Linguistic Competence as Developmental 

It will be a great error not to assume that bulk of speaking is in fact acting, doing something with language, speaking is, in other words, acting verbally i.e. effecting speech acts. Process Grammar begins with this assumption or insight and remains faithful to it throughout its other developments. The descriptions of the structural features such as the phonological, semantic and syntactic are descriptions that do not nullify the above observation. Consistent with this view, the general linguistic competence has to be seen as that which enables and makes in fact possible the effectations of speech acts. It is a specific capacity, perhaps peculiar only to members of Homo Sapiens, that facilitates the effectations of actions through the use of phonological strings by imposing some structures on them. 

 Though peculiar and specific, by virtue of the fact that it is nevertheless the doing of something, effecting actions, linguistic competence shares some features with what can be called 'a competence to do something' which is a general dispositional trait of all living things. And it consists in the abilities :- 

a) to generate an idea of action (in whatever rudimentary manner); 
b) to effect an action and bring about outcomes or change of some sort or other; 
c) to note the conditions under conditions under which actions succeed or fail and generate new actions that are better informed. 

 The effecting of actions, nothing of outcomes, reforming the actions and so forth constitute the learning aspects of living creatures in general. This learning dimension defines the universal behavioural disposition of all living creatures no matter how differentiated or undifferentiated they may be with respect to the scope and range of their abilities. 

 But linguistic competence is larger and more inclusive; it is a higher level of competence to do. It is not also simply a manifestation of new ability to effect actions with phonological strings through the articulation of vocal cords. What is peculiar to speaking is the ability to engage in a specific sort of interpersonal behaviour; a capacity for a new kind of interactional behaviour. The new dimension is more specifically cognitive and it can be stated as follows :- 

a1)  to generate intentions and formulate thoughts reflective of intentions and correspondingly to perceive thoughts and intentions when expressed linguistically. 

 When we replace (a) earlier with (a1) we have a characterisation of linguistic competence which brings out its peculiarity and simultaneously the manner in which it relates to the general competence to act. 

  The concept of 'thought' here needs to be clarified. The form of thought relates not alone to what the person intends but also to what he does. Thought then structures the intention but it structures it in form that allows it to be translated into action that may bring about what is intended. The thoughts when effected as actions emerge as speech acts i.e. stating something, ridiculing, criticizing, ordering, commanding, inviting and so forth. Such a person does something linguistically at all. 

 'Thought' is not simply then, an awareness though it is grounded on such things. For clearly there cannot be thoughts without some form of awareness or consciousness. We could say then that thoughts are transformed awareness, awareness restructured so that it can be acted out and hence changed into means for realizing an intention. 

 Now at least among human beings, as thoughts in general underlie all actions including the non-verbal, clearly in order to explain the peculiarities of linguistic competence, we have to also postulate a specific type of transformation of consciousness that would unfailingly relate thoughts thus formed to speech acts and not other. It is not also purely the interpersonal dimensions for clearly there are non-verbal acts that are interpersonal. Assaulting somebody physically is just as interpersonal as doing the same verbally. 

 Linguistic competence involves, among other things, restructuring consciousness into thoughts that also restricts to linguistic means of acting it out or realizing intentions. Such a restructuring then presupposes performance - the actual effectations of verbal actions. 

 Form the perspectives of Process Grammar then, linguistic competence incorporates in an essential manner the performative knowledge and skills understood here as the knowledge of verbal acts. Linguistic competence facilitates such performative knowledge and in turn is built upon it. There appears to be a cyclic relationship between the two. 

 But does it mean that linguistic competence cannot be described independent of knowledge of some specific language and its uses and hence not a description of anything universal? Certainly not. The specification of verbal acts in the structuring process presupposes only knowledge of acts that can be done only verbally and this is not tying down competence to any specific language. For clearly verbal acts are just as universal as the syntactic and so forth. What is empirical however, is the specific contextual, phonological and other surface structural features and how they relate to effectations of the verbal acts. The concept of linguistic competence as universal aspects of knowledge of language can be described, even in Process Grammar, independent of such empirical knowledge. Specifying how to do certain things linguistically may not be possible without recourse to some specific language. But the description of linguistic competence is not to be confused with this; it is different and it is a description of what enables a creature to act verbally in the first place.  

