ABSTRACT

In this paper I consider some of the most promi-

nent philosophical challenges to the viability of
Chomskyan linguistics. Te challenges in ques-
tion are generated by the work of Quine, Kripke
and Crispin Wright. I respond to these challenges
by developing an account of rule representation
that appeals to the lower level causal workings
of a particular component of the mind-brain
that plays a fundamental role in grounding our
linguistic capacities. Tis account draws upon
various elements of Chomskys core commit-
ments. Tese include his modularity thesis, his
view that the language faculty owes its status as
such to its relations to other in-head systems, his
general conception of the relationship between
the mind and the brain, and his nativist concep-
tion of language acquisition.
Key words: Chomsky, modularity, representa-
tion, nativism, ceteris paribus laws.
Introduction
Despite its massive infuence within
linguistics and cognitive science many
philosophers have been exercised by the
worry that there is something deeply
problematic with Chomskyan linguis-
tics.
1
In this paper I will examine and
attempt to answer what are to my mind my mind mind
three of the most interesting and impor-
tant philosophical objections to Chom-
skys work. Te objections in question
are due, respectively, to W.V. Quine, Saul
Kripke and Crispin Wright. My response
rests upon an account of rule represen-
tation that places substantial emphasis
on the causal factors underlying our lin-
guistic abilities and is motivated by the
scientifc ambitions of Chomskys work,
his commitment to the modularity thesis
and his nativist conception of language
acquisition.
1
Examples of prominent philosophers who have di-
rectly taken issue with Chomskys core commitments
include Putnam 1967, Baker & Hacker 1984, Searle
1990, Dummett 1989, Devitt & Sterelny 1987 and De-
vitt 2003.
5
CONFRONTING PHILOSOPHICAL
OBJECTIONS TO CHOMSKYAN
LINGUISTICS
MARK J. CAIN
Oxford Brookes University
EUJAP VOL. 1 No. 2 2005
ORIGINAL SCIENTIFIC PAPER
UDK: 1:81
81-05 Chomsky, N.
1. Quine and Kripkes objections
According to Chomsky rules are central to an individuals language: when an individual
knows a language there are a whole battery of rules that constitute her language, rules
that she knows and draws upon in language production and comprehension.
2
Chom-
sky characterises the knowledge in question is a species of propositional rather than
ability knowledge and it is typically unconscious or tacit (Chomsky 1980). Tere is
nothing more to an individuals language than the rules that she knows. Consequently,
in seeking to describe an individuals language and explain its acquisition, the linguist is
engaging in cognitive psychology.
3
For such an engagement in cognitive psychology to
have any point there would have to be a fact of the matter as to what rules were known
by the individual and thus constituted her language. What Quine and Kripkes argu-
ments suggest is that there is no such fact of the matter so that Chomsky is searching
for facts where there are no facts to be found. I will examine these arguments in turn
beginning with Quines.
Quines argument runs as follows (Quine 1972). Tere are cases where it makes perfect
sense to attribute to an individual speaker of a given language knowledge of a particular
body of rules (a grammar) and identify those rules as the ones that she draws upon in
language production and comprehension. Consider two groups of Danish children who
have learnt English in diferent ways. Members of the frst group have been taught tra-
ditional schoolroom grammar which they explicitly consult in constructing sentences
of English. Members of the second group have been taught Jespersons grammar which
they explicitly consult in constructing sentences of English. Tese rival grammars are
extensionally equivalent and equally ft the linguistic behaviour and dispositions of all
the children. Nevertheless, the frst group of children are guided by one grammar con-
sisting of a particular body of rules and the second are guided by another grammar
consisting of a diferent body of rules. Hence, it is perfectly legitimate to prefer one
account of the grammar of the frst group and a diferent account of that of the second
group. But, argues Quine, in the case of a typical child whose frst language is English,
there is no such case of being guided by a particular grammar. Such speakers are un-
able to state any rules of the language that they speak; at best a particular grammar fts
the linguistic behaviour and dispositions of the child. But there will always be distinct
grammars that equally ft their linguistic behaviour and dispositions, grammars that
2
With the advent of government and binding theory and the development of the principles and parameters ap-
proach rules were replaced by principles in Chomskys theorising (Chomsky 1981; Chomsky 1986). Tis is because,
for Chomsky, talk of rules carries with it connotations of specifcity to particular languages. If knowing a language
involved representing rules in this sense then it would be difcult to see how a child could acquire knowledge of her
language. Hence, a linguistic theory that talked of rules would not meet the condition of explanatory adequacy, that
is, the condition of explaining language acquisition. For the purposes of this paper nothing hangs on this change as
principles are a species of rules, namely, rules of great generality that Pinker dubs super rules (Pinker 1994). Accord-
ingly, unless otherwise indicated, I will use the term rule to cover both rules and principles in Chomskys narrow
sense of those terms.
3
For an early and classic statement of this psychologistic view of linguistics, Chomsky 1965, Ch. 1.
6
EUJAP Vol. 1 No. 2 2005
are extensionally equivalent. Tere can be no grounds for saying of any one of these
grammars that it is the grammar of the childs language or that the linguistic theory that
attributes this grammar to the child is the correct theory of the childs language. If this
is right then there are clearly going to be rival accounts of the rules of any individuals
language that are equally legitimate even when all the potentially relevant facts about
the individual are known. Consequently, there is no fact of the matter as to what par-
ticular body of rules belongs to a typical individuals language, as to what grammar is
true of her language.
Te second argument is due to Kripke and can be described as follows (Kripke 1982).
Suppose that an individual intends to use the symbol + to denote a particular math-
ematical function and so intends her use of this symbol to be governed by a rule cor-
responding to that function. Te rule is such that she should answer any question of the
form what is x + y? by specifying the number that is the value of the function for the
arguments x, y. At any point in her history she will have applied this rule only fnitely
many times. Moreover, being a fnite creature, she will be disposed to give answers to
only fnitely many questions of this form. Suppose that up to a particular point in time
she has always given the sum of x and y in answer to every problem of the form what
is x + y? that she has encountered but that none of these problems were such that the
value of either x or y was greater than 56. She then encounters the problem what is 57
+ 68? and answers 125. Is that answer correct? Is it in accord with the rule that she
intended to follow in her use of +? If the rule is the addition rule (that is the rule such
that one should give the sum of x and y as the answer) then her answer is correct. But
there will be some other rule (the quaddition rule) that equally fts her past behaviour
but requires her to give the answer 5 to this new problem. Which of these two rules
has she been intending to follow? Which of them is the rule governing her use of +?
