LANGUAGE: A WINDOW TO THE UNIVERSE[1]
I see little difference between a congenital abnormal heart that is still a heart and an ungrammatical sentence that can still be understood.
Language is a singularity. Only human beings have it. There are no known moments in the evolution of species that suggest that a more rudimentary grammar is present in other species, or in older versions of humans. Language is largely a mystery on which creationists and extraterrestrial invasion supporters may speculate on; some read into the angels spoken about in some unorthodox Bibles no other than extraterrestrial beings who, by mating with humans, may have added to their DNA the necessary information for the sudden birth of language in humans. But language remains a mystery whatever the answer to these questions may be. Language is a social art.
And while philosophers analyze language with a synchronic approach as an existing property of human beings as a class, linguists and other scientists go to the historical root of it, and they wonder about the biological aspects of its birth. Biology and evolution may one day find evidence, in some not yet found intermediate pre-human being, of the existence of significant peculiarities announcing the advent of language. But the mystery would remain even in such a case, as language demands a performer and a listener, a speaker and a hearer who understands and interprets what the other is saying. Since we are forced to imagine a moment of the history of humans with no language followed by another moment with language, we may speculate on the idea that a group of people have agreed on onomatopoeic basis to give certain names to certain objects and certain names to actions performed on those objects. So far so good, if we do not question, as I do, the perception of objects as I see them, already furnished with words. But even at the simplest levels language is not talking about you and me, the animal and the stone to kill them. It is much about external and internal events: the sentence ‘I am sad’ contains the mysterious role played by the copula and a reference to an internal state of the individual who utters it. I see only one possible explanation to it: certain words, in spite of their apparent conventionality, must contain a built in ability to evoke what they stand for, within their own culture, and this is possible only if language works on the basis of messenger particles like those they talk about in physics. Whatever its origins are, I see language as inseparable from the mental ability to think.
Time divided in present past and future would be unintelligible and inexpressible without language, so existing and not existing yet, or not existing anymore, states of affairs, facts, counterfactuals are unimaginable without language. With no language there would be no projects, programs, conscious intentions and memories perceived as such: everything would depend on instincts, and there would be no conscious actions, no objects and relations among them,[2] only mechanical behavior directed to immediate satisfaction of biological needs. Some of my claims, I am confident, can be supported by work in cognitive science. Although I see no signs in those works of indistinguishability of thought and language on the one side and language and reality on the other. They offer, in the steps of the tradition, those areas as distinct fields. Still, there are elements of unavoidable interdependence in their works that would support a different view.
In cognitive science, the simplest basic concepts are called ‘mental frames’: ‘apples’, ‘mothers’, ‘horses’, are general concepts somehow acquired by the mind and are connected with complex concepts referring to individuals; so the thought ‘my mother offered me an apple in one spring morning of 1970’ is the outcome of several mental frames that get connected. Concepts, and this is important, are blended independently of their status of referring to anything that already happened or is existing. They may be just imagined: this blending represents the cornerstone of analogical, metaphorical, or hypothetical thinking as we may find them in science. Art, religion, literature are the outcome of blending mental operations. Our mental life is much richer than the physical reality at any moment, and since mind is what allows us to manipulate the physical reality, everything is first in the mind and probably just there.
Having said that, the problem of the coagulation of the object, identity, and sameness are still a mystery for cognitive psychologists and computer scientists. I have something to say about it in the fourth chapter, ‘The psychoanalysis of the self’. But an idea is finding its way to the surface: it may well be that our notion of the individual entity or the particular, separated from the rest of the universe, the entity of which we predicate identity, has its roots in the general concept, the mental frame. The concept of an apple has a fixed, immutable meaning that gets transferred to the particulars of the world; hence the contrast between the immutable objects, the idea of an apple, and the changes that each apple may undergo: once we have assigned an object to a category,[3] it is an object of that category throughout its history, and language provides us with the means to talk about changes of the same object. The general idea of an object infuses its semantic essence into the individuals of this world. The initial monoblock is the model from which objects inherit their identity.
While geneticists, linguists, anthropologists, and psychologists may be interested in the origins of human language and its evolution, philosophers are traditionally occupied by more synchronic problems. They distinguish two main areas of the study of language: Syntax and Semantics.[4] Languages consist of words (the vocabulary or lexicon), rules of formation (the syntax), definitions of the words, or their meaning (covered by semantics). In addition to syntactic rules, there may be semantic rules that are charged with the task of preventing syntactically correct but meaningless sentences to generate. If we do not utter sentences like “quadruplicity drinks procrastinations”, it is perhaps because they are kept out of language by additional semantic rules. However, these rules are not easy to find, since as soon as you add a so called semantic rule to the syntax, all the metaphors that make use of some words in certain sentences perfectly meaningful such as “I don’t drink that”, used metaphorically to mean “I don’t believe that”, would be prevented by a rule concerning the verb ‘drink’ described as a two place relation of which one must be occupied by a mass term referring to liquids, for instance. As discussed later, there are two more notions that are of philosophical interest within the area of semantics: reference and truth.
But first, regarding syntax, the most serious work done on it in the last century was by Chomsky and his school. The main characteristic of Chomsky’s doctrine is that language is innate, probably a biological endowment according to his theory. There is a universal grammar to which all the spoken languages can be connected to, and finally language is infinitively generative as each individual may generate ever new sentences. I have found many of Chomsky’s claims very appealing, but I will not discuss syntax problems; my issues here are meaning, reference and partially truth.