 Though universal, it is universal in a peculiar sense. It is not a description of something uniformly true of all speakers of language. For linguistic competence is not incompatible with a unique realization. Whether it is uniformly true of all or unique to a particular individual is a matter for empirical research and not part of the description of linguistic competence. For we could envisage the possibility of a unique kind of speech act as part of the competence of one person or a linguistic community and not that of other individuals or other communities. In what sense then it is universal? It is a universal in the sense that is structural description that can be used to test something among all individuals within or across different linguistic communities. More generally it provides a means for ascertaining the existence of a certain capacity among creatures including the human that are known or unknown that underlies what they do, how they behave and effect actions. 

 This linguistic competence or a particular description of it could have different components. Some aspects of it may be available among all speakers including the young, the mentally retarded, the abnormal, the underdeveloped and so forth. Some aspects cultured, or the communities spearheading new advances or an individual breaking new grounds in the totality of human knowledge. 

 It also must be noted that a particular description of linguistic competence is that of a linguist and therefore somewhat limited by his own competence towards intuiting the underlying. It may be possible that there are individuals who have competence beyond the intuitive powers of a particular linguist to intuit and describe. 

 One more important point to note is that linguistic competence this conceived is forever a description of open system. Linguistic competence is a growing, developing capacity and therefore in principle not axiomatisable. This point, which is crucial from the perspective of Process Grammar needs to be clarified a bit further. Let us distinguish between recursive creativity on the one hand and advansive creativity on the other. Recursive creativity is repeated application of what one can do under different contextual conditions can be exploited in innumerably different contexts. Advansive creativity is an absolute advance, it is a going beyond what anybody can do linguistically at that point in time. It is the creation of totally new forms of action and totally new forms of meaning that reveals itself in the transformation of language into new shapes syntactic, semantic and phonological. Such changes are deeper and slower but underlie the historical evolution of languages. 

 The concept of advansive can also be utilised to explain the acquisition of language by children. From the perspective of the child the acquisition of a language would require advansive creativity with the difference that is something that happens under the stipulations of established creative advances and as practiced by the adults. When we view language acquisition in this manner, clearly the child could not be considered a non-intelligent device with a certain precisely describable processing capacities. The child should be considered as an intelligent and rational agent who has within his power the capacity to break new grounds in what he can do. This brings us to the view that the behavioural manifestation of linguistic competence is conditioned by at least three important factors :- 

a) a certain kind of creature nature, a psychic maturity, a certain kind of developed subjectivity such as observable among the human beings but not the lower species; 
b) the exposure to exercises of accomplished linguistic competence such as that of parents, peers and so forth i.e. people who enter the psychological space of the individual; and 
c) the psychodynamics of the individual that determines the strength with which he endeavours to realize advansive creativity and assimilates the socially available competence. 

 The linguistic competence of an individual in its totality cannot be then talkad of independent of the development stage of an individual. It develops along with the development of the individual. The developmental aspects of a child ceases when he assimilates the adult competencies but then as an adult he could take it to new heights. In other words there could be no end to the development of the linguistic competence of an individual. 

 The child is certainly not a Language Acquisition Device with the social factors playing the rather trivial role of triggering to action a mechanism already there. The child is an intelligent creature whose intelligence is revealed in the acts the executes i.e. in the exercise of power he manifests. What distinguishes him from the lower creatures is the additional agency he has - the agency he has to execute acts verbally. Given this competence the furtherance of it is subject to at least the three factors listed above. However it is clear that a child is educatable in his linguistic competence however diffuse and complex the relevant strategies may be. The developmental and educatable dimensions the present concept of linguistic competence reveal also lead us to another insight into the nature of the psychic constitution of the creatures in general. 

