\documentclass[11pt,a4paper]{article} \usepackage[pdftex]{graphicx} \usepackage[english]{babel} \usepackage{mdwlist} \usepackage{microtype} \usepackage{hyperref} \usepackage{epigraph} \begin{document} \begin{center} \section*{No such thing as a language?} {\em Andreas van Cranenburgh 0440949 \\ Radical interpretation, hermeneutics and forms of life \\ \today} \end{center} %\begin{quote} \epigraph{ Without language we would have no reason, without reason no religion, and without these three essential aspects of our nature, neither mind [Geist] nor bond of society}{Johann Georg Hamann, S\"amtliche Werken III, p.~231 %S\"amtliche Werken, ed. Josef Nadler (1949-1957), vol. III, p. 231 } %(N III, 231, 10-12). %\end{quote} \abstract{Davidson's later work presents the startling conclusion that a shared language in the form of conventions is not necessary to explain verbal communication. This paper argues that this conclusion can and should be avoided, through a variety of amendments. These allow for recognizing a linguistic division of labor, a role for conventions, and real normativity of meaning. Furthermore, a comparison of Davidson's notion of triangulation with the notion of joint attention suggests that the latter might form an important bridge from precognitive to linguistic cognition. Lastly some reflections on the notion of language suggest that some notion of a public language is necessary to understand verbal communication, firstly because it provides the background for understanding deviations from the norm (such as Davidson's malapropisms), and secondly because it is necessary to make sense of why language evolution is convergent.} \subsection*{Introduction} Davidson's `Nice Derangement of Epitaphs' paper ends with an infamous passage: \begin{quote} I conclude that there is no such thing as a language, not if a language is anything like what many philosophers and linguists have supposed. There is therefore no such thing to be learned, mastered, or born with. We must give up the idea of a clearly defined shared structure which language-users acquire and then apply to cases. And we should try again to say how convention in any important sense is involved in language; or, as I think, we should give up the attempt to illuminate how we communicate by appeal to conventions. --- Davidson (1986) \end{quote} It is easy to balk at such a provocative statement. If there were no such thing as a language, why would there be linguistic discrimination and even persecution, as Dummett (1986) remarks? On the face of it this seems grounds for an unconditional dismissal; all the more because Davidson has fulfilled Lewis's (1975) prophecy that only a philosopher could deny the role of convention in communication.\footnote{``It is a platitude --- something only a philosopher would dream of denying --- that there are conventions of language.''} However, there clearly is a qualification to this claim that demands attention. Davidson refers to a specific conception of language that ``philosophers and linguists have supposed.'' So what is this conception? In Davidson (1994) he explicitly defines it as: \begin{itemize*} \item a language is the ability to operate in accord with a precise, specifiable set of syntactic and semantic rules \item verbal communication depends on speaker and hearer sharing this ability, and depends on nothing else \end{itemize*} % is there a concept of language, ordinary or philosophical, that is basic to communication? % are meanings shared? % Wittgenstein: what distinguishes following a rule from merely thinking % one is following a rule? the answer is: a social practice % are there meanings beyond the beliefs of speaker & hearer? In Davidson's framework the individual has priority over the community. Specifically, individual interpretation has priority over public interpretation. This is associated with the role of conventions in interpretation. In day to day life conventions do play a large role in communication, but Davidson believes they are not strictly necessary. The concept of a language has no role to play for philosophy, because everyone speaks a different idiolect, and even those are constantly in flux. Typically, conventions are characterized as beliefs of those who adhere to them (Lewis 1975). In communication, the meaning of sentences would appeal to these beliefs which constitute the relevant conventions. However, for Davidson the interdependence of meaning and belief precludes this appeal as an explanation on pain of circularity --- conforming to conventions could not be established without also interpreting the language use in question. % why is that problematic? So how could people communicate without conventions? The issue is whether convention is what makes communication possible, or if conventions are merely habits formed through successful communication. As would be expected, Davidson argues the latter. Conventions are crutches, but not strictly necessary. Radical interpretation provides the answer to communication without convention. Imagine two strangers marooned in an isolated location. Their survival most likely hinges on co-operation, which requires communication to establish. Even if they share no common language, there is no principled reason why they should not reach a stage where they can make themselves understood. According to Davidson this should happen through radical interpretation: a process of systematically connecting beliefs about true statements with sentences. In this sense it can be claimed that a language is not necessary for communication, if a language is taken to be a conventional assignment of interpretations to sentences. \subsection*{Defusing individualism} A major feature of Davidson's philosophy is his commitment to \emph{methodological individualism}. This entails the requirement that criteria for the correct application of terms should, ultimately, be grounded in facts of an individual itself. Because of this, accounts of meaning based on interpretation such as Davidson's are perceived to be incompatible with theories that allow for social