These three notions are variously connected. One general way of presenting the connection could be that all sequences of words, in order to be promoted to the status of sentences of a language, must be meaningful, but not all of them may contain referring terms, and finally only a subclass of them may be said to be true or false. To sharply mark meaningful sentences is a problem. If ‘meaningful’ must be, in the ultimate analysis, explained in terms of ‘understanding’, many uttered sentences that are plainly understood by people with a low degree of knowledge of codified rules, both for grammatical and semantic reasons, are not recognized as sentences of the language they allegedly belong to: they are incorrect assemblages of words. And, a problem that has been usually ignored, some sentences deprived of meaning at a certain point in time may become meaningful some day: if somebody would have said in 1836 “George Washington telephoned Joseph Reed in 1776”, such a sentence would have made no sense, but today it is meaningful although most likely false. There is a diachronic dimension of meaning that can be hampered by excessive reduction to synchronic rules.[5]
(Chomsky’s followers, who devoted themselves to the study of semantics within the Chomskyan framework characterize words with semantic marks that are categorical tags in need of semantic characterization themselves; an unsatisfactory solution, I think.[6])
But we may divide theories of meaning into four categories. Following Davidson (1988), who distinguished between distal and proximal theories of meaning within the empiricist tradition, I have added two more subclasses of them that may exhaust the possible range of all the existing doctrines. By proximal theory of meaning, Davidson meant a notion of meaning that originates at the firing of sensory nerves, which is physical but internal (to the agent), and therefore gave rise to a proximal notion of meaning (it was suggested to him by Quine’s notion of ‘stimulus meaning’ in Word & Object, a notion fraught with serious problems). I think it is possible to conceive as a label for intensional, mental meanings, the proximal non physical notion of meaning which would fit the mentalist theories of meaning. By distal theory of meaning, on the other hand, he means a notion of meaning that originates at the physical source of the causal process, the real thing out there. Thus the meaning of “That is a rabbit” is generated by the rabbit itself; here too, we can add its abstract counterpart, the non physical distal theory of meaning, which can characterize the notion of meanings, from Socrates, (the concept), to Kant and Frege, all of whom are somehow responsible for the idea that there is an abstract reality out there that is the language, patrimony of the culture in which we live; the doctrines that can be so described may vary in details.
My suggestion is that a combination of those four categories exhausts the range of the alternative theories. To begin with, the notion of meaning I have in mind is proximal and abstract (pending an explanation of meaning in terms of neural networks, perhaps). So it is only temporarily intensional, so to speak. But, once we accept the distinction between an ante and a post litteram notion and the private-public dichotomy designed to replace the mental-physical dichotomy introduced previously, we can accept that mankind has constructed a public language that is external to each agent in much the same way as an external reality is constructed on the basis of our belief that everybody else perceives reality as we do. The abstract reality of a public language is, in the ultimate analysis, a mental construction we largely enjoy in our web of beliefs.
Rationalism versus Empircism[7]
It is a common characterization of the two schools of thought to say that the former emphasizes the role of innatism in human beings’ ability to learn and speak a language while the latter tries to explain language acquisition on a notion of a tabula rasa, a blank slate, where experience is all or almost all that is needed to explain language learning. Of course no empiricist is so naive to believe that there are no innate structures, as this claim would be easily belied by an overwhelming stock of arguments, among them the most important being the fact that other species cannot learn a human language; but the empiricist objects that the language faculty can be attributed to human’s superior rationality. For the rationalist it is evident that language is independent of other intellectual powers, one of them being the fact that is learned at a very early stage and without formal teaching. So there must be something peculiar to the brain of a human being, the biological foundation of language, yet to be discovered. I am inclined to believe that in order to understand language we have to transcend the individual and consider society as a larger unity to investigate. Each individual is endowed with some innate structures, but society will shape the mind of each individual exploiting those structures. I do not see, contrary to most philosophers, anything taking place during the learning process that is left out of a corpus of natural laws.
Natural language is not, in my view, a system rigidly used to communicate ideas and concepts. It is rather a set of ideas and concepts that inhabits our mind and constitute our essence, and we may need to communicate portions of them to others at different moments. Having structured our internal universe according to rules that are common to the whole of mankind, we are able to trigger in other minds representations that approximate our own. Understanding—and judging in the sense of accepting those representations as true or false, credible, or coherent or not with the preexisting universe that each of us is—will reinforce or weaken our belief in an external reality to share with others. Since mankind, at a very elementary level of perception, largely agrees, our social environment is a reinforcing factor all the way.
Conventionality and Normativity
The two notions are often used synonymously, and yet there are important differences. ‘Conventionality’ is a more general term with respect to ‘normativity’. But, having acknowledged that one term is inclusive of the other, I can put aside such a terminological problem and go to heart of the issue. I have introduced the notion of normativity in the second chapter as a property of those rules that can be violated at will, once one is willing to accept the consequences of one’s own actions; laws of nature cannot be violated nor circumvented. One more characteristic of conventionality is that it involves an agent, a mind, or a sentient.
The first thing that strikes us as conventional is the vocabulary of different languages. Since we tend to believe that things exist independently of language, it is a platitude that the word ‘dog’ is a convention if other languages have different symbols for the same thing (Italian ‘cane’, German ‘der Hund’, Spanish ‘perro’ and so on). But we must acknowledge the fact that for a speaker of any language the words that she learns have little conventionality at all. Even if we grant that reality is independent of language, those words are imposed on the mind of the speaker and they may be the diachronic outcome of an evolutionary process that has its origins in onomatopoeic sounds perhaps; these become morphemes of a language where the structure of that language arises from the impact between local external and internal factors (genetics), both contributing to the birth of a system that has an intrinsic hidden coherence and demands that each word is what it is, and not any arbitrary sound or symbol. This is a possibility neglected by the literature as several opposing views are held as platitudes. Among them is the unjustified claim that language exists apart from the agent as a patrimony of the culture, or an expression of rationality[8].
I see language as certainly cultural in nature, but not separable from the individuals who speak it. Language, so to speak, is the most dramatic element of the set of properties that obscure the boundary of the individual.
But the suggestion that language is governed by natural laws is itself a cyclopean enterprise to prove.
The problem remains that from the perspective of the individual speaker, the words are learned as rigid designators within that language. Aspects of conventionality are discovered at a higher level of culture. But the ascription of conventionality to the vocabulary tout court amounts to an ontological commitment towards abstract realities with no basis in scientific investigations.