   If competence increases with creative advances, the question arises as to why it has to be so. Each advance breaks into new grounds in knowledge and hence brings about new dimensions of consciousness. It is as if the psychic consciousness at any time has its own boundaries, boundaries which are enlarged with each advance made in competence. It is then as if the psyches are constrained deeply within in the forms of awareness they can generate, in the kinds of things they can in fact perceive. We are reminded here of the concept of anavam in Saiva Siddhanta, a concept so central to the whole system an on the basis of which it is to be distinguished from the other philosophical / psychological systems in India. The psyche is an anu by virtue of the fact that it is infected with anavam which is conceptualised as a primordial and deep seated delimiter or constraint that continually and impersoally blocks off or tends to block off the onset of consciousness or restricts any attempt to increase the capacity to generate new dimensions of consciousness. The anavam causes Darkness to prevail in the psychic consciousness, enveloping whatever consciousness it has and tending to eradicate whatever consciousness that has been attained. Advansive creativities are the kinds of events that register a reduction in the hold of anavam in the psychic constitution. 

 The development and educatable aspects revealed also raise another issue. If a child is educatable in linguistic competence then clearly it is something that can be learned. If one can learn then clearly one can also forget. Is this consistent with what we are intuitively aware of in the concept of linguistic competence? 

 Linguistic competence, once required, is not that easily forgotten. It is as if a permanent acquisition that is not lost through lapses in memory, failure to retrieve and so forth. It seems to survive even some forms of severe brain damage. 

 In the light of these generally known facts of linguistic competence, it may appear that the present view of linguistic competence is rather erroneous. 

 But this way may be a hasty conclusion for the concept of learning itself is rather ill understood. We are forced to note at least two strands in human learning - one that increases linguistic competence and another that brings in additional knowledge through the exercise of acquired competence. The bulk of the present studies in cognitive psychology is probably concerned with the later kind of learning where forgetting in commonplace. The former may be a deeper kind of learning where advance are not that easily made but once made not that easily forgotten either. It is a form of learning intimately related to the psychic constitution which once transformed into another remains so until forther transformations are effected. Such a form of learning then would presumably a permanent acquisition of the psyche for even when further transformed the hitherto acquired competence can remain with it, this kind of learning being integrative1. 

2.0. The Description of Linguistic Competence 

 In field work we transcribe utterances and use the transcriptions for various kinds of  linguistic studies. When our focus is phonological, we tend to overlook the fact that what are transcribed are verbal acts of individuals. (Cases where we request an individual to utter a word or recall the names of this and that and so forth are to be exempted.) When we proceed further with this rather refractory frame of mind we tend to forget that what we are analysing are representations of verbal acts. When we become conscious of this fact, we tend to enrich the representation with intonations, pitch, pause and so forth added. We also begin to place the representation in its proper context by specifying for example, the place a particular utterance occupies in the sequence of utterances, who speaks and to whom, and in some cases the various kinds of paralinguistic features that are seen as relevant. 

 Such a description, no matter how detailed and rich it may be, is nevertheless description only of verbal acts and not that of linguistic competence. Linguistic competence is not amenable to such surface descriptions which are descriptions of the observable. The competence we are talking about is that which underlies the effectations of verbal acts; something that is not observable but either perceivable inferentially or through observing criterialogical features that have already been established as constituting evidences for attributing the existence of the competence in question. This gives us an important insight as to the from of the description of linguistic competence. It has to be clearly an independent mode of description utilising elements that would furnish us an idea what elements and in what sense the surface description would constitute evidence for infering linguistic competence of a particular sort. This immediately takes us to the question of the elements that would serve as constitutents in the explication of the criteria. We must enquire into what must be specified so that what we know of the structural features of verbal acts would constitute evidence for ascertaining the existence of competence. 
  
