Dėl tariamos nulinės jungties

ACTA LINGUISTICA LITHUANICA XLVIII (2003), 177–184 NOTES Remarks A TRIFLE The Samogitian dialect adjectives žiaudrus, žiudrus caught the attention of Steven Young, a Balticist at the University of Baltimore, in a very interesting article, “Lithuanian žiaurūs: žiudrus” (Acta Linguistica Lithuanica 47, 109–115). The sentences taken from pp. 453 and 458 of DŪŽ seem to call for a few comments: pākaulinės labai žõdrõs ā.uodeklas; vi.kšrės žõdrõs / pjā.usta liežuvė gi'vuol'ùn. The žõdrõs in the second sentence should be transposed not as žudrus, but as žiudrus: vikšris žiudrus pjāusto liežuvį gyvōliūon (dative singular; nominative gi'vuol'is | g'īvuolis; here, in the old traditional South Samogitian Varniai dialect, a non-etymological n has been added to the dative singular ending -u > -uo). It is not žudrus, but my clear proofreading error probably misled the author, although the headword is žiudrus, i.e. all the sentences are defined by the word žiudrus (the collection of texts, which took six years (1969–1975), had been greatly prolonged by prohibitions, complaints, and so on). In living speech this adjective is now becoming forgotten, and is used mostly by the elderly in the countryside; its roots are žiaud-, žiud-, so the only form is žiudrus, while scholars from distant regions who lack this dialect vocabulary may sometimes hesitate and begin citing a non-existent word, *žudrus. Vytautas Vitkauskas Lietuvių kalbos institutas P. Vileišio g. 5, 2055 Vilnius, Lietuva Received 2003 05 05 ON THE ALLEGED ZERO COPULA This text is primarily intended to discuss sentences such as: My brother is disabled; Persistent coughing is a symptom of a serious illness, and so on. In Grammar of Contemporary Lithuanian (Vilnius: Mokslas, 1994), they are regarded as nominal sentences conforming to a three-part model, since The weather is fine is considered to have a zero copula (p. 612). Before engaging in polemics on this narrow question, I will begin by mentioning two propositions, without which the polemic makes no sense, and which I will endeavor to follow strictly. The first proposition: Nothing has attributes. I also put it another way, jokingly: Nothing grows hair. It seems that the formulation of the proposition belongs to Immanuel Kant, but I do not remember this well; and since I do not base the truth of propositions on authority, it is enough for me that I consider it true. Logic does not operate directly on the word nothing, but its theory of classes admits a null, or empty, class (I note that it is called null after the sign used to denote it, and empty after its 178 | PASTABOS the scope of the meanings of signs, which logicians call the meaning of a name). An empty class has no elements, and only elements can have attributes. To illustrate this intuitively, I offer an example: when we demolish half a house, it is pointless to ask how many windows the now nonexistent half of the house has. If someone were to think that has can also be used to mean “had,” he would still have to admit that in this case he is using has indirectly, and that in a meaningful conversation it should be replaced with the direct form. Words used indirectly or imprecisely can yield only unclear or obviously false propositions, and every science must avoid them. For example, from the proposition that thought is transmitted through language follows the proposition that thoughts enter the air along with words and fly through it. Moreover, if we transmit thoughts, then none remain in the speaker’s mind (in the literal sense) after they have been spoken (and therefore it is no longer possible to ask what he said). This alone shows that it is pointless to set about creating a theory of any worth if it is based on logically unprocessed and nonstandardly used words. To illustrate intuitively what happens when my aforementioned proposition is disregarded, I offer this sentence: Although mermaids do not exist (nothing), they have beautiful breasts (attribute). From the standpoint of logic, this sentence is absurd, a logically contradictory proposition, and asserting such propositions is futile. I regard the explanation of the term zero copula, now favored by linguists, as a proposition of the same kind, because Although there is no copula (nothing), it has the meaning of the present tense of the indicative mood (attribute). It is impossible to escape logical absurdity by means of weighty words, sonorous rhetoric, or appeals to authority, because absurdity remains impervious to all these actions. One can attempt to escape absurdity in two ways. The first is to prove that the proposition nothing has attributes is not categorical and can therefore be disregarded. I regard this approach as entirely unpromising, because this is a proposition of logic and, to get rid of it, one would have to create a new logic of nothing. I regard only the second approach as promising: we need to carefully analyze where our thinking goes wrong, giving rise to absurd sentences. For this analysis I need a second proposition. It is this: Psychologism is a conglomerate of erroneous opinions that impede thought. I do not call it a theory, because the propositions of a theory are logically processed and systematized. The mildest form of psychologism is methodological psychologism. Its adherents acknowledge that the meanings of words and sentences exist objectively, but nevertheless maintain that they must be described in relation to psychological facts (thoughts). Naturally, not just any psyche will do in this case. For example, it is unsuitable when people do not fully understand what they are saying. What