usage to determine meaning. Jackman (1998) introduces another variety of individualism to contrast this with, \emph{ascriptional individualism}. This is the stronger thesis that what one means is \emph{independent} of the usage of others. Burge (1979) introduced the famous example of arthritis. In one situation Bert is confused and believes arthritis also refers to inflammations outside of joints, but is corrected by his doctor. In a counterfactual world, Bert grows up in the same environment except that in this world arthritis does refer to these inflammations as well. A natural intuition here is that in the second situation, Bert correctly uses the word, because it accords with the usage in the community, while in the former he is wrong. Davidson (1994) rejects this claim. %why It is clear that an ascriptional individualist must deny the claim, for there is no difference in Bert's beliefs which can explain that he means two different things with arthritis. One possibility would be that Bert was referring to the actual concept of arthritis in both worlds; however this is not in accord with his beliefs so it is hard to argue. The position taken by Davidson is that he was referring to the extended arthritis in both worlds. But notice that this takes away the normativity, although the doctor ``corrected'' Bert, the latter actually referred correctly to something else in this reading. This contradicts the natural intuition that the doctor is an authority and that consequently his beliefs should determine the relevant meanings. But for a methodological individualist, Jackman argues, matters need not be so one-sided. External factors can influence meaning, so long as the individual in question accepts these factors. Bert's intended meaning for arthritis could be ``whatever experts take arthritis to be.'' In this sense, meaning can be thoroughly determined through social practices, but the individual has the final word on it. What methodological individualism does contradict is the claim that meaning can be determined by something regardless of the speaker's attitude. If a speaker does not intend to speak as others do, then we should not interpret him that way. In practice, of course, almost everyone does have such intentions. We can thus conclude that at least occasional ascriptional non-indi\-vi\-dualism is compatible with methodological individualism. Davidson~(1994) acknowledges as much, mentioning the reliance on experts in identifying elms and beeches. However, he argues that it cannot be the case that everyone means with ``elm'' what everyone else means by it, because that would simply be circular, and fail to provide a reference. This leads him to conclude that the linguistic division of labor cannot be essential to language; the problem is not whether others influence our meanings, but the claim that we depend on such influence necessarily. On the other hand Davidson (1994) also dismisses ascriptional non-individualism summarily: \begin{quote} I am not impressed by [Dummett's] or Burge's or Putnam's insistence that words may have a meaning of which both speaker and hearer are ignorant. I don't doubt that we sometimes say this, and it's fairly clear what we have in mind: speaker and hearer are ignorant of what would be found in some dictionary, or how people with a better or different education or a higher income use the words. This ... imports into the theory of meaning an elitist norm by implying that people not in the right social swim don’t really know what they mean. \end{quote} I think Davidson equivocates prestige and authority here. The former indeed seems incidental and perhaps even orthogonal to meaning, but the latter pertains to the correctness of the application of terms, at least to the degree that individuals recognize and rely on it. What is important is that such deference is extremely pervasive, not whether it is essential to language or communication as such. It is thus not the correctness of Burge's claims, but their generalizability. To sum up, there is no inherent problem in recognizing the influence of social usage on individuals in an interpretational account. This is achieved by limiting individualism to methodological individualism, and limiting the influence of the social to the extent that people actually intend to rely on it. %This probably rules out the constitutive role of social usage for things like function words and basic vocabulary. %\subsection*{How linguistic are beliefs?} %cite Carruthers (2001), Kovacs (2008), Tomasello % => beliefs possible before language, without meaning %As said, Davidson holds that belief and meaning are intimately related. %More generally this derives from Davidson's (1974) position that neither %language nor thought has conceptual priority, and that the one cannot be fully %explained in terms of the other. %In Davidson (1975) the argument is put forward firstly that beliefs and %desires are content-bearing and that they must be expressed in a %\emph{that}-clause. \subsection*{Conventions} %Effective communication requires conventions. Only on pain of rediscovering %or re-establishing them can communication get off the ground. The idea that convention underlies meaning has a long history tracing back as far as Plato's Cratylus: \begin{quote} Hermogenes: [N]o one is able to persuade me that the correctness of names is determined by anything besides convention . . . No name belongs to a particular thing by nature, but only because of the rules and usages of those who establish the usage and call it by that name (Plato 1997, p.