 Let us recall the centrality of the notion of consciousness transformation or structuring / restructuring awareness into thoughts that would translate into verbal acts and no other in the concept of linguistic competence we are expounding here. This immediately makes it clear that whatever symbolic devices we evolve for describing linguistic competence, it must be capable of representing various formed into thoughts that would also specify the nature of the verbal acts. Since consciousness is always consciousness of this and that, it is also clear that what enters into the description of awareness are object-objects as they are in the psychological space of the individual or community. 

 The concept of psychological space is important here and some additional remarks are relevant. We have already noted that there are boundaries would also follow from the distinctions between recursive and advansive creativities. Linguistic competence is also developmental, something that can be augmented through facilitating appropriate kinds of learning. Such notions indicate that what is relevant for the description of linguistic competence is not the totality of objects, their categories and so forth but rather only those that have found a place in the consciousness of the individual or community in some form or other. The range and scope of object-consciousness that exists now constitutes the psychological space in question and certainly it is always bounded and open. It would appear to be always surrounded by the notion of the Unknown - that there are things yet to be known, categories yet to be realized and so forth. Learning then results in the reduction of the Unknown and hence the expansion of the psychological space2. 

 Everything in consciousness is an object and their categoricity determines the mode of consciousness. As it has already been explained in greater detail, we have to distinguish at least two distinct modes of object consciousness particularly with respect to concrete particulars. What we have in mind here is what Tolkappiyar termed 'porulotu punarac cuttunarvu' and derivately the 'porulotu punarum cuttunarvu'. These distinction first articulated in the ancient Tamil linguistic tradition has entered the wider Indian logical tradition under the name of 'niruvikarpa jnana' and 'cavikarpa jnana' i.e. 'non-conceptual awareness' and 'conceptual awareness'. It is not clear (to me) whether these distinctions apply equally to other categories of objects as well. 

 The other categories that have been already mentioned include Time, Space, Qualities, Actions Processes, Concepts and so forth. Perhaps there are more. The categoricity of objects remains part of the consciousness of objects and therefore inalienable from object-consciousness. The categorical distinction are not obliterated when thoughts are formed i.e. when a complex of object-consciousness are brought together to form a coherent whole. 

 But are the notions of forms of consciousness and categoricity of objects sufficient to represent thoughts that would also specify the speech act? 

 Certainly not and the additional objects that have to be listed for this purpose appear to be the case notions. For the cases specify roles of objects and since roles are defined in relation to the acts, it is likely then that the cases that are defined on the objects in the formulation of a thought also encode the specific verbal act that would succeed in translating it into action. When A commands B with 'Go and shut down', the roles that A defines on himself and on B, are not independent of each other and the act effected. Similarly when A states that P to B, another set of roles are defined on A and B. Such roles are interpersonal and arise in the course of interaction. They specify verbal act that relate the interacting agents in a particular episode. Normally we do not include such role definitions in the representation of thoughts generated but clearly they have to be if the relationship between action and thought is also taken to be relevant in the description of linguistic competence. And this can be enormously complicated and subtle as recent developments in discourse analysis would reveal. The differentiation and subtlety of person role definitions in interpersonal interactions directly relate to the linguistic competence of the person or persons. Where there are differences, such interaction become the primary means for the less developed to acquire the competence of the more developed.  

 We have said that thoughts are complex. They define interpersonal roles in the context of certain other consciousness. When A states that P to B, the interpersonal roles between A and B is defined with P as the ground. P is what is communicated and provides the ground for casting A as the one who states and B as the one who listen. P then is that which is acted upon in effecting the action of stating something. It also becomes that which is agreed upon, denied, questioned, doubted, assented to as possible and so forth. P as such has to be a generated form of awareness, something known to some and possibly unknown to others. For other-wise the need to communicate that will not arise. 