In virtue of the fact that the first language sounds are learned by simple association, or fusion of auditive and visual perceptions (at least some of them), we may infer that there are no overt rules governing the mastering of the vocabulary; this may be a guess about what happens: a word like ‘Jessie’ is a sound that may be combined with other perceptual mnestic material like repeated appearances of the same individual, such as a doggie. Some innate mechanisms can compute what goes as essential into the visual perception of different appearances of a dog and extrapolate an ideal individual from the irrelevant content of the entire visual field and irrelevant changes from time to time. This abstract combination of different past appearances gains the status of a module once that mnestic material is combined with the sound ‘Jessie’ and allowed into the growing ontology by being marked with the status of ‘one’ granted by the compactness of the initial monoblock. The mind is able to pick it, create it, form an ideal object, a whole, from a changing environment. And, little by little, the world gets populated by individuals like Mamie, Daddy, and Jessie, where the sound, the word, is the communication factor, capable of fixing in the mind the existence of a physical—however abstract—reality. Each of these modules will go in as physical objects, although they are already abstract mental entities to the extent that Jessie may be recognized under peculiar circumstances, like dressing her up with clothes or other masking devices.
And, once we accept that even the notion of a single individual is an abstract entity, the general term is constructed on the same principles born from scattered (in space/time) perceptions. Both the singular and general term possess predictive ability, as they are applied to ever new perceptual experiences over time. If words are constructs born from an assemblage of perceptions of different natures, they cannot possibly be conventional arbitrary signs. Of course the conventionality of the name ‘Jessie’ is something that will be paid by the learner once she is on the verge of learning the general term ‘dog’, although a kind of intrinsicality of the name is going to stay with us for as long as we do not question it.
We may concede that words are conventional symbols from an external, educated perspective. We see this when we formulate syntactical rules for our own language, extrapolating perhaps from hidden latent rules in the mind, or we ascend to an interlinguistic level. From this it follows that conventionality and normativity are dependent on the perspective at work, since they don’t apply at the learning language level and they do when we rise above that level.
The mistake of considering language conventional arises from a generalization of a particular perspective. We cannot ignore that words and rules of formation are imposed on the speaker of any language from internal and external constraints. We can normativitize language—if I can be permitted this neologism—by constructing norms at will such as, “if you want to drink milk you need to say: ‘I want milk’”, or any equivalent sentence. These rules are no different from laws of nature contained in expressions as, “if you want to build a bridge you must follow such and such steps”.
Normativity overtly arises when we ascend to a higher level and talk about language itself: “What does ‘theopneustic’ mean?” Any correct answer to this question is normative somehow, and obviously the learning of a second language entails normativity at any step. Normativity is present in any secret code created to communicate undetected by intruders, or any artificial language like set theory or Esperanto. All these different perspectives may prompt a generalization that is ultimately wrong.
As I have said it was Wittgenstein who brought to philosophical attention the problem of normativity and Kripke in his work on Wittgenstein elaborated on the normativity of the meaning of ‘addition’ or the sign ‘+’. Here is a good presentation of Kripke’s paradox:
Suppose that you have never added numbers greater than 50 before. Further, suppose that you are asked to perform the computation '68 + 57'. Our natural inclination is that you will apply the addition function as you have before, and calculate that the correct answer is '125'. But now imagine that a bizarre skeptic comes along and argues:
(i) That there is no fact about your past usage of the addition function that determines '125' as the right answer.
(ii) That nothing justifies you in giving this answer rather than another.
After all, the skeptic reasons, by hypothesis you have never added numbers greater than 50 before. It is perfectly consistent with your previous use of 'plus' and that you actually meant the 'quus' function by them, defined as:
'x quus y' = x + y, if x, y < 50. = 5 otherwise.
The skeptic argues that there is no fact about you that determines that you ought to answer '125' rather than '5'. Your past usage of the addition function is susceptible to an infinite number of different quus-like interpretations. It appears that every new application of 'plus', rather than being governed by a strict, unambiguous rule, is actually a leap in the dark.[9]
This leads Kripke to claim that there is no private meaning, which follows from Wittgenstein’s claim that private rules make no sense. My objection to Kripke’s public language starts with an objection to the skeptic. I could escape his objection by telling him that, by the same token, I can doubt his own words and therefore not know what he really means. But the arithmetical problem has a solution that will lead us directly to another reflection on the notion of algorithms (rules).
You are an anthropologist and observe the behavior of members of a tribe in Africa, whose language is totally unknown to you. You grasp that one member is in charge for making purchases for the tribe. He usually goes to a market, 30 miles away, reaching it by camel. On one of these occasions you see that each individual, later guessed as wanting to buy something, approaches the buyer and gives him a little stone that he puts in his pocket. When no more individuals give him stones, he departs. When he reaches the market he addresses a man there, and puts on the table all the stones in his pocket. The seller starts picking up, let’s say, knives,[10] and for each knife he gives a stone back to the buyer; you are observing the transaction and you have counted 57 stones given back to the buyer.
The seller has run out of knives and says something to the buyer, goes away, and comes back with more knives. This time you count 68 knives piled up in addition to the 57 already piled and he gives 68 more stones back to the buyer. At this point, the buyer has no more stones. You conclude that he was commissioned to buy 125 knives and you observe the same individual, on different occasions, doing the same thing: the buyer and the seller both ignore arithmetic, but nonetheless they can carry out their transactions with a very rudimentary system of counting.[11]
You make a simple induction on the future and you surmise that stones stand for objects to buy, under the same circumstances described above. No algorithm is necessary, and yet there are counting, summing and subtracting if needed; the rules of arithmetic are in the mind of the anthropologist, not in that of the natives. Now if the skeptic questions you about your induction he cannot do it on the basis of an algorithm. You tell him that you expect stones to represent items, one by one. At this point the skeptic may reply that meaning of ‘stone’ and ‘one’ are at stake now, and you are faced with a masked case of indeterminacy of translation, in particular, inscrutability of reference! This is Quine’s old problem.
