Description of Language Design

Even more broadly, Martin Joos defines all language as "code" because it is both symbolic and organized. But it will serve our purpose here merely to make a distinction between "language" and a "code." By "language" we shall mean those organically developed systems, whether spoken or scribed, by which humans transmit messages; but the word "cipher," or "code," will be used to mean any invented, self-consistent system, whereby one set of symbols may be transformed into another for certain special stated purposes (i.e., Morse code which converts printed letters into dots and dashes). (Cherry 1977: 93-94)

This is the quote that more than a year ago sparked my interest in Martin Joos. Mainly because this resembles of of Juri Lotman's famous statements, namely, that LANGUAGE = CODE + HISTORY. Or, in Colin Cherry's terms, language is also a code as Martin Joos defines it, but it has developed organically - it has a history (as opposed to a cipher, which is invented and self-consistent). I contacted professor Irina Avramets, a former student of Juri Lotman, with this association. She confirmed that absolute certainty that the Russian translation of К. Черри's Человек и информация : (критика и обзор) was studied by Lotman extensively the very year it appeared, in 1972. Irina Avramets was also kind enough to provide me with the following excerpt by Lotman in Russian:

В основе этих рассуждений — абстракция, предполагающая полную иден­тичность передающего и принимающего, которая переносится на языковую реальность. Однако абстрактная модель коммуникации подразумевает не только пользование одним и тем же кодом, но и одинаковый объем памяти у передающего и принимающего. Фактически подмена термина «язык» тер­мином «код» совсем не так безопасна, как кажется. Термин «код» несет представление о структуре только что созданной, искусственной и введенной мгновенной договоренностью. Код не подразумевает истории, то есть пси­хологически он ориентирует нас на искусственный язык, который и предпо­лагается идеальной моделью языка вообще. «Язык» же бессознательно вы­зывает у нас представление об исторической протяженности существования. Язык — это код плюс его история.

As far as I can make out with Google Translate, a part of it says that it is not as safe to substitute the notion of "language" with the notion of "code" because a code is a structure "that you just created" artificially while a language has a historical existence. The bold sentence is translatable even by my meager capabilities: "Language - it is code plus its history."

So that's why Martin Joos caught my attention. The case seemed even more interesting due to the fact that there are two issues of the Journal of the Acoustical Society of America in Tartu. One is volume 65 issue 1 from 1979 and the other is, well, volume 22 issue 6 from 1950. Before Irina Avramets confirmed that Человек и информация was in Lotman's library, I suspected that he might have read this exact issue that is so neatly present in Tartu. It turned out that Lotman didn't enjoy reading in English. His first foreign language was German and second French. Also, this issue of the journal was donated to our library from Arvo Eek's collection who-knows-when. It kept nagging me only because I was interested in reading this article, "Description of Language Design", but couldn't, because the issue has been misplaced or has gone missing.
In the meantime I have read one philological article by Martin Joos in which he refers to the nonverbal in literature as "details of the life and world that nobody escapes from and yet nobody is aware of" (Joos 1972: 262). Later yet I learned that Edward T. Hall's proxemics was inspired in large part by Martin Joos's 1962 book The Five Clocks, which was his take on communication styles (and, now, one of the books that I desperately wish to read). All in all, Martin Joos is an interesting fella. He seems to be unpopular despite the profound influence he has had on my favourite varieties of semiotics. I'm reading this article now because I've familiarized myself with Charles Morris's strict definitions of language and in Elin Sütiste's course I'm also reading Roman Jakobson's writings on language, so it seems like a perfect time to finally take this on and try to find out what Martin Joos actually thought of the design of language. I can read it now thanks to the Paper Fairy on reddit.

Joos, Martin 1950. Description of Language Design. Journal of the Acoustical Society of America 22(6): 701-708.

Physicists describe speech with continuous mathematics, such as Fourier analysis or the autocorrelation function. Linguists describe language instead, using a discontinuous or discrete mathematics called "linguistics." The nature of this odd calculus is outlined and justified here. (Joos 1950: 701)

This is something that one probably cannot find in Lotman: that linguistics is a discrete mathematics. Instead of discrete systems of signs, here it seems that the method of studying systems of signs is discrete. It also feels somehow appropriate that Lotman's theory of discrete and continuous systems have become so complex and convoluted that they can also be designated as an odd calculus.

The word "design" may be interpreted in at least two senses without violating engineering usage. First, there is work of the engineer who is designing something, for example a telephone system. Second, there is the finished something, such as that same telephone system, considered now as something to be analyzed and described. Our description of it, then, can also be called its "design." It is in this last sense that the word "design" has to be understood when we are talking about the design of language and how it is analyzed for description. (Joos 1950: 701)

"Design" is in this sense just as problematic as the term "rule". Both have a prescriptive and a descriptive dimension.

We can allow other people - telephone engineers or sociologists, for example - to speak artistically, imprecisely, about language. But as linguists we lay upon ourselves the condition that we must speak precisely about language or not at all. (Joos 1950: 702)

This is partly the reason I'm so annoyed when some semioticians speak very vaguely about sign-phenomena.