 The generated forms of awareness are product of actions of a particularly sort viz. cognitive acts. With such acts different category of objects are brought together and interrelated in a specific manner. This interrelating necessitates also definitions of roles on objects that are interrelated. When P is the awareness 'X struck Y with Z', the agentive role is defined on X, the Patient role on Y and the instrumental on Z. 

 It is clear then that there are two distinct types of cases; thought constitutive object roles and act constitutive person roles. Perhaps we can term as object-cases and person-cases. 

 In the Tamil linguistic tradition there has been some controversies with respect to the status of the vocative. Tolkappiyar has stated quite clearly that there would be eight cases altogether provided one also takes the vocative as a case. From the perspectives of Process Grammar outlined above, vocative and an enormous number such other cases would belong to the category of person-cases and not object-cases. They are constitutive of interpersonal actions and are the results of restructuring generated awareness that would relate to the mode of act execution. This area of linguistic enquiry appears to be the main thrust of recent developments in discourse analysis, oral and written. 

 Perhaps what we have termed person-orientations i.e., such notions as first, second and third persons in grammar, are aspects of this person-cases, a matter that remains to be further investigated. The developmental history of pronouns may also reveal important insights with respect to this question. 

 In the above discussions we have alluded to cognitive acts viz. person and object case indexing operations. In the earlier papers the necessity for such other intra-propositional acts as presupposing, object selection and so forth and inter-propositional acts such as conjunctive conjoining, disjunctive conjoining, quantifying and so forth have been argued. Perhaps there are many more such acts that remains to be identified and listed3. 

 The question now arises as to whether the categorical distinctions, we have noted and the listing of a variety of such categories of objects and acts are in themselves sufficient to describe linguistic competence that we have been talking about. Does it for example, allow the description of rules that could be seen as what underlines our knowledge of 'grammatical', 'ungrammatical' sentences and so forth which is part of the competence underlying successful effectations of verbal acts? 

 I think the specification of such rules is superfluous. The combinatorial possibilities and that of being a constituent in a complex that is generated through the exercise of certain cognitive acts is written into the specification of the entity and its categoricity. Specifying the entity 'conjunctive co-ordination' involves stating what category of things it can co-ordinate, under what conditions it would succeed or fail; how it relates to negation and so forth. The specification of 'a' as an inanimate concrete particular rules out indexing it with intelligent agency but not the ascription of certain perceptual qualities and so forth. 

 We appear to have knowledge of such characteristics of  elements for otherwise it would not be possible to generate thoughts that can be effected as actions successfully. However the descriptive specifications may be somewhat incomplete-limited possibly by the depth and accuracy of our own introspection. There could be modification and enrichment even in this direction. 

 If such are the minimally necessary constituents for the description of linguistic competence, then clearly it would be necessary to invent a calculus for describing competence. Such notions as non-conceptual and conceptual consciousness of objects and so forth would be more conveniently captured in the calculus rather than in ordinary language. The design of Process Calculus that has been described elsewhere4, is an attempt in this direction. 

 The specifications of the minimally necessary constituents as above also explain why the Predicate Calculus of Peono, Russel and Whitehead is grossly inadequate for this purpose. This calculus was invented initially for mathematics and later extended to be analysis of language by the British and other philiosophers. The modern mathematical logic is also founded upon the assumption of the validity of this calculus. Some linguistic theories of the recent past such as Generative Semantics and perhaps also Montague Grammar, Relational Grammar and so forth also appear to be inspired somewhat by this Calculus. But such attempts, to be extent they fail to modify and enrich the calculus sufficiently, so that thoughts as detailed here are representable, are doomed to fail - they cannot represent linguistic competence as competence to effect speech acts. 

 Consistent with our notion of open, developmental or learnable competence that allows for creative advances, it must also be stated that the Process Calculus cannot ever be a complete and closed axiom system. The calculus would forever allow for new additions in categories as well as within categories. 

 But how does Process Calculus explain the mechanisms of learning and thereby allow for the growth of competence? 