~384c--d) \end{quote} But this provides only a relatively weak sense of conventional meaning --- distinguishing it from intrinsic meaning such as onomatopoeia or god given meanings. The more interesting questions are whether conventions can determine meaning, and secondly whether they are necessary. What Davidson (1986) purports to show is that malapropisms are incompatible with the idea that communication is governed by regularities and conventions which are learned in advance. A malapropism is by definition not covered by the regularities that make up prior learning; they are only understood when the interpretation of the hearer coincides with intentions of the speaker. Davidson frames this as arriving at a shared but momentary `passing theory.' The critique in Davidson (1984) is aimed specifically at Lewis' definition of convention: that it consists of a regularity in action and belief which perpetuates itself because it serves a common interest (Rysiew 2000). Davidson's conclusion is that there is no interpretation of this definition which could serve as a basis for linguistic meaning. Rysiew remarks the importance of the type of regularity that is considered: \begin{itemize*} \item `an intentional conformity to regularity' (Ramberg 1989, p.~100, as quoted in Rysiew 2000) \item `a regularity in intention to conform to a regularity' (Rysiew 2000) \end{itemize*} It seems apt to call this last version a second-order regularity. It is clear that malapropisms and figurative meanings readily deviate from the former regularities. However, the second version only specifies a regularity in intending to conform to regularities, without specifying that this intending necessarily leads to conforming to the latter regularities. % argh fixme In fact, this regularity of intending to speak as others do is ubiquitous, because speech is typically employed to achieve certain ends. Davidson's mistake is to claim that a convention necessarily consists of interpretations of sound patterns, i.e., of the actual use of language. Rather it is the ``intention to interpret `sound patterns' in the same way'' (Rysiew 2000) which forms the conventions underlying language use. % sum up We seem to have defused Davidson's reservations for conventions to determine meaning, but are conventions necessary? To see that this should be the case, imagine how much malapropisms or other deviations can be accommodated before communication breaks down. Correctly interpreting the deviant usage proceeds through hints gleaned from the situation and the rest of the utterance insofar as it follows standard usage; if that does not supply a substantial context, then there will not be enough to go by. The result is that, because of limits on our sense-making abilities, we are constrained to using most of our words following conventions (if we want to be understood, that is; Dada poetry is another story). Ultimately this does not prove that conventions are necessary in the philosophical sense, but this seems to be rather a scholastic point. Davidson fails to supply or even sketch an example of communication without convention, but just asserts it must be so: \begin{quote} Knowledge of the conventions of language is thus a practical crutch to interpretation, a crutch we cannot in practice afford to do without --- but a crutch which, under optimum conditions for communication, we can in the end throw away, and could in theory have done without from the start. (Davidson 1984) \end{quote} The only ``optimum conditions'' I can imagine are omniscience or telepathy, but those make the whole idea of language irrelevant.\footnote{Although I imagine even telepathy would require that our thoughts follow conventions --- but this is too speculative a matter to even consider.} Either there is something non-obvious that you want to communicate --- which will involve conventions, or what is communicated is simple enough such that immediate radical interpretation is possible (e.g., pointing to whatever passes by and making up a new word). What Davidson presumably has in mind is that we might continually radically interpret each other. This fails to convince me, as I see the point of radical interpretation as the acquisition of a set of conventions, \emph{after which} communication may commence. More importantly, Davidson has a bias for interpretation over speech production. Radical interpretation can help an individual who does not yet possess the requisite conventions acquire them from scratch; however, it does not help in picking out and ordering words to form an utterance --- that, I submit, surely requires conventions. \subsection*{Normativity} It is often stated that meaning must somehow be normative. If it cannot be distinguished whether the use of words is right or wrong, then there is no way to tell believing one is in accord with a rule from actually being in accord with a rule. This problem manifests itself in Davidson's theory through the focus on successful communication, which is defined as getting the speaker's intention across. To see that this is inadequate, consider a request to open a window, which is summarily fulfilled. It appears that the communication must have been successful, but the hearer could also have understood ``open the door,'' while opening the window out of spite (Penco 2007). However, it is unclear where this normativity should come from. It cannot come from language use (at least not directly), because then whatever usage one employs would immediately validate itself --- in other words, facts of usage are merely descriptive. Another option would be that normativity somehow derives from some platonic realm; although this is possible, it is unfortunate that on such an account a speech community is not the creative force behind meanings. Davidson (1993) flatly rejects the normativity of meaning. He argues that there is no distinction to be made between norms of meaning and norms of reasoning; thus when someone uses a word in an unusual way it can be attributed to a misunderstanding of its meaning, but also of its application. He concludes (emphasis in the original): \begin{quote} though our norms guide our judgments of the reasoning ability of a speaker and these judgments in turn affect our understanding of what the speaker means, none of these normative considerations should tempt us to say that the speaker has failed to follow the norms of \emph{language}, for as Bilgrami insists, there are no such norms. \end{quote} Although others have since supported the point of view that there is no normativity in the Davidsonian program (Gl\"uer 2001; Schroeder 