criteria should we use in seeking a suitable psyche? I rule out the psyche of linguists in advance, since in this article I will show that linguists, too, have differing views. Naturally, I also reject appeals to authority, leaving that matter solely to theology. I will note that not only our language but also our non-linguistic experience has regular correspondences in the psyche—even when chopping wood, we have to think about what we are doing. Are we really to describe all our activity in psychological terms? Why do other sciences, zoology for instance, not describe their empirical objects as correlates of the psyche? After all, they could, for example, include zero unicorns, zero Pegasuses, and so on in their classification of animals alongside wolves, moose, and elephants. One might object that language is a system of signs and therefore cannot be compared with animals, which are not signs. There is no consensus on the concept of a sign, but there are two different views: the empirical and the subjective. As an empiricist, I choose the first view and will set it out here in the words of logicians, which do not conflict with the customary understanding of a sign (and accord well with common sense). For them, a sign (they use the term symbol) is “an empirical object used to indicate other empirical or abstract objects”1 (at- 1Логический словарь: ДЕФОРТ, Москва: Мысль, 1994, 208 (sutrumpinimas ДЕФОРТ reiškia: Дедук-тивная формализация теорий. – V.Ž.). PASTABOS 179 I will note that the definition intuitively adheres to the postulate of empiricism). For example, the sign birch, as an empirical—that is, uttered—object, indicates another empirical object: a deciduous tree with white bark; while the uttered sign number indicates the abstract object “number,” one of the fundamental concepts of mathematics. For this reason, I regard the statement in the Grammar of Contemporary Lithuanian (p. 5) that “the structural units of language—signs—are two-sided” as a subjective statement that confuses thought, arising from imprecise and indirect speech. Speaking precisely and plainly, there is no sense in calling a sign two-sided. An empirical object, that is, a sign, and its meaning (or attribute, function) do not in turn constitute an even more genuine two-sided sign. If we adopt this way of speaking, which conflicts with common sense, we would have to say that what we see by the roadside are not traffic signs, but only the sides of traffic signs. Since language use itself goes hand in hand with common sense, can we really detach the theory of language from it and start speaking strangely? Ferdinand de Saussure was the first to call the linguistic sign two-sided, and many adopted this expression without giving it a second thought (perhaps because of his authority?). Saussure had a basis for characterizing the sign in this way: it seemed to him that expression and meaning were inseparable like the two sides of a sheet of paper, because he was an early psychologian and spoke of signs as psychological facts (at the same time, he also asserted the social nature of language, but did not reconcile this nature at all with the individual psyche). Once it is acknowledged that “the planes of expression and content of language are related to each other, but neither directly determines the other” (p. 5), not the slightest basis remains for using the concept of a two-sided sign in its literal sense—it is merely empty rhetoric, ballast left over from the past. The concept of the two-sided sign can, admittedly, be salvaged in one case: the sign must be transferred from objective reality into the realm of the psyche, and the transferred sign must be considered more important and more real—then one can say that, in the psyche, the concept of expression corresponds to the concept of meaning. Perhaps the authors of the grammar would not even object to this, since for them the structure of language is “a system existing in people’s consciousness” (p. 5). This is no longer methodological psychologism, but the most genuine substantive psychologism; it is simply unclear whether it is perceptual (linking language to images) or conceptual (linking it to concepts). If this view is accepted, ancient writings are in principle undecipherable. I see not the slightest need to debate this view, which in my opinion is hopelessly outdated and no longer salvageable by any arguments, now that Leon Zawadowski² has severely criticized it (I have summarized his views in Lithuanian, so they can readily be explored in greater detail³). Strictly speaking, what is in our heads is not the structure of language with its signs, but the ability to operate with it. But, insofar as we are individuals, we can operate only with part of that structure. The entire structure of language could exist in one’s head only in a single case: there would have to be an ideal language user, but no such person exists in reality. Thus, alongside subjective competence, the structure of language must exist objectively. One may ask how to understand this objectivity, and such a question is well-founded. Since language is very complex and many extraneous questions could therefore arise, I will illustrate the situation with a simpler example: the multiplication table. The question is: where do products exist, especially when nobody is multiplying (and products are, after all, the meanings of mathematical signs)? Mathematicians have long rejected psychological explanations, which is hardly surprising, since they are accustomed to thinking much more precisely and rigorously than linguists, whose thoughts, unfortunately, are often in other people’s heads. Mathematicians disagree about the nature of numbers’ existence, because numbers, like all abstract objects, are universals, and the debate about them is not yet settled (this is Bertrand Russell’s view). Anyone interested in the ontology of language must count— 2 ZAWADOWSKI, L. Lingwistyczna teoria języka, Warszawa: Państwowe Wydawnictwo Naukowe, 1966. 