According to Kripke, there is no fact of the matter as to whether or not she intended to
follow the addition rule; her past behaviour and conscious mental states are consistent
with an intention to follow this rule but are equally consistent with an intention to fol-
low some other rule. Appeal to her dispositions wont settle the issue. Her dispositions
might serve to undermine the claim that she intended to follow the quaddition rule.
But as her dispositions are fnite and the demands of rules such as the addition rule are
infnite, both the addition rule and some distinct quaddition like rule will equally ft
her dispositions.
Although Kripke focuses on a mathematical example, one can easily envisage an ana-
logue involving the kinds of rules that Chomsky is concerned with. Suppose that a lin-
guist claims that a particular rule is (tacitly or non-consciously) known by a particular
language user and so belongs to her language. Given the infnite demands of such rules
(that is, given the fact that they typically have implications for the grammaticality or
otherwise of infnitely many distinct combinations of words of the language to which
they belong) if the postulated rule fts the language users history of behaviour, her
conscious mental states and her dispositions to behave, then so will some other non-
M. J. Cain Confronting Philosophical Objections to Chomskyan Linguistics
7
equivalent rule. So there can there be no fact of the matter as to whether the postulated
rule rather than some other competing rule is known by the language user and so be-
longs to her language.
Chomsky has engaged with both Quine and Kripke suggesting that he takes their argu-
ments very seriously indeed. However, I do not think that Chomskys responses directly
confront the arguments as I have described them. With respect to Quine, his tendency
has been to interpret Quine as doing little more than making the point that linguistic
theories are underdetermined by observable evidence. Chomsky is willing to concede
this point but thinks that this does nothing to undermine linguistics as scientifc theo-
ries are generally underdetermined in this way.
4
I think that this is to misunderstand
Quines argument. For, he is not making a point about the epistemic plight of the lin-
guist, the point that she does not have conclusive evidence as to which of the competing
linguistic theories is true, as to which collection of rules belongs to the language of the
target individuals. Rather, his point is that there is no fact of the matter as to which of
the competing theories is true, as to which collection of rules belongs to the language
of the target individuals. Tus, Quines argument bears more in common with his thesis
of the indeterminacy of radical translation (Quine 1960) than it does with the so called
Duhem-Quine thesis (Duhem 1982; Quine 1951).
With respect to Kripke, Chomskys response focuses not so much on the argument for
rule scepticism but on the attempted sceptical solution to the problem of rule follow-
ing, a solution that places substantial emphasis on the role of the community.
5
Conse-
quently a direct response to Quine and Kripkes objections is needed and in the remain-
der of this paper I shall attempt to develop such a response.
An important point to be made concerns Chomskys talk of knowledge. He frequently
says that the rules belonging to an individuals language are items of (propositional)
knowledge, albeit unconscious or tacit knowledge. However, in the face of philosophi-
cal objections he is prepared to drop talk of knowledge, for example, by replacing it
with talk of cognizing rules. Accordingly, I dont think that Chomsky need be commit-
ted to the claim that we know rules of the kind that he postulates in the sense of the
term know that philosophers normally operate with. Rather, his commitment is that
determinate rules are represented within us where representing a particular rule is a
psychological state that can difer in salient respects from a state of knowledge.
6
Tus,
the rules that constitute an individuals language are represented within her mind but
she is not typically conscious of these rules. Consequently, the task becomes that of
4
For example, see Chomsky 2000, Ch 4.
5
See Chomsky 1986, Ch 4.
6
Here is a typical expression of this view: Evidently, each language is the interplay of two factors: the initial state
and the course of experience. We can think of the initial state as a language acquisition device that takes experi-
ence as input and gives the language as an output an output that is internally represented in the mind/brain,
Chomsky 2000, p 4.
8
EUJAP Vol. 1 No. 2 2005
giving an account of how a determinate rule can be represented within an individuals
mind that survives Quine and Kripkes objections.
It might be thought that producing such an account of rule representation would re-
quire producing a naturalistic theory that appealed only to scientifcally respectable
properties that resided at a level lower than the psychological or linguistic.
7
Even as-
suming that there would have to be a true naturalistic theory of this kind for a Chom-
skyan linguistic theory to be true,
8
it is not incumbent on us to produce such a theory.
Tis has to do with Chomskys repeated insistence that linguistics should be treated
no less charitably than any other science. To be guilty of methodological dualism is
to operate with double standards by making demands of one science (linguistics, in
particular) that one doesnt make of other more established sciences. Now very few
scientists, be they geologists, biologists, or whatever, would be in a position to present a
relevant naturalistic theory in order to defend their scientifc claims but no-one would
dream of questioning the legitimacy of their endeavours for that reason alone. Tus, an
inability to present a naturalist theory in the case of linguistics does not in and of itself
undermine the scientifc legitimacy of delivering a particular theory as to what rules
are represented in an individuals mind/brain.
Te most signifcant consequence of not needing a naturalistic theory is that we can
legitimately appeal to psychological and linguistic phenomena in responding to the
challenge of Quine and Kripke so long as the reality of those phenomena is not a matter
of controversy. Indeed, there are many psychological and linguistic facts the reality of
which most philosophers would concede. Here are some examples. Most human indi-
viduals have mastery of a language and know the meanings and sounds of many words
and expressions of the language that they have mastered. Individuals routinely make
correct judgements as to the grammaticality of expressions composed of words of their
language. Tere are syntactic generalisations true of languages that their speakers typi-
cally satisfy: for, example, in English the head of a phrase comes before its complement
and speakers of English typically produce sentences that satisfy this generalisation and
reject as ungrammatical sentences that they encounter that do not.
9
7
Such a theory would be the analogue of the naturalistic theories of content that Jerry Fodor has attempted to
develop (Fodor 1987; Fodor 1990).
8
Tis is an assumption that I doubt that Chomsky would endorse given his opposition to physicalism. Chomsky
(Chomsky 2000) argues that physicalism is an incoherent doctrine as a result of Newtons destruction of Cartesian
contact mechanics. For discussion see Lycan 2003, Strawson 2003, Poland 2003 and Chomsky 2003.
9
To endorse the view that it is legitimate to appeal to non-controversial linguistic and psychological phenomena
might be seen as problematic in the context of the discussion of Kripke as his scepticism is intended to range so
far as to undermine the idea that there are any determinate facts concerning what we mean by our words. My
answer to this worry is to point out that my concern is specifcally with the viability of linguistic theories that
many philosophers who are not generally sceptically inclined fnd problematic. Tis entitles me to take for granted
phenomena the reality of which they would not ordinarily be inclined to question, such as those described above.