2002; Engel 2008), I believe this to be an overly pessimistic position. I take it for granted that there should be normativity (although some dispute this as well); this intuition is supported among others by McDowell~(1984), Gampel~(1997) and Wedgwood (2007), who argue that normativity follows from the principle of charity. I will presently discuss one such account. Following Jackman (2004), normativity in an account of meaning can be defined as the result of the following conditions: \begin{enumerate*} \item Non-Reducibility: No fact about what someone means is reducible to some purely descriptive property of that person. \item Bindingness: The norms must be in some sufficiently robust sense binding on us. Anyone who means anything by their words will not be able to `opt out' of such norms. \item Autonomy: The norms in questions should be genuinely semantic norms. That is, the norms should be meaning-determining, and not (like some prudential and epistemic norms) simply applicable to items that are independently meaningful. \end{enumerate*} These are strong conditions. For example, they disqualify the supposed version of normativity which prescribes that words should be used so as to make sentences true, because this fails to fulfill the third condition. Such an account boils down to making words refer correctly, which arguably can be accounted for in a purely descriptive theory. Worse yet, someone who decides not to call a spade a spade could violate the second condition as well, while still conveying a message (albeit most likely an ineffectual one). It appears that generally, use-based theories of meaning suffer from a lack of normativity, because usage can be noted descriptively without norms entering into the picture. But although Davidson himself was opposed to the idea of normativity of meaning, it turns out that the principle of charity can be employed to account for it. Most other accounts of normativity in meaning stress the social aspect of norms. Judgments then derive from the contrast between the usage of the community and that of the speaker. Davidson is skeptical of this because these social norms then seem independent of the intentions of speakers. However, Jackman (2004) presents a take on normativity in terms of charity which is compatible with Davidson's methodological individualism. For this Jackman uses a version of charity which is slightly generalized: \begin{quote} The semantic values of the words in a speaker's language are the values in the set that maximizes the truth of the speaker's commitments. \end{quote} The modifications are `semantic values' instead of `objects,' `commitments' instead of beliefs (to cover presuppositions and assumptions as well), and lastly the maximization is intended to allow for commitments with different weights. Now let us see how this version of the principle satisfies the conditions for normativity. The first condition is non-reducibility. This seems to be satisfied in virtue of the holism of belief. Meaning cannot be reduced to descriptive facts because meaning depends on beliefs which are revised holistically, through a process that does not follow descriptively from the facts at hand. Beliefs are revised according to what is rational to believe, and it appears highly unlikely that a non-normative theory could account for rationality. The result is that meaning supervenes on usage, but cannot be reduced to usage. The second condition prescribes that semantic norms must be binding: it must be impossible to ignore all norms and still mean something. Jackman (2004) argues that this implies the norms should have an internal relation to the language user. Norms with an external relation would allow one to deviate from them while still being meaningful. In contrast, if charity is followed and a statement has been accepted as true, one is committed to keep that statement consistent with the other commitments in the belief set (by dropping one or more commitments, or changing the meaning of terms). Failing to do so implies giving up on making sentences true, and consequently on assertion, judgment and meaningful discourse. Ergo, a speaker is bound by norms with the principle of charity, where the norms consist in maintaining a consistent belief set. The third condition, autonomy, seems the weakest link. This is because of the interdependence of meaning and belief in a Davidsonian framework. When someone misidentifies a wolf for a dog, it is more likely to be an epistemic than a semantic norm which is violated. But this should not have to pose a problem, because in the larger picture of the whole web of beliefs, the norms are more readily recognized as semantic. These norms are meaning-determining, because they do not apply to things that already have meaning, but are constitutive of it. % to sum up? %\subsection*{The reality of literal meanings} %Davidson (1986) %first meanings \subsection*{Triangulation} Aside from charity, there is another major principle in Davidson's framework. This is the idea of triangulation (e.g., Davidson 2001). Triangulation involves two persons interacting with each other and with an object to which they react in a similar fashion. With the reactions as two sides of the triangle, the third side represents the mutual recognition of both persons of the stimulus. This process then allows to establish a form of objectivity derived from intersubjectivity, which would otherwise be impossible. This objectivity amounts to a sense of right and wrong derived from the fulfillment or violation of expectations. But two kinds of triangulation can be discerned. The first is this basic, precognitive kind which serves to establish a shared world and objective thought. It is basically a form of conditioning of correlated stimuli. The second is a special kind of triangulation, which is reflective and depends on the ``exploitation of reflexive communicative intentions'' (Eilan 2005) --- i.e., language. How, though, should one get from precognitive triangulation to reflective, communicative triangulation? The former is purely causal, the