3 ŽULYS, V. L. Zavadovskio pažiūros į kalbą. Linguistics 24 (1), 1972, 111–119. 180 | PASTABOS …ty philosophers. But language can also be studied without resolving the ontological question. Physicists, too, do not actually know what gravity is, yet they study it successfully. By way of an aside, I will say that once the aforementioned definition of a sign is accepted, there is no room for zero signs (they are also precluded by Kant’s nothing account). Whoever began using the concept of a zero sign probably reasoned as follows: if an empirical object has meaning, then its absence also has meaning and is therefore a sign. It may be that the idea is derived by implication, but according to the postulate of the empirical sign it remains only an idea, not a sign—it requires an empirical object. Some people get confused about zero and say that zero is a sign—either a straight ellipse or a circle crossed diagonally. But that is notation in writing, not in natural language. In any case, of course, we have signs for nothing and related concepts, cf. the words nothing, emptiness, hole. Unfortunately, we have no contrary case. I regard the use of the concept of a zero sign as a (perhaps unacknowledged) variant of idealism (more specifically, subjective idealism), whereas the empirical sciences, among which I include linguistics, take a realist position. Once the existence of a language system in the minds of subjects has been rejected, one must adopt the postulate that language not only can, but must, be studied as an objective phenomenon, like other empirical objects. Such objective things are utterances, in other words, communicata. Communicata have meaning even apart from communication, because the communicative function of language is secondary—it rests on the representative function, the conventional depiction of some fragment of reality (were it not for this function, language would be pure play). A communicatum communicates only what it represents. It does not communicate the speaker’s subjective thought (all thoughts in the minds of subjects are subjective by virtue of the fact that they exist), but a certain fragment of reality. It is only on the basis of this fact that one can lie, mislead, or confuse (in such cases, thoughts clearly do not correspond to what is said). The sentence Black swans exist does not indicate that such a thought exists in the speaker’s mind (whether it does or not is of no concern to us; besides, how could you verify it, and can clear matters be based on unverifiable ones?). This sentence is an assertion about reality, and its truth can be verified by traveling to Australia and seeing for oneself; one can also rely on visual material or, for example, refer to biologists’ descriptions. On the basis of what has been said, I clearly distinguish two concepts: the meaning of a communicatum, which is objectively given, and the interpretation of that meaning—the manipulation of one’s own thoughts (I note that interpretation is understood here more narrowly, on the assumption that the meaning of a communicatum is understood without interpretation, according to relations that objectively exist in the language system). I maintain that only the meaning of the communicatum is the meaning of language. All interpretations belong not to language, but to thoughts “around language and its meaning.” I will illustrate this with an example: from the sentence Kepler died in poverty, by implication we can derive the true thought that Kepler lived (the so-called existential presupposition). If someone were to say that it is already implicit in the meaning of the sentence Kepler died in poverty, according to my account they would be wrong, because it lies not in the meaning, but in the sense. The search for senses, the continual reinterpretation, may well be a never-ending process, whereas the meaning of a sentence is “closed” with respect to sense. For example, for a linguist, one meaning of the word eye is the organ of sight, although one can think and say (using other words) a great deal about an eye. It seems to me that quite a few American linguists, unfortunately, do not distinguish the meaning of a communicatum from thoughts about that meaning. Now that I have clearly set out my position on the meaning of language, I will turn to a specific investigation, taking the sentence My brother is disabled. To avoid any misunderstanding, I will point out at once that this sentence does not belong to “speech” if that term is taken to include individual utterances, each tied to a specific speech act (it is easy to see that this sentence can be repeated, in which case we would have two “speech” utterances). But this sentence does not belong to PASTABOS | 181 language either, since only an abstract sentence schema can belong to it. It is “suspended” between a concrete utterance and an abstract language system. Thus, it is not possible to make do with just the concepts of language and “speech.” Leonas Zawadowskis quite reasonably stated that such sentences belong to a third entity—the potential language system—and I accept his account. Grammars give as examples not utterances concretely tied to a speech act, but facts of the potential language system (although they do indicate their authors). Understood