. “Pejoratives And Conceptual Truth”. In Inference, Consequence, and Meaning: Perspectives on Inferentialism, ed. L. Gurova, 47–67. Cambridge: Cambridge Scholars Publishing. Miščević, N. 2011. “Slurs & Thick Concepts. Is the New Expressivism Tenable?” Croatian Journal of Philosophy 11 (32): 159–182. Miščević, N. 2014. “The Geometry of Offense – pejoratives and Conceptual Spaces”. Croatian Journal of Philosophy 14 (41): 195-214. Potts, Ch. 2005. The Logic of Conversational Implicatures. Oxford: Oxford University Press. Pryor, J. 2007. An Epistemic Theory of Acquaintance. Unpublished manuscript. Recanati, F. 1993. Direct reference: From Language to Thought. Oxford: Blackwell.

Reimer, M. and E. Michaelson, 2014. “Reference”. The Stanford Encyclopedia of Philosophy (Winter 2014 Edition), Edward N. Zalta (ed.), URL = . Salmon, N. 1988. “How to Measure the Standard Metre”. Proceedings of the Aristotelian Society 88: 193–217. Soames, S. 2003. Philosophical Analysis in the Twentieth Century, Vol. 2. Princeton: Princeton University Press. Williamson, T. 2009. “Reference, Inference, and the Semantics of Pejoratives”. In The Philosophy of David Kaplan, ed. J Almog. and P. Leonardi, Ch. 9. Oxford: Oxford University Press.

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.