Tis might involve engaging with an argument that isnt quite what Kripke had in mind. Hence, it might be more
accurate to talk of an objection inspired or suggested by Kripke rather than Kripkes objection. However, for ease
of exposition I shall continue to talk of Kripkes objection. Te justifcation for not directly engaging with Kripkes
radical scepticism is that if such scepticism is a problem for the Chomskyan then it is a problem for everyone. If
M. J. Cain Confronting Philosophical Objections to Chomskyan Linguistics
9
2. Te goal of linguistics
What is needed then is an account of how a determinate rule of the kind that Chomsky
postulates could be represented in the mind/brain of an individual. I dont think that
Chomsky has fully developed such an account. In response to the related issue as to
what it is for a rule to be psychologically real Chomsky has tended to reply that it is just
a matter of the truth of the linguistic theory that attributes the rule.
10
I sympathise with
those psychologists and philosophers
11
who complain that Chomsky might do more
to tell us what is involved in a linguistic theorys being true or how rules could be rep-
resented in the mind/brain. However, I do not think that the situation is hopeless and
prominent elements of Chomskys output do provide suggestions that can be developed
in such a way as to provide the required account of rule representation. To see this it is
necessary to take a more detailed look at Chomskys core commitments.
From the Chomskyan perspective the goal of linguistics is to uncover the workings of
the language faculty. Te language faculty is a distinct component of the mind/brain
that plays a fundamental role in knowing and acquiring language and is implicated
in language production and comprehension. Chomsky (Chomsky 1980) compares the
language faculty to other components of the mind/brain such as the visual system and
with organs of the body such as the heart and the liver (indeed, he sometimes calls
the language faculty a mental organ). Taking these comparisons seriously suggests that
Chomsky is committed to a modularity theory akin to that of Fodor (Fodor 1983) ac-
cording to which the language faculty is a distinct causal mechanism located in the
brain that plays a fundamental role in language processing.
12

Consider the heart, a distinct component of the body whose defning function is to
pump blood around the body and so deliver oxygen to other organs. Te heart couldnt
perform this function unless it was suitably connected to other organs. Something
could be intrinsically just like my heart without being a heart because it was hooked up
to other bodily organs in fundamentally diferent ways. Nevertheless, we can study the
heart in relative isolation from other bodily organs seeking to discover how it works
and so performs its defning function. Something similar is true of the language faculty
as conceived by Chomsky. Te language faculty interacts with other systems of the
mind/brain such as thought, perception and articulatory systems and owes its status as
a way were found of dealing with such a scepticism so as to vindicate our commonsense assumptions about lan-
guage and the mind, then that response could combined with the account of rule representation I shall develop to
constitute a complete defence of Chomsky against Kripkes radical scepticism.
10
For example, Chomsky identifes a claim for the psychological reality of the posits of a particular linguistic
theory with a claim for the truth of a certain theory. Chomsky 1980, p. 191.
11
For example, Pylyshyn 1991 and Rey 2003a; Rey 2003b.
12
Tis involves rejecting Fodors (Fodor 1981; Fodor 2000) own reading of Chomsky according to which his ver-
sion of the modularity thesis is not an architectural one but the idea that speakers of a language have a particular,
largely innate, body of knowledge or beliefs specifcally about language that they draw upon in language produc-
tion and comprehension. For an interesting discussion of Fodors reading of Chomsky see Collins (Collins 2004).
10
EUJAP Vol. 1 No. 2 2005
the language faculty to its relations with those systems. Similarly, Chomsky describes
an individuals language as a state of her language faculty (and, thus, a state of her
brain). Tis brain state would not be a language were it not for the connections that it
bears to components of the mind/brain distinct from the language faculty.
13
If the task of linguistics is to uncover the workings of the language faculty then the
question arises as to the precise function of the language faculty. For Chomsky, when
it has reached a mature state, the language facultys function is to generate expressions
from the lexical items belonging to the individuals language by means of the operations
of merger and move.
14
Te expressions so generated have two components, a sound and
a meaning. Tus, the language faculty works to efect a particular pairing of sounds and
meanings. When the language faculty operates so as to produce an expression, that ex-
pression is an internal phenomenon that resides at the interface between the language
faculty and related systems concerned with thought and articulation. In particular, the
expression consists of a pair of representations: an LF (Logical Form) representation
that resides at the interface with the thought system, and a PF (Phonetic Form) repre-
sentation that resides at the interface with the articulatory system. Chomsky sometimes
describes such representations as instructions. For example, a PF representation is an
instruction to the articulatory system to produce a particular sound. Refecting the
above point that the language faculty is a language faculty (and its state a language)
only insofar as it is appropriately related to other systems of the mind/brain, an internal
phenomenon is an expression (a particular pairing of an LF and a PF representation)
only insofar as it is appropriately related to other systems of the mind/brain. Such a
relationship might involve having a particular causal impact on such systems so that,
for example, a particular LF representation owes its status as such because it causes the
articulatory system to produce a particular sound.
Chomskys systematic use of the term mind/brain and his comparison of the language
faculty with bodily organs refects a general view of the relationship between the mind
and the brain that can be described in the following terms. Te mind and the brain are
not two distinct systems. To describe an individuals mind is to describe her brain in
mental terms; it is to describe her brain at a higher level of abstraction than the neuro-
scientist does. Alternatively, to describe the brain in neuroscientifc terms is to describe
the mind at a lower level of abstraction than the psychologist does. Tus, the language
faculty is ultimately a component of the brain and the rules represented within it are
somehow physically realised in the brain. Hence, A person who speaks a language has
developed a certain system of knowledge, represented somehow in the mind and, ulti-
mately, in the brain in some physical confguration (Chomsky 1988, p. 3).
13
See Chomsky 2000, p. 27 for a particularly clear expression of such views.
14
Merger is the operation of concatenating lexical items and movement is the operation of moving an element of
an expression from one place in the expression to another.
M. J. Cain Confronting Philosophical Objections to Chomskyan Linguistics
11
So the task of linguistics is to uncover the way in which the language faculty, a system
housed in the brain, generates expressions. If we combine this with the idea that engag-
ing in this task involves specifying a body of rules represented in the language faculty
then the implication is as follows: the rules represented in an individuals language
faculty is a matter of the way in which that faculty generates the expressions of her
language.
3. Rule representation
Te claim that the rules represented in the language faculty is a matter of the way in
which that faculty generates expressions suggests two things. First, that there are rules
that ft the behaviour of the language faculty. In order for a particular rule to ft the be-
haviour of the language faculty that system must be capable of behaving in accord with
the rule in question, say producing expressions permitted by the rule. Moreover, the
language faculty must not be capable of behaving in confict will the rule, for example,
of producing an expression not permitted by the rule.