Furthermore, if the skeptic insists on seeing an algorithm in that rudimentary way of calculating a sum, and he focuses on the normative character connecting past and future uses of that system, you may question him in turn by asking in what sense learning to walk is not normative! And most generally in what sense negative charged particles observed in the past are expected to attract positive charged ones in the future! Since, if explicated, walking consists in moving one leg after another, keeping an erected posture and so on, and attraction among particles is a form of behavior that we attribute to non minded entities; and this is a hypothesis.[12] They are both forms of behavior. This is true, I believe, even before conceding any space to Quine’s analytical hypotheses or Davidson’s principle of charity.[13]
Normativity in the modern debate configures into complicated further distinctions. It is one thing to say that meaning is normative itself, and another to say that it causes norms (like in the case of analyticity mentioned in the second chapter). However, the study of normativity is not within the scope of this work. I mention it only because in my system normativity does not seem to arise at least at the level of learning a language; it is at most a post litteram notion, and as such it echoes natural science. Let me remind you the basic ideas of the theory I would support:
Words are the public aspects of ideas, but there is no sharp separation between words and ideas and if there are rules, they are written into the physical world.[14] Words contribute to the formation of ideas and they fuse with them, forming one single entity in the cognitive psychological reality of the speaker.[15]
Davidson[16] reminds us that Gilbert Ryle was supportive of ideas we may find in Graham Wallas. In The Art of Thought Wallas puts these words in the mouth of a young girl who was told to be sure of her meaning before she talked, “How can I know what I think till I see what I say?” Along the same lines, Robert Modwell is known to have said, “I would say that most good painters don’t know what they think until they paint it”. Davidson, in line with Ryle, reads in this idea that we know our mind exactly as we know about other minds, but rejects the tenet. As it often happens I see rather something different in these, according to me, indisputable points. I see the inseparability of words and thoughts. If, as I maintain, words are thoughts, then the girl was inadvertently claiming a hyperbolic platitude; she must see the words to know her own thoughts! There are no preverbal thoughts in the mind of a person endowed with language. A difficult thought may give the impression that you must find the right words to express it; I rather believe that if the thought is there, the making of its verbalization consists primarily in reordering its configuration; words are the necessary means of delivering the mental object and we need to find the verbal side, so to speak, of a thought in order to communicate primarily with ourselves.
Jean Paul Sartre[17] once said that literature is a mirror of the soul. I understand this claim as a contribution by literature to the reordering of thoughts: you read something and you know that those words are a good or excellent formulation of what you were incapable of saying (reordering thoughts may be a very challenging task). Finally, don’t we write to give systematization to our thoughts?
To believe in ideas does not necessarily amount to a commitment towards mentalism. As a matter of fact, although in need of future evidence, I believe that one possibility is that ideas, in the ultimate analysis, may consist in configurations in the neural network in spite of all problems posed by reduction. Ideas are still first person entities, although we may find out one day that they are supported by a physical substrate. The first language therefore, as far as meaning goes, does not entail normativity.[18]
A celebrated semantic theory rejects meanings as entities as they generate an infinite regress. It is not a wholly new idea. A theory of meaning that rejects meanings at the level of lexical items was made viable by the Stoics, then Jeremy Bentham and finally Quine. All held that we can talk of meaning only at the level of sentences or beyond. The theory put forward by Donald Davidson,[19] that meaning is fixed by knowing the truth of a sentence, is unsatisfactory, in my view, for at least three main reasons: 1) If meaning is what we know and understand in the presence of a symbol or a word uttered, we know that words have meaning, we understand the words, even of a syncategorematic type, in isolation, i.e., we know what to do with them and where to put them; ‘of’ ‘then’ ‘for’ and so on, are immediately recognized by English speakers who know their function within the language. 2) Truth is a more advanced notion with respect to meaning (whatever it is), and truth seems to be a function of meaning, not the other way around. You know if a sentence may be true or false (or neither) because you understand the meaning of the sentence, which in turn is understood via the understanding of its words in isolation.[20] 3) Finally, only a subclass of sentences can be said to be true or false, so the meaning of sentences not true or false is indefinite under such a theory.
The words ‘meaning’ and ‘…means’ appear in a variety of contexts. Consider the following examples. A) What does ‘GPR’ mean?, B) ‘When he becomes red in the face he means business’, ‘The clouds in the sky mean rain’, ‘A slippery road means an increase in the car’s accidents’ and so on. In (A) we are clearly referring to linguistic meaning, whereas in (B), the word ‘means’ stands for an underling intention by an agent, a physical cause, or the knowledge of a sign in nature that stands in turn for an event to follow. This is the usual explanation that you find in any introduction to the problem of meaning in the literature.
On the above track, consider now the following two questions:
a) “What does ‘GPR’ mean?”, b) “What is a GPR?”. One is a question about meaning and the other is a question about ontology, a question about an object in the world. In spite of this dramatic difference we may get the same answer to both question: “A GPR is an electronic gadget…”.
I see a car, a tree, a monkey, a GPR and I know what they are. I hear words like ‘car’ ‘tree’ ‘GPR’ and I know what they mean, having learned English as a second language. And yet both observable things and words are objects of perception, one visual and the other auditive. If I do not know what a GPR is, we have noticed above, I may ask, ‘What is this thing?’ and a more knowledgeable person nearby may answer, ‘It is a GPR a gadget, which…’ and the explanation of its function would follow. If I ask instead “What does ‘GPR’ mean?” in the absence of any GPR to the same person, I would get most likely the same answer; she would give us what we call the meaning.
It seems that we get the same explanation for two different questions, one referring to visible objects, the other to abstract, mental objects. Why in the first case do we say it is an explanation of what something is and in the second we say it is a definition when the answer to each is more or less the same?
It seems that the knowledge of what something is and the meaning of the word that labels that thing are the same thing. The difference may consist in one being a particular (‘what is this?) sometimes, and the other a general object (the idea in the mind that covers all the particulars in conjunction with other words that form descriptions of individual objects). But the particular depends on general idea.
Keep in mind that my point is that words are parts of objects, inextricably connected with them in the space of what I have called the individual universe. In the Lockean philosophy of language, words refer to or are names of ideas, and ideas are mental objects.
The distinction between the idea and the name of the idea is a later superimposition with no corresponding primary reality,[21] a mistake that Locke made as well: if he acknowledged that the word and what it refers to in the mind are the same thing, he would have not fallen into to the problem of his wavering perplexity about kinds of references (the idea and the object out there). The idea is a peculiar product of minds designed to capture an infinite class of particulars in appropriate contexts.