 It must be noted that when the calculus is supplemented with elements of the surface structure i.e. the means for representing the surface structural features of the execution of speech acts, we would have a symbolic system that would model the deep and the surface features of any speech act. It will not be exactly a simulation model if by that what is meant is a very detailed program that on being run in a computer would result in a similar kind of behaviour. However it would be a simulation model, if by that what is meant is a modeling of the critical skeletal features that are somehow intuitable as actually present in the behaviour that is being modelled. The model the Process Calculus provides is a clarification of the intuitive knowledge we already possess and hence something that can be used to gain a better insight into the structural peculiarities that underlie language behaviour. Where there is conflict between the insights gained and the intuitive knowledge we already have, we perceive the need to rebuild the model and hence reshape the calculus to facilitate a more accurate modelling. When linguistic competence has grown to new dimensions, clearly models that were adequate before would become inadequate now. The perception of such inadequacies is simultaneously perception of the improvement in competence that has taken place, whatever the causal factors underlying this growth. The necessitated search for and identification of the new elements that have to be provided for in the calculus, gives us an insight into the kind of learning that has taken place. On the basis of the perception when we further investigate the experiential or other factors that could have contributed to the emergence of the new elements in the competence, we would get an idea of the mechanisms of learning that underly the growth of linguistic competence. 

3.0 The Incorporated and the Represented in Linguistic Competence 

 D. N. S. Bhat (1983) has drawn attention to the need to distinguish between two species of knowledge: the incorporated and the represented. It is only knowledge that is individually represented in the brain that would be accessible for introspection. The knowledge incorporated would be part of the structuring principles of the device and hence, we can add in Wittgensteinian strain, something that can be shown in behaviour but cannot be stated. The incorporated knowledge would be inaccessible for introspection because it would not be individually represented. A person's grammatical competence is knowledge of this variety i.e. something unrepresented but incorporated accounting for the reason why it remain inaccessible for introspection. As he says 'It would be impossible for our innate knowledge of grammatical rules and elements to contain any representation as such. The rules and elements will have to be of the general nature, and therefore, they can only be incorporated into the structuring of our neutral network. This incorporation would take place as the neural network grows up as result of the activities of the genetic material' (1983:174). He also feels on the strength of these observations that the Transformational Grammar of Chomsky fails to describe either the introspectable or the non-introspectable grammatical knowledge of a speaker-hearer. (Ibid P. 178.) 

 While we can accept the distinction between the incorporated and represented forms of knowledge, the further identification of these with introspectable and non-introspectable seems to be grossly mistaken. It can be argued that there is no such a thing as 'non-introspectable grammatical knowledge' and if further this is taken to be the sense of 'innate knowledge' then knowledge of linguistic competence is certainly not of this sort i.e. not innate at all. Such knowledge is available for intuitive grasping and while it may be unconsciousness but by that very token something that can be brought up to reflective consciousness. 

 When we reexamine the Process Calculus keeping in mind these distinctions, it can be seen quite clearly that such things as concrete particulars in their conceptual and non-conceptual modes of awareness, qualities, processes, actions, differentiated time and space and so forth are elements that are individually represented in the brain as D. N. S. Bhat would put it. They are experiential and would not be there unless the individual had been exposed to the appropriate empirical stimuli. In contrast to these are the remaining elements - the cognitive acts - which because they are not tied to any particular experiential content, would not be represented in the brain as individual items. Such cognitive acts are capacities to do and therefore would be incorporated as part of the structure of the neural network. When we view them after subtracting away the psychological aspects of acts, then we would see them as processes the neural networks are capable of under suitable stimulus conditions. The calculus symbolises then two kinds of entities experientially derived knowledge of individual objects and processes which are dispositional or capacities that can be activated under suitable stimulus conditions. 