latter intentional. Davidson does not have a story about this fundamental transition. It is as if language is conceded to appear by magic. His claim that thought is essential social in nature follows from the sense of right and wrong established by the precognitive triangulation; however, Eilan (2005) argues that it is unclear how this sense could survive into the reflective triangulation. Moreover, incontrovertible evidence about 1 and 2 year old's language development presents a problem for Davidson's insistence that belief and meaning are intertwined: infants of this age undoubtedly have some degree of success at communication, yet they fail the false belief task (they only attribute what they themselves believe to be true to others). Apparently there can be communication without higher-order intentions and the concept of belief. We can remark, with Brinck (2004), that we are in a dilemma: \begin{quote} Either triangulation is pre-cognitive and cannot explain how speakers converge on a common cause, or it requires higher-order thought, and cannot provide for language entry, nor individuate the content of individual beliefs and utterances. \end{quote} In other words, a language learner must break into an ``intentional circle'' (Brinck 2004), but triangulation does not offer a way in. For an alternative, it is instructive to look at developmental psychology. Triangulation is strikingly similar to the concept of joint attention in the account of Tomasello et al.\ (2005) of shared intentionality, with the difference that shared intentionality is a much broader concept and consequently deals with more than just objectivity. Specifically, it includes the ability of recognizing others as persons, and ultimately to collaborate in shared goals. %What makes humans unique among animals? Common answers include language, %reasoning or the pursuit of knowledge and art. But upon reflection these %phenomena are only displayed after humans acquire these skills from their %caretakers and other conspecifics. The more fundamental ability underlying %these phenomena is shared intentionality, according to Tomasello (2005). %Shared intentionality makes it possible to participate in collaborative %activities with shared goals and intentions. First infants start to follow gaze directions and recognize animate actions (6 months). Later infants recognize the goals that underlie these actions (9 months). Lastly they come to understand intentional action and selectively attend to aspects relevant for the goal in a situation (14 months). These three steps allow for imitative learning, which is a crucial mechanism for cultural transmission. As long as previous achievements are passed on reliably to new generations, culture grows with a ``ratchet effect,'' i.e., knowledge is monotonically increasing. What is important to realize is that this implies that most individuals make use of tools (including language) which they could not have invented themselves, and those that could have get the chance to invent new tools instead of re-inventing things. Shared intentionality arises when these skills are applied to solving problems together or in groups. In dyadic interaction infants notice each other’s emotions and behavior, which allows for turn taking. In triadic interaction two or more individuals work towards a common goal and recognize each other’s behavior as directed towards this goal. The final step is collaborative engagement, in which actions are coordinated toward a shared goal using joint attention. To reach this stage agents must represent the situation from a neutral point of view, in order to allow for role reversal or helping out the other. This results in the development of dialogic cognitive representations, which not only include the shared goal but also the roles of the participants. Aside from these capabilities, these processes also rely on a motivation to share one’s psychological state with others --- this explains why children make disinterested comments such as ``doggie gone.'' --- this is in contrast with animal communication, even chimps who have been taught some sign language mostly utter imperatives, e.g., to procure food. % summarize Brinck 2004 here. To sum up, it seems clear that compared to shared intentionality, the stimulus-response nature of triangulation is only a very basic form of social interaction. From developmental evidence it is clear that before children acquire language, they develop skills of intentionality; the desire to share mental states appears to be an innate driving force behind language acquisition and co-operation. What this entails philosophically remains to be worked out, but that an understanding of the mental states of others must precede linguistic understanding seems clear. The recognition of stimuli in triangulation is not enough to get to language, the recognition of the other \emph{qua} person is an important threshold. Joint attention could serve as a bridge towards reflective triangulation; this bridge provides a middle ground between the purely causal and correlative character of precognitive triangulation and the full blown propositional nature of reflective triangulation (Brinck 2004) % fixme \subsection*{What of the concept of language?} Davidson's (1986) position is that there is no interesting concept of language. Speakers can at best be said to speak compatible idiolects, and at worst even those disintegrate into continually changing passing theories. A language would then be a set of more or less overlapping prior or passing theories. It is clear that such a concept is not basic or fundamental to communication, but a mere afterthought. This leads Davidson (1994, p.