Since we posit objects of perceptions that are particulars, the idea is above all particulars in a suspended virtual space to incarnate real things. The existence of ideas, separated from the particulars, is justified by many facts: it is obviously a fantastic peculiarity of the human mind and it is a necessary creation to talk about the world. Words often come to our ears before we know what they mean. Similarly, we may know a thing by a different name, like when we know what an ‘unmarried man’ is and we do not yet know that ‘bachelor’ is synonymous with ‘unmarried man’. This problem is intra-cultural as well as intercultural. You may well tell a woman apart from men when you go to China by looking at women and men, but you may not know that 婦女 is a translation of ‘women’ in Chinese pronounced ‘fù nu’. In such a case you recognize the particular thanks to the general idea in your language. When you learn the word for ‘woman’ in Chinese, the conventionality of words gains support in spite of the fact that 婦女 may enter contexts completely alien to English.
Words have come to enjoy specific spaces in our minds and they have been slowly and artificially separated from their bodies in a surgical operation that is as useful as misleading due to a variety of reasons. One being the fact we are exposed to symbols, which are the only physical perceived entities. Their interpretations, decoding, that in my view are inseparable from meaning, seem to be another fact, just because this fact is an internal operation subconsciously performed.
“The wine is ON the table.” Any English speaker knows how to use that particle (on) in discourse even if she doesn’t know that it expresses a relation. Similarly, you may know how to use a knife in China without knowing the word for ‘knife’ in Chinese.
Things become more complicated with a syncategorematic word like ‘on’, ‘of’, and the like. In the English dictionary we may find different meanings of ‘of’ (it stands for different relations).[22] A natural way out is to resort to contextual definition, and this problem may have forced linguists to deny meaning to words in isolation.
But it is not something peculiar to language only. The use of this particle may be compared with the use of a knife for completely different uses: like cutting bread, killing your neighbor, unscrewing a bolt, pinpricking an abscess, and so on. And there is no telling what the official use of a knife must be.[23]
Words are parts of objects in the mind, parts of ideas[24]
The idea in the mind and the word are one entity at the outset. Only later may reflections of a cultural nature separate them into the (ideal) physical linguistic expression, and their properties, meaning and reference, artificially constructed. [Mr. Garofolo: I am not clear what you mean by this sentence.] Words are like plants with deep roots (ideas), and just as deep root plants have a high hydraulic conductivity ultimately capable of modifying the climate, the new content of a concept will percolate through the linguistic system modifying it in a holistic way: change irradiates language, correcting or adding new knowledge reflected in different or new word sequences that in turn will modify human behavior and thus the universe.
There may be people who speak perfectly well the language of the culture they are born in and never heard anything about dictionaries, meanings, or definitions. Language is structured according to certain necessary, mentally indispensable operations. In reality we have trees, plants, fish, dogs, wolves, human beings and so on. Sometimes the distinction between one species and the next is vague or arbitrary, but language needs general terms that ignore some differences in order to be learned. For, think of a perfect ideal language that adheres to reality up to the last detail, i.e., every significative difference is reflected in the language to the extent that you have a word for each thing and its changes. Its lexicon would be infinite and it could not be learned. As Hume pointed out, language is a fiction.
Many words are learned in connection with observations, but the idea in the mind is an abstraction that, once created, enjoys an independent life. The great power of the mind consists in generating ideas that can be used to refer to particulars out there once placed in the right linguistic context. We can construct sentences like, “That one is my dog” one using a word, the general term ‘dog’, that is the observable peak of the iceberg, the idea ‘dog’. As a consequence of the formation of a general idea, the properties of objects are abstracted and generalized in words that will be predicated of objects. We do not see properties in objects as entities, we mentally abstract them and then we see them. There are no visible colors free of objects to observe, they are a mental product.
The subject-predicate structure is a universal of language as it follows from the above considerations. Everything we encounter out there belongs to a class or category, and we want to say something about it; hence the S-P structure where S is the object we want to say something about, and P contains what we want to say about it. The copula ‘is’ is a mental entity, a mysterious connective generated by minds in every language.
As said, even syncategorematic terms are ideas. We have a mental representation of what words like ‘of’, ‘in’, ‘then’, ‘but’ stand for, although we must use contextual examples to explain them; they cannot be easily represented to refer to anything that can be visually perceived that would illustrate their general use.
However, these primitive reflections may induce a beginner to believe that nouns usually refer to either abstract or concrete objects, and syncategorematic particles do not.
There is no dog out there that resembles the general idea of
‘dog’[25]—although a particular dog may have been the triggering factor in learning the word—just as there is no containing relation that is more than a case of what we mean by the little word ‘in’, although we may have learned it in situations of a particular containment. Language may have been born to talk about what we call reality, but soon after, it has become independent of it and is better described as a tool to represent one’s own mind.
We accept abstract reality like numbers for their unquestionable usefulness, and many apparent ontological linguistic commitments are just devices to simplify our communication tools. We also accept ideal entities like ‘death’, ‘dance’, ‘speech’. There are no correspondent entities named by these words. ‘Death’, to a closer look, is better described as a process or event, and so is ‘dance’. People and animals die and dance. However, if you ask around, before recognizing the problem many people would immediately defend the thesis that ‘death’ exists like ‘dance’ and ‘speech’.
Try to introspectively analyze the problem: when you think what goes through your mind? Words, images, and what else?[26] Do you perceive any non-propositional or conceptual thought, naked thoughts of a non-verbal nature? The problem was dealt with by Plato and Aristotle, who made a distinction between discursive (dianoia = propositional) thinking and non-discursive thinking. According to them, or interpretations of their doctrines, the mind may contemplate what they called the concept. The more mystical Plotinus transformed some Aristotelian ideas into a more religious or mystical notion of contemplation of concepts. Whatever these philosophers had in mind, I find, in my consciousness, just images (memories of the past) or collages of images (let me call them potential realities, what cognitive psychologists call ‘blending’, sometimes images of the future like when we are about to meet with a person never seen before, and we have a mental representation of what she may look like according to what we know about her); or we find whole propositions (what we may say in a future encounter with that person, or what she may tell us, or something said in the past). Of course we may focus on single concepts, accompanied or not by images, but I hardly see any thought that can be said to be preverbal. Of course you may practice a kind of radical removal of any thinking activity from your mind and after that, label that experience as mystical or religious; you may focus on the notion of nothingness. But anything you can do with your mind, no matter how bizarre it may sound, will be explained, in the ultimate analysis, primarily to yourself, in terms of words.