in this way, the sentence I am examining is doubly abstracted from the speech act. First, we no longer include in the sentence the speakers, the specific location of the act, or other parameters. Second, the sentence itself does not refer to a speaker in its words either (it does not say, for example: I, Vladas Žulys, say that...). Linguists often claim that every sentence has modality (and perhaps is thereby connected with the speaker). From the standpoint of logic, the sentence under consideration is an assertoric proposition (Kant’s classification), and assertoric propositions have no modality (possibility, necessity)4. Thus, in this respect too, it is abstracted from the speaker (linguists who disagree have two options: either prove that the logicians are wrong, or reformulate the concept of modality and show that it can justifiably be used in its reformulated sense). The sentence My brother is disabled is completely abstracted from space (here I emphasize that I consistently use the method of abstraction). And since space is usually mentioned together with time (Kant even considered both these concepts a priori with respect to thought, but we now know that he was mistaken), I turn to what is essential for me: as a proposition, this sentence is also abstracted from time. In other words, it is a timeless sentence, and it is quite understandable why: it has no element indicating time. It exists in its meaning where all other abstractions exist (in the class of timeless concepts). There is nothing unusual about this: we are human beings, and our capacity for abstraction is so great that we readily construct propositions detached from both place and time. Of course, if one wishes, time can be indicated with other modifying words, for example, by saying My brother has been disabled since childhood, but, strictly speaking, this is not an expansion of the same sentence, but a different sentence. Using the method of abstraction, I have here characterized the meaning of the communicatum (of course, I have not explained the words my, brother, disabled, since that is trivial to me). This meaning is the meaning of language, and I consider the fact that time has been abstracted away here to be very important. Thus, I consider it unjustified to try to insert processless and timeless sentences into the paradigm of the verb to be as present-tense sentences (p. 291). In the paradigm, they should be set apart as timeless. But traditional grammar, including the Grammar of Contemporary Lithuanian, which has not shed psychologism, explains the sentence I am examining differently. It is usually said that the copula is omitted, left unsaid, or understood. I have already presented the absurdity that follows from treating the zero copula as the absence of a copula, so I will say no more about it. In scholarly language, I regard the words omitted, left unsaid, understood as imprecise and insufficiently clear. To make them usable more rigorously, I will qualify them with the words accidentally and necessarily. If a copula is “omitted” accidentally, this can be understood as the fact that it occurs in some sentences and not in others. Thus, we are dealing with a stochastic property of the system, and such properties are explained not univocally, but through their probability distribution laws5. Thus, one can describe only the frequency of copula use, and that is all. 4 See R. Plečkaitis’s account (Universal Lithuanian Encyclopedia 2, 2002, 69). 5 See Lithuanian Soviet Encyclopedia 10, 1983 (“Stochastic system” entry), 422. 182 | PASTABOS But if the copula is necessarily omitted, the omission must be explained unambiguously. In that case, omission is non-expression. That is, a sentence with a copula was present in thought, while one without it was realized. But that is not enough: we must also rely on the assumption that the sentence in thought is the real, “better,” original one, whereas the realized sentence is defective and secondary. Here we have reached the essence of psychologism: the true, more perfect system of language is in our heads (I have already mentioned that this is a variant of subjective idealism, which the empirical sciences reject). In fact, thinking and acting this way does not establish the objectively communicated meaning of a sentence; rather, whether consciously or without full awareness, it moves from the communicated meaning to its reinterpretation (I have already said that reinterpretations, however small, do not belong to linguistic meaning). The procedure is as follows: Mano brolis neįgalus, detached from time, is reinterpreted as Mano brolis yra neįgalus, which, because of yra, is linked to time. An abstract statement thereby becomes less abstract. Clearly, yra is supplied from an extraneous thought. Thus the objective meaning of a sentence is conflated with subjective thought (and this is precisely why absurd claims arise, but I leave their dissection to readers who are fixated on “zero copulas” and similar things). Functional linguistics (I do not know whether all of it consistently) regards such confusion as an error, wholly rejecting psychologism. For it, the sentences Mano brolis neįgalus and Mano brolis yra neįgalus are two sentences differing both formally and in meaning, and both are objective, although they have much in common. Agreeing with the view in Dabartinės lietuvių kalbos gramatikos (p. 5) that we should follow “the principle of dynamic synchrony,” I regard such sentences as competing: tenseless sentences, despite being very characteristic of the Indo-European type, are being displaced by tensed sentences. The present situation, and