~100) to state that: \begin{quote} neither the usual concept [of language] nor the philosophical concept is very important in understanding what is essential to verbal communication. \end{quote} I think that Davidson is overstating his case here. From the perspective of a single instance of communication, indeed, the concept of a language does not add much to our understanding. But this instance of communication is situated at a point of massive agreement on semantic and non-semantic information, and this is what affords the possibility to deviate freely. In parallel to Davidson's claim that irrationality is only recognized relative to massive rational agreement, deviance from common language should also be defined as such and interpreted relative to or in contrast to more common usage (or perhaps `arbitrary agreement'). This requires the individual to consider their idiolect as a more or less imperfect reflection of a shared language (we might even call this the `public theory' following the prior and passing theories); in turn prior and passing theories are projections from this idiolect. Although mrs.\ Malaprop's prior theory is indeed different from ours, it is crucial to observe that it is not \emph{that} different; consequently we share her prior theory to a great extent. Furthermore, from the perspective of language origins and acquisition, it is even more problematic to deny a notion of a shared and stable language. The only way that communication will work is if individuals imitate each other to such a degree that language use converges sufficiently. If such a commitment is not there, language will not function or even get off the ground, and consequently success at communication will be random instead of reliable. Speaking similarly is not just an ``enormous convenience'' (Davidson 1986, p.~256) but a \emph{sine qua non} of sufficiently complex communication (i.e., the kind which would typically not succeed without language). In any given situation the commitment is defeasible, of course, so Davidson's malapropisms can be interpreted, so long as they are infrequent enough not to topple the critical mass of agreement among language users (conversely, if they are frequent enough they will simply be assimilated as recognized meanings). Without sufficient stability there is, game-theoretically, no incentive to continue using language --- interpretation will be too costly for the hearer, and the speaker risks misinterpretation. Davidson's insistence to consider just two speakers at a time (Hacking 1986) prevents these matters from being recognized. Although Davidson does speak of idiolects converging, he has no way of explaining \emph{why} they do so, except in the local case of a speaker and hearer successfully communicating. The matter is even worse concerning \emph{how} convergence occurs, because Davidson frankly admits this to be a mystery.\footnote{Davidson (1986) talks of a ``mysterious process by which a speaker or hearer uses what he knows in advance plus present data to produce a passing theory''} My contention is that local convergence of speaker and hearer depends on global convergence in the speech community as a precondition. Another argument that Davidson employs to deny the concept of a language is to claim that there is no essential difference between language abilities and other skills. While it is plausible that mostly domain-general cognitive capabilities are recruited by language, on the other hand it must be admitted that the nature of the skill is exceedingly unique. Language is all-pervading, a medium for cultural transmission, a means of self-expression, \&c. Therefore, to understand what it means to be human, language is the most indispensable tool. Davidson limits his theory to \emph{communication by language}, but this is too narrow a focus to understand how language is implicated in the bigger picture. Finally, there is Davidson's claim of the priority of idiolects over public language. Although it is necessarily true that individuals can only invoke their idiolect in interpretation, specifying what a particular individual's idiolect comprises is practically impossible. Save for some amazing future brain-reading device which would be able to extract and exhaustively describe idiolects, the only sense we can make of someone's idiolect is as an instantiation of a readily recognized ethnic language. Thus, from a pragmatic point of view we are much better off recognizing common ethnic languages as basic, with the proviso that this is an idealization. While technically this concept of language will be a fiction, it is to be preferred over insisting on an ultimately unhelpful nominalism. Davidson (1986) anticipates alternative concepts of language and posits: \begin{quote} none of them satisfies the demand for a description of an ability that speaker and hearer share and that is adequate to interpretation. \end{quote} However, I think it is a sleight of sophistry to insist on both ``share'' and ``adequate'' at the same time. The former insists that any speaker has the particular ability uniformly, the latter requires that it exhaustively describes everything going into interpretation. Both of these demands are too strong, and conjunctively this makes for an easy case of linguistic denialism. If we loosen these criteria, we can re-interpret ``share'' as sharing to a sufficient degree; it is undeniable that the idiolects of speakers from the same speech community have a strong family resemblance, and that this is no accident. Secondly, ``adequate'' is an unfairly stringent demand, because the fact that non-linguistic information is necessary for complete interpretation does not detract from the idea that specifically linguistic abilities are brought to the table as well. %http://plato.stanford.edu/entries/meaning-normativity/ %http://plato.stanford.edu/entries/convention/#ConLan %http://plato.stanford.edu/entries/meaning/ % ... 