Considered against the background of Chomskys views concerning the relationship
between the mind and the brain, the second point has to do with how the expression
generating behaviour is grounded by lower level processes. As the language faculty is a
component of the brain whenever that faculty produces an expression a causal process
will take place that can be described both in linguistic terms and in terms that reside at
a level lower than the linguistic (for example, in neural terms). Describing the workings
of an individuals language faculty in linguistic terms involves describing the workings
of a component of her brain at a higher level of abstraction than the neuroscientist
does. In efect, the linguists task it to describe the workings of the brain in terms of
its linguistic signifcance. Consequently, which linguistic descriptions are true will be
bound up with, and constrained by, which lower level neural descriptions are true.
What all this suggests is that for an individuals language faculty to represent a rule not
only must the rule ft its behaviour. But, in addition, all instances of behaviour covered
by the rule that the system engages in must exhibit a salient similarity from the lower
level perspective. In other words, there must be a common lower level causal process
going on whenever the system behaves in accord with the rule.
To see how this applies to the linguistic case consider the following example. A promi-
nent hypothesis within the principles and parameters approach that emerged in the
1980s is that it is a rule of English that the head of a phrase comes before its comple-
ment. Suppose that this rule is represented in the language faculty of English speakers.
Ten, whenever the language faculty of an English speaker produces an expression, it
will consist of phrases where the head comes before its complement. In addition, all ex-
pressions will have this characteristic as the result of the workings of a common lower
level causal process. To claim that the head frst rule is represented in an individuals
12
EUJAP Vol. 1 No. 2 2005
language faculty is therefore to make a substantial explanatory move. For it is to tell us
that the expressions produced by her language faculty have a salient characteristic and
it is to tell us, in abstract terms, how those expressions are generated and why they have
that salient characteristic.
Te explanatory signifcance of the claim that the head frst rule is represented can
be seen by contrasting it with an alternative hypothesis. According to the alternative
hypothesis, the individual has several distinct rules represented in her language faculty
that together ensure that any expression produced will consist of phrases where the
head comes before its complement. Tese rules include the rule that in a noun phrase
the noun comes before its complement, the rule that in a verb phrase the verb comes
before its complement, the rule that in a prepositional phrase the preposition comes
before its complement, and so on. An implication of this hypothesis is that distinct
lower level processes are implicated in the construction of diferent types of phrases so
that there is no single, general explanation as to why heads precede their complements.
Rather, that general feature of expressions is to be explained as a product of several
distinct and independent lower level processes. Hence, the alternative hypothesis ofers
a contrasting account of the workings of the language faculty.
Generalised, the idea is this. Te task of linguistics is to explain how the language fac-
ulty works in generating expressions. Te linguist does this by specifying a body of
rules for generating expressions that are represented in the language faculty. Such an
attribution of rules implies that those rules ft the language facultys expression gen-
erating behaviour and provides an abstract account of the lower level causal basis of
this behaviour. Tis account is abstract because it does not tell us anything about the
specifc nature of this causal basis. For example, it does not provide any neural details of
the workings of the language faculty. Tis leaves the way open for an incorporation of
linguistics into neuroscience when these neural details are uncovered, something that
Chomsky sees as the ultimate goal of linguistics.
None of this implies that linguistics reduces to neuroscience for two reasons. First, we
should expect the rules represented in the language faculty to be multiply realisable
so that individuals whose brains were physically quite diferent could still share a lan-
guage. Second, because the language facultys status as such depends upon its relation-
ship to other subsystems of the mind/brain, in principle something could be just like
the human language faculty from the neuroscientifc perspective without representing
any linguistic rules at all. Tis would be the case if the subsystem in question drove
locomotion rather than being connected to modules concerned with thought and ar-
ticulation.
In suggesting that the manner in which the language faculty represents rules should
be seen as a species of representation that is to be understood in causal terms, I have
committed myself to the idea that there is a complex two way relationship between the
output of the language faculty (understood to include its efect on modules concerned
M. J. Cain Confronting Philosophical Objections to Chomskyan Linguistics
13
with thought and articulation) and its internal workings. On the one hand, the identity
of the rules internal to the language faculty partly depends on properties of the expres-
sions that it produces that are independent of the precise workings of that faculty. Tat
is to say, on properties that competing linguists could attribute to expressions whilst
holding divergent views of the language facultys internal workings. For example, for
the head frst rule to be represented it is necessary that the language faculty gener-
ally produces expressions which consist of phrases where the head comes before its
complement.
15
On the other hand, some salient properties of expressions are directly
determined by the workings of the language faculty so that it is necessary for the lan-
guage faculty to work in a particular way from the lower level perspective in order
for expressions to have that property. To see this consider the following. Typically an
expression will be the product of the application of more than one rule so that more
than one distinct lower level process will have been involved in its generation. Te
fne-grained syntactic structure of the expression (as represented by a linguists tree
diagram) will be an interaction efect of these lower level processes. So, the extent to
which two distinct expressions share a common syntactic structure partly depends on
the extent to which they were generated by means of the interaction of common lower
level processes. Once generated an expression is subject to further processing as when
an element is moved from one place to another in the expression. Once again, any such
processing will have a lower level description. Just as the lower level causal processes
implicated in the generation of an expression play a role in determining its fne grained
syntactic structure, so do the lower level processes that it can in turn be subjected to.
So, the extent to which two distinct expressions share a common syntactic structure
partly depends on the extent to which they can be subjected to further common lower
level processing.
To talk in terms of states and processes that reside at a level lower than the linguistic is
to leave unanswered the question of how such states and processes are to be described
and individuated. Tis is not a question that one should expect the linguist to answer
but there is no problem here. Tis is because it is commonplace for the scientist to
develop theories whose truth-value is sensitive to facts concerning the nature of lower
level phenomena the precise details of which the scientist is ignorant. For example,
Mendel was able to make considerable progress in genetics long before the discovery
of DNA.
16

15
Tis implies that there is a fundamental distinction between the language faculty and other systems whose be-
haviour might be said to accord with rules. For example, suppose that it were claimed that the behaviour of a planet
accorded with a rule corresponding to Keplars law of planetary motion. Tis would provide no reason to think of
the planet as representing that rule as its behaviour does not have any psychological description. Te planet merely
follows a particular elliptical trajectory whereas the language faculty is part of a complex system that understands
and employs a language.