One possible exception is the ‘I’, the ‘self’. It is not a concept and it is not a feeling. No matter how long you focus on it, it will eschew any attempt to characterize it in a way that is not metaphorical. But even the metaphors are just corollary to its essence, which is behind understanding and emotions. The ‘I’ seems a natural product of some mental operation that ignores multiplicity in favor of a non specifiable unity.
--------------------------------------------
Philosophy and linguistics have dealt primarily with the communication function of language, which can be disputed. Language is born to shape the universe and the universe is created upon an ontology of objects and their relations that we see through the lens of language. We talk for a variety of reasons: if displaying one’s own universe to oneself is a form of communication, then this is the main function of language. Granted, this is not what linguists and philosophers mean: they have in mind communication as a relation between two or more agents. Still, Wittgenstein’s theory that language is a picture of reality, if referred to internal reality, is a good intuition.
We not only represent to others our internal world, but also often expel portions of our internal world that we want to better look at. Ideas that we struggle to verbalize may just hide their verbalization to the mind, but the words to express them must be there or else there would be no ideas. We can imagine that the preverbal child has a non-verbalized notion of ‘motion’ and ‘steadiness’ to the extent that she may discriminate, in her mind, things that are not moving from things that are. This requires the knowledge of the notion of objects and things, and the difficulty arising here was dealt with in previous epistemological considerations.
Freud’s studies on schizophrenia show that patients tormented by nightmares would enjoy a temporary relief by simply reporting the nightmares to the psychoanalyst: but the ultimate purpose of it seems to be expulsion from one’s own mind of undesired states of affairs. It is a form of liberation that—according to Freud—would not be achieved if the narration hadn’t taken place.[27] But liberation from what? If narration can remove something from your mind, doesn’t it indicate that words have some bulkier consistency than what we usually attribute to them?
The ritual of confession as an institution of the Catholic Church is based on the same intuition: to report one’s own sin to another person is a form of liberation. But again, liberation from what? From facts inhabiting one’s own universe, of which words are just the public handle.
I have said that words are objects. The notion of object is a post litteram notion. Each word has a dimension in the mind, and that dimension is what we call the meaning. We combine words together according to certain rules, and sentences become tools to either cut and pick a portion of sense data or cut and pick a portion of inner life. In either case, we add a piece to the representation of our universe.
When we say, “This book is not the same as that”, we are construing a metaphysical relation that is mental in nature and has no echo in the alleged reality out there. When we say, “I have a pain in the neck”, we are attempting to break an opening in other minds, with a representation of our internal reality for a variety of purposes, of which communication is not necessarily the primary one. You could not make a person that is born with a congenital insensitivity to pain understand what pain is.
My understanding of language is not ‘symbol tethering’ (following Aaron Sloman and Jackie Chappell’s terminology instead of ‘symbol attachment’), which is a notion of meaning that spread from Ferdinand de Saussure and others who saw meaning as determined by structure. Mine is rather ‘symbol grounding’, which is determined by experience—provided that experience in turn is taken as a total internal process genetically and culturally determined.
Going back to the separation between form and meaning, one may wonder what it is due to. One reason is certainly the fact that if words elicit meaning, while words are always perceived in a physical context, their conductivity to meaning is a subconscious process. The other reason may be historical: in spite of the fact that thinkers like Levi Strauss, who showed that different myths may share the same meaning in virtue of their structure, great minds like David Hilbert’s have clearly cut form from meaning. In his axiomatization of geometry, the primitive concepts, point, line, plane, are deprived of their intuitive meaning and dealt with exclusively in terms of their relation. And, little by little, notions like truth, validity, proof, have been shown to depend solely on form. In the light of these new steps, it became clear that even Aristotelians syllogistics depends upon symbolic forms. Our modern computers have added an additional sense of confidence in the idea of form as they operate by the manipulation of symbols. Nonetheless, if there is no human mind at the desktop of a computer, there is no intelligibility.
Time and language
Not all languages have a clear distinction between present, past and future. The old biblical Jews had a distinction between things happened and things not happened yet, a qualitative distinction that was supposed to bridge past and future.[28] Some old languages seem to represent time by using spatial categories. Although we may speculate on this inextricability of space and time in ancient languages as a latent intuition of general relativity, our perception of time is still that of Newton, absolute time as an entity separated from space. Our western modern languages have all three tenses and they seem to capture our intuition of time, or is it the other way around? Has language has shaped our conception of time? Time, as we instinctively conceive it, is like a vector with a direction and the universe including us, the speakers, is at a precise point on the line along which the vector is moving. Everything before us on that line is the past and, whatever is in front is the future. We know the past as fixed and immutable and gone forever, while the future is characterized by potentiality.
To a closer look, however, the present seems to evaporate in the light of the fact that any temporal segment can be infinitely divided. So my simplest action or thought displays itself in a time that seems to include all three tenses. If I say, “please be silent”, my sentence is perceived as present, and yet the three words are temporally related in such a way that they belong to different times. When I utter ‘be’, ‘please’ already belongs to the past and ‘silent’ to the future. I could, of course, use a more accurate fragmentation, and divide words into distinct sounds, with each sound can being further temporally divided.
But even this rough distinction of words suffices to show the problem. Past and future have no meaning without a robust notion of present, because they must have enjoyed the status of being present or will enjoy the status of being present. Unfortunately, we lack an acceptable notion of the present.