 Now there is an intimate link between such processes and the experiential knowledge of individual objects. For certainly the discriminative and analytical knowledge of the space-time-manifold is a product of the processes that are set into motion under various kinds of circumstances in which the presence of the stimuli is one and the need to act is another. Unless the processes are of a particular sort there could be no representation of knowledge of the space-time-manifold in terms of this and that and describable as so-and-so. This means an introspective awareness of the represented items should also reveal, in some specifiable manner, the nature of the processes that underly the production of such representation. Such revelations constitute our intuition of such processes and clearly constitute part of what we are aware of however vaguely it may be. While it may not be introspectable in the same sense as the individual items represented, it is nevertheless accessible for consciousness. It is vague and unclear and perhaps that is the main impetus for modelling the processes, an endeavour that may bring about the clarification, a gain in perspicuity and clarity that we seek. 

 The processing cap cities can be isolated and distinguished within the linguistic competence we have articulated. Let us call this component 'cognitive-act-competence'. Then clearly it would be seen that there could be no linguistic competence without this cognitive-act-competence. What establishes the capacity for speech acts is precisely this competence. And this competence must be available in whoever is capable of language. In other words it is a universal we could deduce also the presence and function of this competence. 

 What can be modeled and studied cannot be innate. Our knowledge of even this underlying cognitive-act-competence is then certainly not innate. It is to be wondered whether there could be any 'innate' forms of knowledge at all, whether it is self contradictory to describe one and the same thing as knowledge on the one hand and innate on the other. From the perspectives gained from Process Grammar we are inclined to the view that 'innate' as articulated by Chomsky is a vacuous notion, a chimera that has arisen from the fundamentally erroneous ideas Transformational Grammar has generated. It may be possible that what Chomsky was trying to articulate was this cognitive-act-competence for certainly it is that which underlies the syntactic structure of language. But he give a misleading description of it by falling to note that speaking is not simply a performance, an activation of the neural processes but rather that of doing something, effecting an action. 

 We are also obliged to explain in what sense Process Grammar is not a species of Cartesian Linguistic. Since acting is not to be equated with performance, it is not something that can be done by non-intelligent entities such as the neural processes we have been describing. We have to postulate a 'ghost' in the machine, as Ryle would describe it or a 'miniscule of a man' as Skinner would put it as the agent of the actions effected. The individual items represented in the brain, and the processes available in the network facilitate the realization of a range of actions. But since in each action there is activation of only a selected range of entities and processes, it is clear the agent must be 'ghost' distinct from these elements and processes with a capacity for selection and decision. The ghost must be then an entity capable of consciousness for clearly selection and evaluative decisions are impossible without this capacity. 

 But the ghost we are postulating, it must be emphasized, is certainly not the Cartesian Ghost. The Cartesian ghost is an unchanging intelligence, an entity that already knows all that needs to be known and therefore without any need to learn. The psyches we are postulating are the sorts of entities that are termed pasu in Saiva Siddhanta and which are distinguished carefully from Pathi, the unchanging and all knowing intelligence that is the real source and ground of all activities. 

 The agency of speech acts has led us to postulate a psyche as distinct from the neural processes and object representations and with a capacity for consciousness. But this in itself is not sufficient to conclude the reality of innumerable and distinct psyches, one perhaps in each machine. But we are led to this view on the grounds of person-roles that are seen as that which specify into what act and what speech act a particular thought would be translated into i.e. whether it will be commanding, requesting, questioning, stating and so forth. Such roles are indefinable unless the interaction is between independent psychic entities. 

 But how are we to make sense of the similarities and differences among these innumerable psyches? If there is perfect uniformity in behaviour, clearly we would fail to note any difference except perhaps the one of spatio-temporal location which is however inconsequential for establishing the reality of innumerable independent psyches. The absence of such uniformity and the observable differences in behaviour which reveal differences in needs, knowledge, competencies and so forth are the facts that lead us to an intuitive knowledge of the reality of innumerable independent psyches. 