16
To account for rule representation in terms of lower level causal processes echoes the account of tacit knowl-
edge of the axioms of theories of meaning devised by Evans 1981 and developed by Davies 1987. In an important
paper Wright (Wright 1986) criticises this account. Whatever its power, I dont think that Wrights points have any
direct bearings on my position due to salient diferences between Chomskys project, on the one hand, and that of
14
EUJAP Vol. 1 No. 2 2005
4. Dealing with Quine and Kripke
How does the account of rule representation that I have been developing deal with the
challenges generated by Quine and Kripke? Recall that Quine argued that numerous
distinct yet extensionally equivalent grammars of English will ft the linguistic behav-
iour of an individual whose frst language is English. In such a case, in contrast to
the Danish children of his example, the English speaker is not guided by any particu-
lar grammar. Te account of rule representation that I have developed deals with this
problem in a straightforward way by suggesting that there is a real sense in which an
individuals linguistic behaviour is guided by a particular grammar. Te correct gram-
mar is that of the competing extensionally equivalent grammars that corresponds to
the lower level processing executed by the language faculty in the course of generating
expressions. For example, suppose that there is a choice between a grammar containing
a general head-frst rule and one containing numerous more specifc rules that together
have the same implications for linguistic behaviour as the general rule. Ten, if it is the
case that all expressions conforming to the general rule are the product of the execution
of a common lower level process by the language faculty, that fact will tell against the
advocate of the latter grammar. If, on the other hand, there is no such commonality,
then that fact would tell against the former grammar. Alternatively, suppose that two
competing grammars consist of equally specifc rules, but difer with respect to the
syntactic structures that they assign to various expressions of an individuals language.
Once again, patterns of commonality and diference in lower level processing involved
in the production of expressions and the further processing that the expressions can be
subjected to may tell against one grammar and count in favour of its competitor. For
example, suppose that grammar G
1
assigns a particular structure to expression E
1
and
an analogous structure to E
2
whereas grammar G
2
assigns distinct structures to each
of these sentences. Were it the case that a common lower level process was involved
in the production of the expressions and that both expressions were subject to com-
mon further lower level processing, then that fact would undermine G
2
and support G
1
.
However, were it the case that quite distinct lower level processes were involved in the
production of the expressions and that the expressions were subject to quite diferent
further lower level processing, then that would count for G
2
and against G
1
. In short, the
nature of the lower level processes implicated in the generation and further processing
of expressions determines which of the distinct yet extensionally equivalent grammars
is true as an account of an individuals language.
I now come to the problems generated by Kripke. Recall that Kripke argued that an ap-
peal to dispositions could not succeed in explaining how it is possible for an individual
to grasp a determinate rule as, being fnite creatures, our dispositions are fnite whereas
the demands of a rule are typically infnite. Tis threatens to undermine my account
those philosophers who have engaged in the armchair project of constructing a theory of meaning, for example,
Dummett 1993 and Davidson 1983.
M. J. Cain Confronting Philosophical Objections to Chomskyan Linguistics
15
of rule representation as that account requires that linguistic rules that have an infnite
character ft the behaviour of the language faculty.
Te frst element of my response to Kripkes objection is to question the assumption
that a fnite creature cannot have infnite dispositions. Te behaviour of plenty of f-
nite physical systems are governed by laws of nature that are such as to imply that the
system is disposed to behave in a distinctive way in each of infnitely many possible
situations. For example, a particular body of gas that at this moment has a particular
volume, temperature and pressure. Te gas laws determine what volume the gas would
occupy when subjected to any of infnitely many distinct pressures at any given tem-
perature. Hence, the laws imply that the gas has infnite dispositions with respect to the
volume it occupies in response to pressures that it could be subjected to.
Te case of laws governing the behaviour of fnite systems can help defeat Kripkes
objection if there are appropriate laws governing the workings of the language faculty
and here there are grounds for optimism. For one thing, Chomsky conceives of linguis-
tics as being a science, the science of the language faculty. Sciences typically uncover
laws and construct explanations that appeal to such laws so that one might expect the
enthusiast for the scientifc status of Chomskyan linguistics to hold that the language
faculty is governed by laws and that these laws directly correspond to linguistic rules.
Moreover, the nativist commitments of Chomskyan linguistics suggests that we should
think of rules in terms of laws. For Chomsky a characterisation of Universal Grammar
(UG) describes the initial state of the language faculty, a state that places constraints
on the mature states that the language faculty can assume and determines how the lan-
guage faculty develops in response to impoverished linguistic stimulation during the
language acquisition period. As UG is innate it is part of our biological endowment, an
aspect of us that is genetically determined. Tis seems to imply that an accurate speci-
fcation of UG would capture a body of laws about the initial state of the language fac-
ulty and how its state changes in response to external stimulation. With respect to the
mature state of the language faculty it seems to imply that there are laws governing the
operation of that faculty when it is in any such state. For example, given that I was sub-
jected to expressions of a head frst language when I was acquiring my frst language, it
is a law of the workings of my language faculty that it generates expressions that accord
with the head frst rule. In short then, there are reasons for thinking that linguistic rules
are closely related to laws governing the workings of the language faculty so that the
behaviour of that system has the required infnite character.
Suppose that it is accepted that law governed fnite systems can have infnite disposi-
tions. Tere is still a problem as gases always behave in accord with the gas laws where-
as speakers of a language routinely violate the putative rules of their language. Te
language faculty is conceived as being connected to performance systems so that is
operations are refected in the linguistic behaviour of the individual. So, one might rea-
sonably ask, how can it be a law that the language faculty of individuals who encounter
16
EUJAP Vol. 1 No. 2 2005
head-frst phrases as children acquiring their frst language operates in such a way as
to ft the head frst rule? Te answer to this question involves pointing that any such
linguistic laws will, like all special science laws, be ceteris paribus laws (Fodor 1975;
Fodor 1987; Fodor 1991). Ceteris paribus laws have exceptions; it can be a law that,
ceteris paribus, As cause Bs, even though an A is sometimes tokened without causing a
B to be tokened. At this point it will be helpful to give an account of the nature of ceteris
paribus laws. Te account that I favour is explicitly championed by Pietroski and Rey
(Pietroski & Rey 1995) and implicit in Fodor (Fodor 1987).
Te natural world consists of many distinct systems. Any given special science will be
concerned with the workings of one such system. Te special scientist will conceive of
her system as relatively self-contained and independent of all others and will attempt
to uncover how it operates when free of external interference. When she states that a
particular law holds she is making a claim about how the system of her concern works
when free of external interference. But she recognises that in reality such interference
does take place. Tus, she qualifes her statement of the laws governing the workings of
the system in question with a ceteris paribus clause implying that any putative violation
of the law constitutes a case where an external factor has interfered with the natural
workings of the system. In such a case, the violation of the law can be explained by ap-
peal to the external interference. Fodor presents an illustrative example (Fodor 1987).
It is a geological law that, ceteris paribus, a meandering river erodes its outer bank.