20 pages later: \subsection*{Conclusion} To sum up, we seem to have defused much of the more controversial aspects of Davidson's conclusions. Firstly, it has been shown that the Davidson's commitment to individualism is actually compatible with the linguistic division of labor. This allows for a clearly delineated amount of social influence on meaning. This relies on taking the middle ground of an individual who is able to choose the amount of outside influence that is relied on. Secondly, the role of conventions has been re-instated by re-defining them such that it is not the usage of words that constitutes conventions, but rather a collective intention to use words in the same way. This makes for defeasible conventions, although it was argued that getting away with deviant usage is still parasitic on agreement on other conventions. Lastly, the lack of normativity in Davidson's theory has been shown to be problematic, because intentions and apparent success are not sufficient to establish actual success at conveying a message. This can be fixed by a modified version of the principle of charity, from which normativity of meaning appears to follow. Aside from these more or less straightforward modifications of the Davidsonian program, some further problems have been put forward. Triangulation is either correlative and thus lacking in intentionality, or it is reflective and linguistic and thus cannot explain the acquisition of language. A possible solution lies in the concept of joint attention as a limited form of intentionality without propositionality. It has been argued that some idea of a public language is necessary to understand communication after all, particularly if the perspective of a speaker and hearer is broadened to that of a speech community, including the phenomena of language acquisition and evolution. %It must be concluded that Davidson's claims are limited to shared conventions %not being \emph{sufficient} to explain communication, which is not that %surprising, while the claim that conventions are not \emph{necessary} is %dubious. %We can conclude that there is such a thing as language, % priority of individual versus public interpretation % Davidson argues that individual has priority over public % radical interpretation is not actually radical; language could only have % originated if people who did not already share a language could come % to share a language; however, an encounter with an alien life form or % communicating intimately with another species would be radical -- % but this is likely not possible, at least not with our ethnic languages. % communication is grounded in sharing certain basic interests such as % hunger and desire for companionship et cetera; this already prevents % interpretation from being truly radical, if it is interpretation at all % since a non-linguistic understanding precedes attempts at interpretation, % and linguistic understanding can develop from there. % => does davidson trivialize the notion of language? rorty & haack say so % but: seems to be consistent and coherent with his views of radical interpretation % malapropisms and mistakes can only be accomodated if they don't stray to far % from expectations given by shared knowledge of language. % the only way a speaker can predict that an utterance will be correctly % understood is when the hearer is assumed to be part of the same speech % community. %two individuals radically interpreting one another and coming up with %the language spoken in the rest of the community would be as fortuitous %as accidentally reinventing an equivalent language. \vspace{3em} \begin{center} $ \infty $ \end{center} \vspace{3em} %\newpage \subsection*{References} \begin{description*} \item[] Brinck, Ingar (2004). Triangulation and radical interpretation, \emph{Dialectica}, 58(2). \item[] Burge, Tyler (1979), ``Individualism and the Mental,'' in French, Uehling \& Wettstein. Midwest Studies in Philosophy IV: Studies in Metaphysics, Minneapolis: University of Minnesota Press. \item[] Davidson, Donald (1984). ``Communication and convention,'' in Inquiries into truth and interpretation, Oxford University press, 2nd edn., 2001. \item[] Davidson, Donald (1986). ``A nice derangement of epitaphs,'' reprinted in The essential Davidson, Oxford University Press, 2006. \item[] Davidson, Donald (1994). ``The social aspect of language,'' reprinted in Truth, Language and History. Clarendon Press, Oxford, 2005. \item[] Davidson, Donald (2001). ``The second person.'' In: Subjective, Intersubjective, Objective, pp.~107--122. Oxford University Press, Oxford. \item[] Dummett, Michael (1986). `A nice derangement of epitaphs': some comments on Davidson and Hacking. In E. Lepore, ed., Truth \& Interpretation: Perspectives on the Philosophy of Donald Davidson. Oxford: Blackwell. \item[] Hacking, Ian (1986). The parody of conversation. In E. Lepore, ed., Truth \& Interpretation: Perspectives on the Philosophy of Donald Davidson. Oxford: Blackwell. \item[] Eilan, Naomi (2005). Joint attention, communication, and mind. In N. Eilan, C. Hoerl, T. McCormack \& J. Roessler (Eds.), Joint attention: communication and other minds (pp.~1--33). Oxford Clarendon. \item[] Engel, P., 2008. “Belief and Normativity,” Disputatio, 2: 179--202. \item[] Gampel, E.H., 1997. “The Normativity of Meaning,” Philosophical Studies, 86: 221--242. \item[] Gl\"uer, K., 2001. “Dreams and Nightmares. Conventions, Norms, and Meaning in Davidson's Philosophy of Language,” in Interpreting Davidson, P. Kotatko et. al. (eds.), Stanford: CSLI Publications, 53--74. \item[] Jackman, Henry (1998), Individualism and interpretation, Southwest Philosophy Review, V.14, n. 1, Jan 1998. \item[] Jackman, Henry (2004). Charity and the normativity of meaning. Talk presented at the APA Pacific division meeting, March 2004, \\ \url{http://www.yorku.ca/hjackman/papers/normativity.pdf} %\item[] Johansson, Sverker (2004). The individual and the species in the cultural evolution of language. \url{http://hem.hj.se/~lsj/EELC04.pdf} %\item[] Lewis, David (1969). Convention: a philosophical study. Cambridge, Mass.: Harvard University Press. \item[] Lewis, David (1975). ``Languages and Language,'' reprinted in Philosophical Papers. Vol.I. Oxford: Oxford University Press, 1983. \item[] McDowell, J., 1984. ``Wittgenstein on Following a Rule,'' Synthese, 58: 325–363. \item[] Penco, Carlo (2007). Idiolect and context. In R.E.Auxier and L.E. Hahn (eds.), The Philosophy of Michael Dummett -- Library of Living Philosophers, vol. XXXI, (567--590). \item[] Plato (1997). ``Cratylus.'' Trans. C. D. C. Reeve. In Complete Works. Ed. John Cooper. Indianapolis: Hackett \item[] Ramberg, B. (1992). Donald Davidson's philosophy of language: an introduction. Oxford Basil Blackwell. \item[] Schroeder, T., 2003. ``Davidson's Theory of Mind is Non-Normative,'' Philosophers' Imprint, 3: 1--14. \item[] Tomasello, Michael, Malinda Carpenter, Josep Call, Tanya Behne, and Henrike Moll (2005), Understanding and sharing intentions: The origins of cultural cognition. Behavioral and brain sciences 28, pp.