 One of the things we have evoked here is the differences in linguistic competence among individuals. Clearly there is such a thing as language instruction where one is defined as the instructor and another as the learner. Learning is just as real as instruction. The psyche is the sort of entity that needs to be instructed to facilitate learning. 

 This learning process can be seen as being very complex involving at least two distinct strands. One will be enlarging the number of individual things and perhaps also the addition of new category of things within the cognitive processes that can in fact be activated through stimulating the neural network. Such learnings are assimilatory and are products of the recursive creativity we have mentioned earlier. But overall there is a limit to such assimilatory learning; there are boundaries to the range and scope to such acquisitions. But what are the factors underlying this perceived limit to knowledge? We can identify the following factors :- 

a) the exposure to the stimuli and the demands made by the social and environmental factors upon the psyches; 

b) the range and scope of cognitive processes that can in fact be activated; and 

c) a constitutional limitations inherent in the psyche itself which sets boundaries to the selection and decision and hence sections it can in fact effect. 

 The first of the above is that which facilitates assimilatory learning and therefore cannot be cited as the cause to the upper boundaries to this form of learning. It has to be either factor two or three and we have to investigate whether both or only one of them. 

 Now we have mentioned earlier that there is such a thing as advansive creativity which results in new processing capacities and along with it new dimensions of consciousness. It is as if a complex computer restructuring itself with additional, new and qualitatively different processing capacities. When advansive creative thrusts occur, what we have is perhaps additional neural structures being built into the nervous system. 

 Now the question arises: Where does the impetus for this transformation come from? It cannot be traced either to the first or second factor for both are non-intelligent and hence incapable of such creative advansive thrusts. We can also rule out Pathi as it is not in need of any such new competencies. We are left with the psyches as the only remaining candidate as the genuine source of this thrust. 

 But this again raises the question of why it has to be so? Why the need for the psyches to effect such advances or transcendences? There is apparently a pressure within the psyches to reconstitute themselves so that they become capable of new dimensions of consciousness. This means, as Meykandar (13th. century) has concluded, the psyches are delimited or constrained deeply within by some impersonal obstruent (called anavam) that continuously acts to reduce the range and scope of the consciousness of the psyches. Such delimited psyches are termed pacus and once we postulate such 'ghosts' in the machine, the learning or developmental dimensions of linguistic competence is well accounted for. The impetus the psyches generate is an impetus to remove this affliction within and it would probably originate when the limit to the existing capacities are perceived somehow. 

 The post Meykandar Saiva Siddhanta literature if full of details about the structures of pathi, pacu, and passam (to which belong the deep constraints and the rest) and the dynamic relations between them all to account for the differences and developmental growth of competencies to act. Since they are being discussed elsewhere, we shall refrain from discussing such matters here5. 

 This discussion, it is hoped, should make it abundantly clear that Process Grammar is not a species of Cartesian Linguistics. More appropriately perhaps it can be called 'Meykandarian Linguistics' in honour of Meykandar, certainly one of the most brilliant philosophers India has produced and who seems to have anticipated the concept of mind theory of knowledge implicit in Process Grammar. 

REFERENCES 

1.Bhat, D. N. S. 1983. Language Productivity and Grammatical Rules. IJDL vol. XII No. 2. 

2.Jantch, Eric 1976. Evolution: Self Realisation through Self-Transcendence in Evaluation and Consciousness: Human systems in Transition. (Ed.) Erish Jantch and Courad H. Waddington, U S A: Addison - Wesley. 

3.Muttharayan, K. Loganathan 1983 (a). The Tantric Theory of Learning. Saiva Siddhanta Vol. XVII. 

4.----, 1983 (b). Process Grammar: An Integrated Theory of Acts, Speech acts and Language structure IJDL Vol. 12 No. 2. 

5.----, 1984. Agamic Psychology and its Implications for child Development Studies. Asian Resional Conference on child and Adolescent Development university of Malaya.