which type prevails, can be established only through a specific study. From the standpoint of logic, the sentences under consideration also differ in that the second sentence expresses the actual logical predicate6 and the predication operator (yra), indicating a special binary logical relation that cannot be explained in terms of other relations. Formally, this is represented as (x ⇐ X), where x is the subject, ⇐ the logical operator, and X the property (this interpretation is now considered more acceptable than representing predication by a propositional function, in which the predication operator7 is omitted). In the sentence Mano brolis neįgalus, there is no predication operator as a linear sign. Whether one should look for it elsewhere—in word order or intonation—is very unclear. But even in a sentence such as Dirbu, predication is clearly expressed without Aš, which syntacticians awkwardly posit as the subject; it is clear, too, that the subject is indicated by the ending -u, but the predicate itself is not very clear (one can only say that the subject and predicate are expressed syncretically). Interpreting Dirbu as Aš dirbu is possible only on psychologistic grounds. In the great majority of cases, Aš is used pleonastically, as a second marker of the subject. The texts show that it is used not necessarily, but contingently, and is therefore linked to the verb only by contingent relations (this must change the current understanding of coordination, but correcting it is not my concern). Logicians rightly point out that “in traditional logic the relation of predication is formalized only very approximately by the word yra in propositions such as (S yra P), where S is the subject of the proposition and P is a property possessed by proposition S”8. When logicians, in forming propositions such as (S yra P), abstract away from everything that is not a form of thought, they, strictly speaking, treat yra (at least intuitively and in practice) not as a copula in the full scope of the concept, but only as a copula, leaving aside what 6 Dėl termino žr. Логический словарь: ДЕФОРТ, Москва: Мысль, 1994, 191–192. 7 See ibid., 171–173. 8 Ibid., 171. PASTABOS | 183 does not belong to the copula. I think traditional linguistics failed to become aware of this process of abstraction and, in describing language, also treated yra only as a logical copula; from this, two further conclusions were drawn: 1) that būti, as a syntactic copula, has no independent lexical meaning, and therefore 2) in a sentence it “is used only as a marker of a certain tense and modality, i.e. it gives the property expressed by a nominal word the meanings of tense and mood” (p. 479). There is no reason to believe these claims as long as it remains unclear what the logicians abstracted away from. Let us reason as follows: (1) There was (intense) heat. The proposition is existential. From the standpoint of logic, such propositions have no predicate (Kant already said that būti is not a predicate, since it is not a property). They are now formalized with the existential quantifier ∃: (∃ x), but this quantifier actually indicates not existence but quantity (there is at least one, i.e. the set is nonempty). In the sentence under consideration, buvo has the meaning of existence, of being present, and this is the lexical meaning of the word (let us call this use of būti existential, or existential-being). From the standpoint of thought, existence is more important than objects and their properties—it is the foundation of objects in the sense that no empirical object is possible without being, and only such objects can have properties (it should now be clear what can ground the claim Nothing has properties). Karštis ‘heat’ is a fully substantivized word—a noun. (2) It was hot. This sentence also has existential-being, while karšta differs from karštis as a word that is not fully substantivized. (3) – hot is a property, incomplete both semantically and syntactically (this is indicated by the dash). Semantically, it is incomplete because it is not independent (like karštis), but is someone’s property. Syntactically, it is incomplete because the bearer of the property is absent. (4) – was hot has the same existential-being, indicating existence, as the logical foundation of the property karšta. The construction buvo karšta does not yet indicate in any way that buvo has any property of a copula. It acquires the function of a copula only in a sentence, for example: (5) The day was hot. That is, a syntactic copular meaning has been added to the lexical meaning of existence, but the lexical meaning has not thereby faded. It is itself secondary (here we may note in passing that the copula does not join the subject to the predicate, but attaches the predicate to the subject—as the earlier discussion of the predication operator shows). That buvo, by its very form and not by means of any property markers, indicates existence is evident from the comparison: (6) The pot seemed hot. 184 | PASTABOS (7) The pot was hot. The seemed in the first sentence needs to be compared not with the properties “conferred” by the noun, but directly with was. The former (seemed) indicates appearance (which is why it may perhaps be considered a word with modality), whereas the latter (was) indicates genuine existence (it has already been said that assertoric statements have no modality whatsoever). This analysis has shown what remained in natural language beyond the boundaries of logicians’ abstraction, and that is enough. 2003 04 20, at Easter Vladas Žulys Tako St. 1–7 LT-4324 Akademija, Kaunas District, Lithuania Received 2003 05 09