~675--735 \item[] Wedgwood, R., 2007. The Nature of Normativity, Oxford: Oxford University Press. %\item[] Hurley, S., 1998. Consciousness in Action, Cambridge, MA: Harvard University Press %\item[] Williams (2000). The sociality of language %\item[] Wittgenstein, Ludwig (1969). Uber Gewissheit. On Certainty. Blackwell, Oxford. %Also Hacking 1986, Dummett 1986, Rysiew 2000 and Talmage 1996. % Rorty? Haack? \end{description*} \end{document} %look at chomsky new horizons? from http://www.ucl.ac.uk/philosophy/LPSG/Language.htm#Conventions Language, Conventions \& Idiolects Is there such a thing as a common language shared by a social group? Or is the notion of a shared language simply a socio political fiction? Chomsky gives one reason for rejecting shared languages and Davidson another. Davidson, D. 1986. 'A Nice Derangement of Epitaphs'. In E. Lepore, ed., Truth & Interpretation: Perspectives on the Philosophy of Donald Davidson. Oxford: Blackwell. Dummett, M. 1986. 'Reply to Davidson'. In E. Lepore, ed., Truth & Interpretation: Perspectives on the Philosophy of Donald Davidson. Oxford: Blackwell. Burge, T. 1975. 'On Knowledge and Convention'. Philosophical Review 84: 249-255. —.1989.'Wherein is Language Social?'. In A. George, ed., Reflections on Chomsky. Oxford: Basil Blackwell. George, A. 1990. 'Whose Language is it Anyway? Some Notes on Idiolects'. Philosophical Quarterly 40: 275-298. Higginbotham, J. 1989. 'Knowledge of Reference'. In A. George, ed., Reflections on Chomsky. Oxford: Basil Blackwell. Laurence, S. 1996. 'A Chomskian Alternative to Convention Based Semantics'. Mind 105: 269-301. Lewis, D. 1975. 'Languages and Language'. Reprinted in Philosophical Papers. Vol.I. Oxford: Oxford University Press, 1983. Peacocke, C. 1976. 'Truth Definitions & Actual Languages'. In G. Evans, and J. McDowell, eds., Truth & Meaning: Essays in Semantics. Oxford: Clarendon Press. Schiffer, S. 1993. 'Actual-Language Relations'. In J. Tomberlin, ed., Language and Logic. Philosophical Perspectives, vol. 7. Atascadero, Calif.: Ridgeview. ----- Here be dragons Davidson: "no such thing as a language" "I conclude that there is no such thing as a language, not if a language is anything like what many philosophers and linguists have supposed. There is therefore no such thing to be learned, mastered, or born with. We must give up the idea of a clearly defined shared structure which language-users acquire and then apply to cases. And we should try again to say how convention in any important sense is involved in language; or, as I think, we should give up the attempt to illuminate how we communicate by appeal to conventions." Davidson changed his views on language in his Epitaphs paper (1986), from insisting on a finite formal system to mutually interpretable communication -- but according to Williams (2000), he doesn't go for enough in approaching a Wittgensteinian view of language. I will argue that his conclusion that there is no such thing as a language is a natural outcome of the way his definitions are set up, and that there is ample reason to believe there is such a thing as language, namely in the form of shared conventions. Most notably, Davidson's individualistic bias prevents him from reaching a sufficiently social conception of language. In fact, to me Davidson's claim that there is no such thing as a language seems so counterintuitive as to be a violation of what Wittgenstein would call a certainty -- because if there would be no language, one wouldn't be able to observe this, let alone advertise it in a paper. I want to contend that regardless of the specific theoretical notions of language employed in linguistics or philosophy, it is beyond question that we have a common sensical, pretheoretical concept of language. Philosophy and linguistics are about idealizations of this concept of language, but I think it simply does not make sense to say whether these idealizations "exist" or not, as Davidson does. References: Davidson (1986). A nice derangement of epitaphs Williams (2000). The sociality of language Also Hacking 1986, Dummett 1986, Rysiew 2000 and Talmage 1996. ------ "Wittgenstein's point in these extended passages is that the divide between these two positions is, practically speaking, uninteresting.39 Both stances are effectively indistinguishable when we consider them in the realm of the ordinary." -- Hutto (1996) comparing idealism vs. realism Communication requires correlated responses to shared environmental stimuli. Speakers, it would seem, need not share specifically linguistic conventions, so Davidson can retain his position that communication does not require shared “language”. -- Williams (2000) -- timecube - gibberish Davidson relies on notion of first meanings, in other words literal meanings. however, this category is introduced by linguists to abstract over figurative meanings and other deviations from "normal" usage. In practice, however, no such sharp dividing line can be drawn, and the division plays no role in everyday language usage (except perhaps in pathological cases such as autists who are less competent with figurative language). "Incidentally, Haack's view of Davidson was off the mark as well. In the talk that I attended, she actually agreed with Rorty's criticism of Davidson's views in "A Nice Derrangement..." and charged Davidson with "trivializing the notion of language" in his later work. Both Rorty and Haack are wrong on this. Davidson's latter views are the logical consequences of his earlier work on interpretation and are just part of his developed "Unified theory"." -- http://www.newcriterion.com/articles.cfm/rortyism-haack-3261