Examples of cases where a meandering river fails to erode its outer bank include those
where the bank is concreted over, where the tiny abrasive particles are sieved from the
water and so on, cases where some extra geological event has interfered with the work-
ings of the geological system. Such putative violations of the law can be explained by
appeal to the interfering extra-geological event (for example, the river failed to erode
its outer bank because the bank was concreted over).
Tis account of ceteris paribus laws suggests the following line of thought.
17
Lingusitics
is concerned with the workings of a distinct system of the natural world, namely the
language faculty. Te laws governing the workings of the language faculty are ceteris
paribus laws so there can be exceptions to them. Te cases where I produce a phrase
where the head comes afer its complement, where I misjudge the grammaticality of
a sentence, where I am unable to deal with a long complex sentence, and so on, are
cases where some interfering, extra linguistic factor prevents my language faculty from
producing its natural efect. For example, my failure to process a long and complex
sentence may be due to memory limitations of the system feeding input to the language
faculty. My production of an ungrammatical sentence may be due to physical damage
to the language faculty or a breakdown in the workings of an adjacent system such as
the articulatory system. What some of these examples bring out is that the language
17
Tis line of thought is inspired by Rey and Pietroskis appeal to the nature of ceteris paribus laws in responding
to Kripkes critique of dispositionalism (Rey and Pietroski 1995).
M. J. Cain Confronting Philosophical Objections to Chomskyan Linguistics
17
faculty interacts with other in-head systems and requires the co-operation of these sys-
tems to ground our observable linguistic capacities. Consequently, the language faculty
can lose its ability to ground such a capacity either as a result of its own breakdown or
as a result of the breakdown of such adjacent systems. An example of the latter type of
case would be that where damage due to a stroke prevents an individuals articulatory
system from responding to the language faculty. But in such a case, the putative viola-
tion of a linguistic law is due to, and can be explained in terms of, the occurrence of
some extra linguistic phenomenon.
It should be noted that there are cases where all else isnt equal in the linguistic sphere
which do not involve the interference of extra-linguistic factors. Suppose that it is a rule
of English that the past tense form of a verb is the verb with -ed added to the end and,
thus, that it is a law that the language faculty of English speakers adds -ed to the end of
verbs to make their past tense forms. However, the past tense of sleep is slept and not
sleeped. All else is not equal when I say I slept well last night rather than I sleeped
well last night. But my behaviour in this case, though not constituting a genuine coun-
ter-example to the law, has a linguistic explanation. Te explanation has to do with the
lexicon, something that Chomsky describes as a list of exceptions (Chomsky 1995, p.
235). Te lexical entry for sleep will contain information about that item that is unique
to it, for example that its past tense form is slept. In contrast, the lexical entry for regu-
lar verbs such as talk will contain no information concerning their past tense form.
When I produce a sentence containing slept and reject as ungrammatical one con-
taining sleeped my so behaving is an interaction efect, the product of an interaction
between a general rule and an atypical lexical entry that instructs the language faculty
to treat the lexical item in question in a non-general way. Tus, the apparent counter-
example to the general rule or law and why it is not a genuine counterexample is to
be explained by appeal to the lexicon and so is to be explained in linguistic terms. Tis
example indicates another respect in which the linguist, like any special scientist, is
engaged in a process of abstraction; for, she is trying to separate out the distinct factors
that are involved in the occurrence of everyday phenomena that are typically interac-
tion efects.
In sum then, there are good reasons for thinking that there are linguistic laws and that,
despite our being fnite creatures, the language faculty may well operate in a way that
rules that have an infnite character ft.
Kripke ofers a second criticism of dispositionalism that, once again, needs to be dealt
with. Tis has to do with normativity and runs as follows. Rules have a normative di-
mension in that they determine facts about how we ought or should behave. Tere is a
gap between how we ought to behave and how we actually behave as we ofen do things
that we ought not. By tying the demands of the rules that govern our behaviour to
how we would behave in certain possible circumstances, the dispositionalist obliterates
this distinction. She implies that whatever we do or would do is right or correct. But if
18
EUJAP Vol. 1 No. 2 2005
whatever we do or would do is right or correct then it makes no sense to talk of right
and wrong or correct and incorrect and, thus, it makes no sense to talk of rules. Directly
applied to my account of rule representation the objection is that laws of nature have to
do with how a system for example, the language faculty does or would behave and
that there is a distinction between this and how we ought to behave if we speak a rule
governed language.
My response is to question the centrality of normativity to rules.
18
I do not wish to deny
that there are facts to do with right and wrong, correct and incorrect, associated with
many of the rules that philosophers have focussed their attention on. But such facts
might well be grounded in facts concerning human social practises and psychological
phenomena that need not be present in every case where rules are operative. Consider
the example of the rules of chess. In teaching children how to play chess we typically
explicitly state the rules of the game, encourage them to consciously consult the rules
when deciding how to behave and aim to instil in them a respect for the rules and a
desire to act in accord with them. When they violate the rules we criticise them and ad-
vise them to consider the rules and think again. And when we realise we have violated
a rule we reproach ourselves, regret our actions and, even, feel shame. My suggestion is
that normative facts surrounding chess are the products of such practices of teaching
and evaluating and the desires, regrets and states of shame that they give rise to. Tus,
in principle, such normative facts need not be present in all cases of rule governed be-
haviour. More specifcally, they need not be present when our behaviour comes to be
governed by a particular rule without any antecedent process akin to that involved in
cases of learning the rules of chess. In the case of the rules that Chomsky envisages, the
rules do not come to govern our behaviour as a result of such a process. Te rules have
a substantial innate basis and so there is no explicit instruction, criticism by appeal to
statements of the rule and resultant desires, regrets, shame, and so on. In short, there
is an explanation for the absence of normative facts surrounding linguistic rules such
as the head-frst rule and it is a mistake to generalise on the basis of a consideration of
rules that come to govern our behaviour on the basis of a process of learning in a social
context.
5. Wrights objections
I will now turn my attention to a paper by Crispin Wright (Wright 1989) in which
he explicitly considers the impact of Wittgensteins rule following considerations on
Chomskys work.
19
Wright suggests that the rule following considerations generate a
18
Such a move is far from unprecedented. For example, Horwich argues that meaning isnt inherently normative
(Horwich 1998).
19
See Wittgenstein 1953, 138-242 and Wittgenstein 1967, particularly Part VI, 15-49 and Part VII, 47-
60).
M. J. Cain Confronting Philosophical Objections to Chomskyan Linguistics
19
pair of related objections to Chomsky. Te frst runs as follows. Chomsky views the
rules belonging to an individuals language as a theoretical matter to be determined by
reference to factors that the individual is not necessarily in the best position to deter-
mine. Consequently, Chomsky cannot make sense of a salient feature of language, the
fact that language users have non-inferential knowledge of the rules of their language.