The mathematical description for reality has been most illuminatingly called a "map." Now when holds a map (in the ordinary sense) in one's hands, one may say, "I feel that there is no way to prove logically that it is not instead a map of a piece of Australia or perhaps of some imaginary Treasure Island. Undeterred by the impossibility of logical justification, explorers use maps, scientists use mathematics, and linguists use the descriptive technique called "linguistics." All three have the same attitude toward the map. One proceeds across the terrain and simultaneously traces a line across the map; one notes discrepancies between the map and the direct apprehension of the real world, until the discrepancies seem to form a pattern themselves; then one corrects the map and starts all over again. All this is intuitive behavior, and logically unjustifiable. Nor does it need justification. The place for logic is inside the map, not between the map and the real world. (Joos 1950: 702)

Something for the semiotician of space.

Ordinary mathematical techniques fall mostly into two classes, the continuous (e.g. the infinitesimal calculus) and the discrete or continuous (e.g. finite group theory). Now it will turn out that the mathematics called "linguistics" belongs in the second class. It does not even make any compromise with continuity, as does statistics, or infinite-group theory. Linguistics is a quantum mechanics in the most extreme sense. All continuity, all possibilities of infinitesimal gradation, are shoved outside of linguistics in one direction or the other. There are in fact two such directions in which we can and resolutely do expel continuity: semantics and phonetics. (Joos 1950: 702)

This is more specific than simply stating that these categories come from the "mathematical systems theory".

Every language has "meaningful" molecules called "morphemes," sub-assemblies which roughly correspond to subdivisions of the real world outside language. The word nose, for example, consists of one such morpheme; the word noses consists of two, the second having as its rela-world correspondent the conventional category of numerousness, one of our customary subdivisions or categorizations of the real world. Now consider these facts. The English word nose may refer to a part (how much?) of an airplane. With a lipstick draw a loop on your own face enclosing your nose and nothing else; another person will say that you have enclosed either too much or too little. Say in English: "The councillors all put their glasses on their noses" and then get the sentence translated into German; for English noses you will get the singular noun-form Nase, not the prular Nasen. The German knows that numerous councillors have equally numerous noses, and he has a word for "noses," but in this sentence he uses his word for "nose" instead. In linguistic terminology we simply say that the form Nase belongs to the category which we have called "singular." (Joos 1950: 702)

Jakobson uses a similar example, but his is much more limited.

The linguistic categories, then, are absolutes which admit of no compromise. They correspond roughly to favorite categorizations in the real world, and it is generally held that every community subdivides the phenomena in the real world according to the categories of its language, rather than the reverse. But the correspondence between the discrete categories of the language and the continuous phenomena of the real world is not and cannot be precise. Our reaction, as linguists, to this situation, is very simple: all phenomena, whether popularly regarded as linguistic (such as the tone of anger in an utterance) or not, which we find we cannot describe precisely with a finite number of absolute categories, we classify as non-linguistic elements of the real world and expel them from linguistic science. Let sociologists and others do what they like with such things - we may wish them luck in their efforts to describe them precisely in their own terminology, but no matter whether they describe them with discrete categories or not, for us they remain vague, protean, fluctuating phenomena - in a word, for us they represent that "continuity" which we refuse to tolerate in our own science. (Joos 1950: 703)

Very much related to concursivity. The linguist, or at least Joos's variety of linguists, does not deal with these matters. The precept here is that different languages have different linguistic categories for non-linguistic reality. I think it would be great to study these categories in specific languages. In Estonian, for example, there are many neat little words that categorize bodily behaviour that most of us youngsters aren't even aware of. "Kurekrõnks" is an example.

So much for the semantic end of language, with its intuitive bridge between language and the real outside world, the bridge across which we banish continuity. We do exactly the same thing at the other end, the phonetic end of language. (Joos 1950: 703-704)

Joos is excluding exactly those aspects that Jakobson urges the linguists to include.

Thus the word hotel cannot be spoken the same twice, from the physicist's viewpoint, no matter whether the same person or different persons should utter it; and yet from the linguist's viewpoint it always has in the middle a phoneme /t/ which is identical through neglect of unimportant variations in reality; rather, it is identical by definition, simply because it is a linguistic atom, a category, and these are either identically the same or absolutely different. (Joos 1950: 704)

And we are again grappling with the philosophical problems of sameness and otherness. That is, types and tokens, sinsigns and legisigns, etc.

...the codes called "languages" have numerous layers of complexity instead of only two, and in each layer there are severe limitations upon the combinations permitted. This much, we find, all languages have in common. But the particular code is different from one language or dialect to another, as the Baudot and Morse codes differ. (Joos 1950: 705)

Seems like a valid point. The cultural semiotician, with his or her interest in the interrelation of sign systems should appreciate this.

First analysis: cut it into the councillors and all put their glasses on their noses. The first part goes into several categories; similarly, in the four-unit analysis of Morse, the "dash" belongs to the two categories "long" and "mark." It is a "noun-phrase" because it fits into the sentence the same way as a plain "noun" might, sich as the word councillors; its further analysis will of course agree with this, for example it contains a "noun" and the word the, and such combinations generally make up noun-phrases. It is an "actor-expression," so marked by the absence of any other candidate for the job, while the rest of the sentence, the "action-expression," can use one and will couse anything to be sa classified which is not marked as having a different value: this one is rather marked as probably being an "actor-expression" by the fact that it is a "noun-phrase" and stands first, and its classification as an "actor-expression" is not forbidden by any such remark as the s of man's. (Joos 1950: 707)

I'm not sure how to feel about this. Is this better than the subject/predicate distinction? I'd rather not go either way and stick with generality for now. At best what I can take away from this is that there are indeed at least two forms of descriptions: those aimed at the actor and those aimed at the action. E.g. s/he is vs s/he does.