In modern physics there is nothing to support our categorization of time, a momentary now, as opposed to the other two categories. While we firmly believe that we cannot go back in time, in physics there is nothing that prevents it. This is another serious problem that shows the mind does not capture reality if reality must be understood as something explained by natural science. Furthermore, from physics we learn that every object of our perception belongs somehow to the past, so we perceive, if at all, a reality that could not be there anymore, overtly evident when we talk about stars, less so when we talk about closer objects, as the light takes less time to reach our retina. Any possibility of construing a present is therefore subjective.
But physicists are misleading when they talk about space and time; first they want us to understand the inextricability of space and time, but then they speculate on them separately. Brian Greene, to mention just one among others, begins his story of space from Newton’s attempt to show the existence of absolute space with his famous example of the spinning water in the bucket.[29] Mach’s reflections on the problem that allegedly influenced Einstein’s relativity follows, where the example of yourself in empty space serving the purpose of understanding the idea in subjective terms. Suppose you are placed in empty space away from any point of reference (matter in the form of stars or whatever may be gauged for its mass). Suppose furthermore that someone comes and starts spinning you in such an empty space. Here, the spinning loses any meaning. There is no spinning at all in an empty universe.
This neglects the fact that if someone came to spin you, he would also fill the space with his mass!!! This makes us wonder if physicists are not forgetting the effect of this presence—the space is not empty anymore! Furthermore, since spinning is just a case of motion, in this empty universe you shouldn’t be able to perform any motion since, according to Mach, in an empty universe, motion does not exist; one cannot feel it. One wonders what happens to you if you lift an arm or move a leg, since you are moving the arm or the leg with respect to the other arm or leg…or the entire body!
Time is in no better shape. They provide no definition of time and, once again, they start by claiming that in an empty universe, time is meaningless, or non-existent, as they see time as a measure of change or flow; one physicist said that time is what prevents everything from happening at once. It is hard to imagine how everything could happen at once: take a piece of music and the aging of a butterfly, how could they ever happen at once?
They seem to believe that there is an objective and subjective time. The objective time is not supported so much by mundane events, as they are relative to agents, but rather by the universe in its entirety. Two scientists, one in our galaxy and another in the next, would reach the same conclusions about the age of the universe and its intermediate stages. These remarks about time seem the ultimate evidence that there are notions peculiar to the mind that, however useful or even necessary to our understanding of reality, are not grounded into reality. All attempts to define time are doomed to be tautologies, for look at my explanation that could have been welcomed by Einstein: reality is like a building that you are walking around for the first time. In order to know everything about the building you must go around it, at least for the entire perimeter, and that ‘takes… time(?)’ and ‘cannot happen at once(?)’; what you see at any instant is the present, what you have seen, the past, and what you will see, the future? They are all tautologies and there is no escape from them.
Final considerations on Language and reality
My opinions on language, expressed so far, may be read as supportive of a radical logocentric thesis. I have no objections to such claims, except perhaps that I do believe in the external reality as everybody else and that I want to try to evaluate the incidence and the role of language in a situation where reality is overwhelmingly present. Consider the human enterprise of landing on the moon in on July 20, 1969. I have no doubts that some men boarded a spacecraft named Apollo11 and landed on a satellite existing out there that we call ‘the moon’. The success of such an enterprise was made possible however by a political background involving decisions, the study of materials, and other engineering problems together with computations of gravitational problems, speed, motions of the celestial bodies involved, study of the human resistance to unknown, up to that point, reactions to novel physical conditions and so on. All these preliminaries were collected on a two dimensional reality in the form of notes, instructions, and information, and taught to all the people involved in the enterprise. These data were the backbone of every second of the trip. Moreover, we know about it because we have collected not only all the above information, but also a description of the entire trip, complete with images[30] and comments. If we did not have a language, those images would be lost in the obscurity of an indistinct flux of meaningless photons. Is there a trip to the moon for any living being not endowed with language? I think not, and we could expand on this by attributing to ourselves the impossibility of seeing important aspects of reality—our language is constructed by our system of perception and our perception is not sophisticated enough to grasp those detailed elements of reality; we are already dramatically perplexed at the idea of particles that are also waves!
The Mississippi Rivers is about 2, 340 miles long, the second longest river in the United States. Mount Everest is 8,848 meters or 29,029 ft high. New York is the most densely populated major city in the United States, with an estimated 8,274,527 people occupying just under 305 square miles (790 km2). We learn early in school a large number of data like these. For a middle school class to learn only one of these data by direct experience would take a lifelong commitment.
Can you imagine a class of students going to measure the Mississippi river just to make sure that books are not fooling them? Today we can use our bank accounts by sitting in our bedrooms, talk to a Chinese or Philippine individual on the phone or internet, where we can study, teach, or even fall in love by not going anywhere; much of reality can be collected with no major loss into a bidimensional reality. They are perfecting techniques to connect minds with computers and, without touching the keyboard, seriously handicapped people can communicate. As science evolves, the universe shrinks and the hard reality out there becomes more and more superfluous, let’s say; and language is essential to these achievements.[31]
Last word on Rules and Laws
You cannot jump out of the window and fly, but you can utter meaningless expressions. Is this the main difference between rules and laws? A meaningless expression is not a representation of anything, but you can utter it in the action of reporting a dream to a friend or a psychologist, just as you can dream of flying and tell the dream to someone. In both cases there is some real connotation to them: dreams are free of grammatical rules and physical science laws. They obey different natural constraints, and in the context of a dream they may acquire meaning. So, just as you cannot fly in the real universe, you cannot utter ungrammatical meaningless expressions and hope to be understood. I see differences only in the quality of the constraints. Of course a treatment of this problem would require more sophisticated and deeper considerations, which I do not intend to pursue here. I have already hinted at the problem within the section about normativity.
[1] This chapter has no claim to be a complete presentation of a theory of language. It is rather a discussion on some basic principles that, according to me, should eventually develop in a theory of language.
[2] Studies on animals that reveal apparent creative behavior, like that of tricking a potential predator into believing something false, are couched in terms of individuals and relations among them, but of course it is a projection of our own vision of the world; we have no way of knowing how an animal sees the world and what kind of identity characterizes their environment.
[3] It is an illusion to believe that particulars of this world are not abstracted entities, as reality changes and only the mind fixes objects over time. Individuals too are mental constructs.