Wright relates this objection to Dummetts (Dummett 1976) views on knowledge of
language suggesting that he thinks that it is part of the concept of following a rule that
the rule follower has such non-inferential knowledge.
Te second objection draws upon Wittgensteins rejection of the idea that rules are inf-
nite rails whose demands are independent of the judgements of the rule follower whose
task it is to track those rails. Rather, argues Wright, for Wittgenstein, the demands of a
rule are bound up with the rule followers judgements in a manner that echoes the case
of colour concepts. Our judgements of colour determine the extension of our colour
concepts rather than refect a pre-determined extension. Similarly, the demands of a
given rule are related to our best judgements as to those demands. Wright accuses
Chomsky of endorsing this rules as rails mythology as he portrays the demands of the
rules of an individuals language as being determined by a syntactic mechanism that
settles which expressions of her language are well formed and which are not in advance
of her making any grammaticality judgements.
I will consider these objections in reverse order. Tere is a respect in which, for Chom-
sky, the demands of an individuals linguistic rules are determined by the state of her
brain in advance of her making any judgements. However, this is not to say that there
is no connection between those demands and the judgements that she would make, at
least on the account of rule representation that I have developed. Te state of my brain
causally grounds facts about the grammaticality judgements I do make along with facts
concerning the grammaticality judgements I would make in various possible circum-
stances. Were this link between my brain state and my judgements not to exist then the
demands of the rules of my language would be other than what they are. So, it might be
said that for Chomsky the demands of a rule are bound up with the individuals judge-
ments.
I take it that Wittgensteins target is Platonism. Now Chomsky is clearly not a Platonist
as for him language is a psychological entity. Hence, when an individual makes a lin-
guistic judgement she is not making a judgement about anything existing indepen-
dently of her; she is not trying to track rails laid down in some Platonic heaven. Tis is
why Chomsky (Chomsky 1986; Chomsky 2000) insists that we mustnt be mislead by
the term representation to think that the representations he talks about are representa-
tions of or about anything.
Nevertheless, even though Chomsky is an opponent of Platonism his picture of linguis-
tic rules is at odds with that Wright seems to have in mind as that picture fnds a sub-
stantial role for judgement in an ongoing process of shaping and settling the demands
20
EUJAP Vol. 1 No. 2 2005
of our rules. Whereas, for Chomsky, the demands of our linguistic rules are settled in
advance of our making any judgements as to their demands. Now there is considerable
plausibility in Wrights view when applied to many of the most familiar rules that we
follow. For example, think of rules governing the games that we invent. When we invent
a game we need to lay down rules governing the behaviour of the games participants.
Rarely can we envisage every situation that might be thrown up in the course of a game
so we do not explicitly specify what the rules imply with respect to every such situation.
Consequently, once we begin to play the game a contestant will occasionally behave in
a manner whose legality is far from clear to the contestants, the referee and any specta-
tors. In response to such cases judgements are ofen made sometimes there and then,
sometimes subsequently as to precisely what the rules demand in such situations. In
this way an ongoing process of judgement can shape the precise demands of a rule.
As portrayed by Chomsky, linguistic rules are not shaped by any such process of ongo-
ing judgement. Rather, the demands of any given rule are fxed prior to the individuals
actually making any judgements. However, there is nothing problematic about this as
why there should be such a diference between linguistic rules and the rules governing
games that we invent has a ready explanation. For, according to Chomsky, linguistic
rules have a substantial innate dimension. Consequently, unlike the rules we self-con-
sciously invent, there is no need for any ongoing process of judgement to shape and
determine the demands of the rule. In efect, the accusation is that Wright, noticing a
prominent feature of certain rules, makes an illegitimate generalisation to the conclu-
sion that all rules have this feature.
A similar response can be made to Wrights frst objection. We do have non-inferential
knowledge of many of the rules that we follow. But this doesnt refect a general fact
about the nature of rules. Rather, it refects a contingent fact about the manner in which
we acquire mastery of many of the rules that we follow. Te rules of chess are not in-
nate. On the contrary, the chess player has to learn them through a process of explana-
tion, deliberation, and correction. Tis process will ofen involve a conscious confron-
tation with a statement of the target rules as the learner is told what the rules are and
refects on such statements in her early attempts to play the game. Consequently, it is
hardly surprising that the competent chess player has non-inferential knowledge of the
rules of chess. As the rules that Chomsky postulates are innate rather than learnt, the
acquisition process does not involve any such conscious confrontation with statements
of the rule. Hence, it is no surprise that the individual fnds it difcult to state the rules
of her language. In sum, the accusation is that Wright once again makes an illegitimate
generalisation on the basis of a consideration of learnt rules and fails to appreciate the
signifcance of Chomskys nativism.
Both Kripke and Wrights objections draw upon interpretations of Wittgensteins rule
following considerations. Tis raises the general question of the relationship of the rule
following considerations to the line of thought that I have developed. Te rule follow-
M. J. Cain Confronting Philosophical Objections to Chomskyan Linguistics
21
22
EUJAP Vol. 1 No. 2 2005
ing considerations are concerned with the question of how, if at all, it is possible for a
particular rule to belong to an individuals language. A key aim of Wittgensteins is to
undermine mentalist attempts at answering this question as such attempts are either
circular or lead to infnite regress. In other words, they are explanatorily bankrupt. If
the account of rule representation that I have developed in this paper is viable then an
important implication is that Wittgenstein has not undermined mentalist attempts to
explain rule following per se. At best, he has only defeated the particular mentalist lines
of thought that he considers, for example, the appeal to mental images and the appeal
to interpretations.
Conclusion
In this paper I have attempted to defend Chomskyan linguistics against objections gen-
erated by the work of Quine, Kripke and Wright. I have done this by developing an
account of rule representation that makes a crucial appeal to the lower level causal
workings of a particular component of the mind/brain that plays a fundamental role in
grounding our linguistic capacities. But what exactly is the status that I accord to this
account? I should make it clear that I have not attempted to provide a constitutive ac-
count of rule representation, an account of the necessary and sufcient conditions for a
rule to be represented in the mind/brain. Rather, my aim as been to give an account of
one way in which determinate rules could be represented in the mind-brain that does
not rule out the possibility that there are other ways in which determinate rules could
be represented in the mind/brain. In other words, I have aimed to present a sufcient
condition for rule representation rather than a necessary and sufcient condition.
20
My
account draws upon various elements of Chomskys core commitments. Tese include
his modularity thesis, his view that the language faculty owes its status as such to its
relations to other in-head systems, his goal that linguistics ultimately be incorporated
into neuroscience, and his nativist conception of language acquisition. To that extent,
my account is motivated by Chomskys work and is designed to be attractive to the
Chomskyan.