[5] I have included the notion of ‘definition’ within the area of semantics, although from a philosophical point of view you wouldn’t find a doctrine taking definitions seriously as meanings anymore.
[6] The word ‘girl’ would be semantically characterized by tags like ‘living’, ‘human’, ‘feminine’, ‘young’, and so on. Of course such an approach would generate an infinite class of tags when it comes to give the meaning of the tags. To consider them primitive is a way of putting an arbitrary limit to our semantic questions or, if you like, a way of moving the entire semantic problem to these primitives.
[7] Once again this chapter is a collection of fragments of reflections on the problem of language over a number of years and there is no pretension of a formulation of a complete theory of language. Entrenched with aphorisms rather than arguments it is directed to the reader who shares in advance some common ground or else to the reader who finds these reflections appealing to her mental framework.
[8] As I have said, meta notions such as linguistic conventionality can be granted the status of post-litteram notions.
[9] This is Kripke’s paradox as it is explained in Wikipedia under the title ‘Wittgenstein on Rules and Private Language’. This page has a good version of Kripke’s problem.
[10] I have simplified the story by not including different items to purchase, that can be visually presented or simply mentioned by the members and different stones, signifying different objects, perhaps carried in separate bags, each containing stones referring to particular items, let’s say knives, arrows, shoes and so on.
[11] It may well be that the rudimental arithmetic I am describing is the first step to a more general and extended usage of the built in recursive motor machine present in animals too: walking, eating, calculating distances that allow the predator to land on the body’s weakest spot of the animal he intends to victimize, are all instinctive recursive abilities. What makes recursion typically human may be its transferring from this behavioral setting to more abstract forms of investigating reality. But just as we could not question an animal about how it knew that it had to jump at a precise moment, we cannot question the African native about his computation. It is still based on an instinct.
[12] Physical events are explained in terms of a silent order and why, for instance, negative and positive particles attract is not a question to be asked, it is ultimately a fact; how correct is this assumption is debatable. I am not claiming here to have solved more than Krepke’s paradox, I am well aware that by observing behavior we cannot get to some mental states, at least at the present knowledge of the mental as a physical reality.
[13] These two principles are quite similar, but for a better understanding of them there is a large literature on them that can be consulted: they are mainly constructed on the assumption that other minds are similar to ours.
[15] I exchanged emails with an Artificial Intelligence scientist, Giuseppe Trautteur, who seems to be sympathetic to my idea as he believes that my use of the notion of ‘mind’ as a container of ideas is misleading. Also, he adds that only numbers can exist independently of consciousness; they do not need interpretations. This is his claim as I understand it. Does he mean that arithmetic, for instance, is autonomous, a self explanatory that words are indistinguishable from ideas. He calls this position the reduction to consciousness system where syntax suffices to fix the semantics as well? If this is what he means, I wonder whether some AI scientists are not under some kind of illusion. The manipulation of symbolic forms in math may well provide an exhaustive, self-contained set of rules and relations among its entities, the numbers. But, when it comes to applications, math needs interpretations of unity and multiplicity posited, and these are ontological issues. One example could be the following: the atom is one entity, but it contains subatomic particles, which constitute multiplicity. How can anybody, say an extraterrestrial, understand the application of math to this aspect of physics if she did not have an ontology like ours, with unity and disintegration of unity to represent internal multiplicity? Numbers are only apparently immune to understanding.
[20] I use the notion of meaning for its explanatory power, although I do not believe in meaning as it is understood in the tradition. This is a problem with many of my terms used just to bring the problem to the attention of the reader, the problem I am trying to clarify according to my point of view.
[21] For explanatory reasons, we can allow a distinction between primary reality and secondary or constructed reality, where the first is also the reality investigated by science and the second is what the mind has added to it, although the distinction is fictitious.
[22] 1) Used to indicate distance or direction from, separation, deprivation, etc.: within a mile of the church; south of Omaha; to be robbed of one's money. 2) Used to indicate derivation, origin, or source: a man of good family; the plays of Shakespeare; a piece of cake. 3) Used to indicate cause, motive, occasion, or reason: to die of hunger. 4) Used to indicate material, component parts, substance, or contents: a dress of silk; an apartment of three rooms; a book of poems; a package of cheese. 5) Used to indicate apposition or identity: Is that idiot of a salesman calling again? 6) Used to indicate inclusion in a number, class, or whole: one of us.
[23] The problem with syncategorematic words becomes even more dramatic in translation. In Italian they say 1) ‘Io vado a Napoli’ and 2) ‘Io vivo a Napoli’ whose correct translations are 1) ‘I go to Naples’ and ‘I live in Naples’. From which we may conclude that ‘a’ translates to both ‘in’ and ‘to’ in English; but this is a tentative, incorrect, if useful, way of putting things. As in Italian we may say ‘Io sto in casa’, meaning ‘I am at home’. This problem is persistent: it does not affect only syncategorematic words, but also verbs and nouns. An example of verbs is the translation of ‘have’ and ‘be’ respectively with ‘avere’ and ‘essere’, but I may construe a grammar perfectly compatible with the data and switch the translation. The noun ‘pilot’ only up to a point enters the same contexts into the two languages. These are considerations that opened the path to indeterminacy of translation in Quine. My proposal bans meaning altogether at every level. Semantics becomes a science of mental ontologies relative to the culture.
[24] While semantics is the target of my criticism and my innovative proposal, syntax is not affected by it. However, I do believe that syntactic rules are dependent on semantics, one of the reasons being the fact that the sentence is structured according to ontological needs. In a rational reconstruction, we may separate the two fields and deal with them separately. Grammars and dictionaries are, in the ultimate analysis, general approximation that are designed to meet a large consensus.
[25] Nor can our ideas of particulars be said to adhere to what they refer.
[26] This reinforces my conviction that visual and other perceptions are intimately connected with language: syntax and semantics are not separated at the psychological level.
[30] Images are a byproduct of the impact between language and perception and as such they are part of language.
[31] To predict further elimination of the need of living in the real dimension is a hopeless enterprise, we cannot even imagine how far it can go.
No comments:
Post a Comment