20
Cf. Fodors approach to naturalising intentional content, Fodor 1990.
REFERENCES
Antony L. & N. Hornstein eds. (2003), Chomsky and His Critics, Oxford: Blackwell
Baker G.P. & P.M.S. Hacker (1984), Language, Sense and Nonsense, Oxford: Blackwell
Barber A. ed. (2003), (2003), Epistemology of Language, Oxford: Oxford University Press
Chomsky N. (1965), Aspects of the Teory of Syntax, Cambridge: MIT Press
Chomsky N. (1980), Rules and Representations, Oxford: Blackwell
Chomsky N. (1981), Lectures on Government and Binding, Dordrecht: Foris
Chomsky N. (1986), Knowledge of Language: Its Nature, Origins and Use, New York: Praeger
Chomsky N. (1988), Language and Problems of Knowledge, Cambridge: MIT Press
Chomsky N. (1995), Te Minimalist Program, Cambridge: MIT Press
Chomsky N. (2000), New Horizons in the Study of Language and Mind, Cambridge: Cambridge
University Press
Chomsky N. (2003), Replies, in Antony & Hornstein 2003, pp. 255-328
Collins J. (2004), Faculty Disputes, Mind and Language, 19, pp. 503-533
Davidson D. (1984), Inquiries into Truth and Interpretation, Oxford: Oxford University Press
Davidson D. & G. Harman eds. (1972), Te Semantics of Natural Language, Dordrecht: Reidel
Davies M. (1987), Tacit Knowledge and Semantic Teory: can a fve per cent diference matter?, Mind,
96, pp. 441-462
Dennett D. (1978), Artifcial Intelligence as Philosophy and as Psychology, in Ringle 1978, pp. 57-78
Devitt M. (2003), Linguistics is not Psychology, in Barber 2003, pp. 107-139
Devitt M. & K. Sterelny (1987), Language and Reality: An Introduction to the Philosophy of Language,
Oxford: Blackwell
Duhem P. (1982), Te Aim and Structure of Physical Teory, Princeton: Princeton University Press
Dummett M. (1975), What is a Teory of Meaning?, in Guttenplan 1975, pp. 97-138. Reprinted in
Dummett 1993, pp. 1-33
Dummett M. (1989), Language and Communication, in George 1989, pp. 192-212
Dummett M. (1993), Te Seas of Language, Oxford: Oxford University Press
Evans G. (1981), Semantic Teory and Tacit Knowledge, in Holtzman & Leich 1981, pp. 322-342
Fodor J.A. (1975), Te Language of Tought, Cambridge: Harvard University Press
Fodor J.A. (1983), Te Modularity of Mind, Cambridge: MIT Press
Fodor J.A. (1997), Psychosemantics, Cambridge: MIT Press
Fodor J.A. (1990), A Teory of Content and Other Essays, Cambridge: MIT Press
Fodor J.A. (1991), You can Fool Some of the People all of the Time, Everything Else Being Equal:
Hedged Laws and Psychological Explanation, Mind, 100, pp. 19-34
Fodor J.A. (1998), Concepts: Where Cognitive Science Went Wrong, Oxford: Oxford University Press
Fodor J.A. (2000), Te Mind Doesnt Work Tat Way, Cambridge: MIT Press
Fodor J.A. (2001), Doing without Whats Within Mind, 437, pp. 99-148
George A. ed. (1989), Refections on Chomsky, Oxford: Blackwell
Guttenplan S. ed. (1996), (1996), Mind and Language, Oxford: Oxford University Press
Holtzman, S. & C. Leich eds. (1981), Wittgenstein: To Follow a Rule, London: Routledge and Kegan
Paul
Horwich P. (1998), Meaning, Oxford: Oxford University Press
Kasher A. ed. (1991), Te Chomskyan Turn, Oxford: Blackwell
Kripke S. (1982), Wittgenstein: On Rules and Private Language, Oxford: Blackwell
Lewis D. (1970), How to Defne Teoretical Terms, Journal of Philosophy, 67, pp. 427-446
M. J. Cain Confronting Philosophical Objections to Chomskyan Linguistics
23
Lycan W. (2003), Chomsky on the Mind-Body Problem, in Antony & Hornstein 2003, pp. 11-28
Pietroski P. and G. Rey (1995), When Other Tings Arent Equal: Saving Ceteris Paribus Laws Te
British Journal for the Philosophy of Science, 46, pp. 81-110
Pinker S. (1994), Te Language Instinct, Harmondsworth: Penguin
Poland J. (2003), Chomskys Challenge to Physicalism, in Antony & Hornstein 2003, pp. 29-48
Pylyshyn Z. (1991), Rules and Representations: Chomsky and Representational Realism, in Kasher
1991, pp. 231-251
Putnam H. (1967), Te Innateness Hypothesis and Explanatory Models in linguistics, Synthese, 17,
pp. 12-22
Quine W.V. (1951), Te Two Dogmas of Empiricism, Philosophical Review, 60, pp. 20-43
Quine W.V. (1960), Word and Object, Cambridge: MIT Press
Quine W.V. (1972), Methodological Refections on Current Linguistic Teory, in Davidson & Harman
1972, pp. 442-454
Rey G. (2003a), Chomsky, Intentionality and a CRTT, in Antony & Hornstein 2003, pp. 105-139
Rey G. (2003b), Intentional Content and a Chomskian Linguistics, in A. Barber 2003, pp. 140-146
Searle J. (1990), Te Rediscovery of Mind, Cambridge: MIT Press
Ringle M. ed. (1978), (1978), Philosophical Perspectives on Artifcial Intelligence, New York: Humanities Press
Strawson G. (2003), Real Materialism, in Antony & Hornstein 2003, pp. 49-88
Wittgenstein L. (1953), Philosophical Investigations, Oxford: Blackwell
Wittgenstein, L. (1967), Remarks on the Foundations of Mathematics, eds. G.H. von Wright, R. Rhees &
G.E.M. Anscombe, 2
nd
ed., Oxford: Blackwell
Wright C. (1986), Teories of Meaning and Speakers Knowledge, in Wright 1986, pp. 204-238
Wright C. (1986), Realism, Meaning and Truth, Oxford: Blackwell
Wright C. (1989), Wittgensteins Rule Following Considerations and the Central Project of Teoretical
Linguistics, in George 1989, pp. 233-264
Received: April 8, 2005
Accepted: August 30, 2005

Oxford Brookes University
Harcourt Hill Campus
Oxford OX2 9AT
United Kingdom
[email protected]
24
EUJAP Vol. 1 No. 2 2005