Bits and pieces of language functions

Jakobson, Roman 1962[1929a]. Remarques sur l'evolution phonologique du russe comparée à celle des autres langues slaves. In: Rudy, Stephen (ed.), Selected Writings I: Phonological Studies. s-Gravenhage: Mouton, 7-116.

Les éléments structuraux de la langue s'usent comme les parties de toute structure et de tout mécanisme. Il faut des réparations qui en renouvellent l'efficacité. Par exemple, les éléments de la langue affective s'automatisent, perdent leur teinte émotionnelle, changent de fonction. En conséquence, la langue affective se crée un nouveau répertoire de moyens d'expression. En outre, il peut y avoir changement de l'attitude même des sujets parlants à l'égard de la langue, changement des styles linguistiques dominants, et modification de la hiérarchie des fonctions. Un point essentiel, c'est, dans un ensemble linguistique donné, le rôle relatif de la langue affective et de la langue intellectuelle, de la langue poétique et de la langue de communication, de la langue théorique et de la langue pratique, du langage intérieur et du langage manifesté.
[The structural elements of the language are used like parts of any structure and any mechanism. Must repair that renew efficiency. For example, elements of emotional language automate lose their emotional tone, change function. Accordingly, emotional language creates a new directory of means of expression. In addition, there may be a change of the attitude of speakers towards the language, change the dominant language styles, and change the hierarchy of functions topics. A key point is that, in a given language set, the relative role of emotional language and intellectual language, poetic language and the language of communication, theoretical language and practice language internal and language expressed language. Google Translation.] (Jakobson 1962[1929a]: 18)

This is from the 9th item Jakobson ever published (according to "Works by Roman Jakobson"). What we have here is a theory of emotive function quite in keeping with the Russian Formalist theory of the genesis of functions. That is, like any artistic device, elements affective language can, too, become automatized, lose their emotional tone and change their function. In response to automatization and loss of emotional tone, new elements of emotive language are created.

Jakobson, Roman 1962[1949a]. Principes de phonologie historie. In: Rudy, Stephen (ed.), Selected Writings I: Phonological Studies. s-Gravenhage: Mouton, 202-220.

La capacité expressive du discours affectif est obtenue par une large exploitation des différences phoniques extraphonologiques existant dans la langue en question, mais au plus haut degré de l'affectivité le discours a besoin de procédés plus efficaces, et ne s'arrête même pas devant la déformation de la structure phonologique; par ex. divers phonèmes se confondent, phonèmes dont l'articulation est modifiée en vue de surmonter l'automatisme du discours indifférent; l'emphase va jusqu'à la violation des corrélations prosodiques existantes; certains phonèmes sont «avalés» en vertu de l'accélération du tempo. Tout cela est favorisé par le fait que dans le langage affectif le renseignement cède le pas à l'émotivité et que par là la valeur phonologique de certaines différences phonologiques s'atténue. De même la fonction poétique pousse la langue à surmonter l'automatisme et l'imperceptibilité du mot — et cela va également jusqu'à des déplacements dans la structure phonologique.
[The expressive capacity of emotional speech is obtained by a wide operating extraphonologiques phonic differences in the language in question, but the highest degree of affectivity speech needs more efficient processes, and does not even stop before the deformation of the phonological structure; eg. various phonemes merge, phonemes whose articulation is amended to overcome the automation of speech indifferent; emphasis goes to the violation of existing prosodic correlations; certain phonemes are "swallowed" under the accelerated tempo. All this is supported by the fact that in the emotional intelligence language gives way to emotions and thus phonological value of some phonological differences fades. Similarly, the poetic function pushes the language to overcome automation and imperceptibility of the word - and it will also travel to the phonological structure. Google Translation.] (Jakobson 1962[1949a]: 219)

Here he seems to agree with Martin Joos that the emotive elements of speech are nonlinguistic elements of the real world. E.g. emotive language makes use of sound differences that are outside of the phonological structure of the language in question; it may even deform the phonological structure - phonemes may merge or in fast emotional speech ever be skipped. The poetic function of language, also, tries to overcome the imperceptibility or automatism of practical, intellectual language.

Jakobson, Roman 1962a[1939]. Zur Struktur des Phonems. In: Rudy, Stephen (ed.), Selected Writings I: Phonological Studies. s-Gravenhage: Mouton, 280-310.

Man könnte vielleicht glauben, dass der Frageton eine besondere Satzbedeutung angibt. Doch wäre es vollkommen unberechtigt den Fragesatz als eine der Darstellungsarten zu betrachten. Der Fragesatz ist keine Darstellung, sondern blos eine Art von Aufforderung zur Darstellung. Das Fragen gehört nicht zur Darstellungsfunktion, sondern — nach Bühlers Terminologie — zur Appelfunktion. In der schematisierenden Projektion auf die Schrift ist also der Gegensatz zwischen Fragezeichen und Punkt demjenigen zwischen Ausrufezeichen und Punkt grundsätzlich ähnlich. Der Frageton, der Ausrufton und jedes lautliche Mittel des Appels und der Kundgabe steht zum Ausgedrückten in direkter, unmittelbarer Beziehung. So z.B. dei Überdehnung des betonten Vokals und des vortonigen Konsonanten im Deutschen (Jjjeesus!) oder die Zurückziehung der Betonung im Französischen (formidable !) besagt an sich die Gefühlsstärke. Der Frageton symbolisiert die Frage unabhängig vom Satzinhalt. Der Frageton kann sogar ohne Worte auskommen und mittels eines einfachen Murmeins verwirklicht werden. In den Dialogen der Kunst- oder Zeitungsprosa finden wir manchmal diese Art von Fragen durch ein alleinstehendes Fragezeichen wiedergegeben: — ? — .
[One might believe that the Frageton indicating a special sentence meaning. But it would be totally unjustified to consider the question set as one of the display modes. The question set is not a representation, but merely a kind of invitation to representation. The questions are not part of the representation function, but - according to Bühler's terminology - the Appel function. In the schematizing projection on the font so the contrast between the question mark and exclamation point to that between point and basically similar. The Frageton, the Ausrufton and each phonetic means of Appels and the proclamation is for that which is expressed in direct, immediate relationship. So, for example, dei overstretching of the stressed vowel and the consonant pretonic in German (Jjjeesus!) or the withdrawal of emphasis in French (formidable!) states the feeling of strength in itself. The Frageton symbolizes the question, regardless of the sentence content. The Frageton can even do without words and can be realized by a simple Murmeins. In the dialogues of art or newspaper prose sometimes we find these types of questions by a stand-alone question mark reproduced: -? -. Google Translation.] (Jakobson 1962a[1939]: 289)

Frageton = interrogative intonation ["question-tone"]; Appelfunktion = conative function; Ausrufton = to call, to call up; to call for; Gefühlsstärke = strong emotion, emotional strength. This makes the conative function (rather than the emotive function) comparable to Morris's modors. Although, the "Jjeesus!" example works perfectly with the example of emotive function in "Linguistics and Poetics"; e.g. [bIg] vs [bI:g].

Erste Klasse: das Bezeichnete fungiert in der Aussage, oder, breiter gesprochen, in der Sendung, als Inhalt. Jeder Satz, jedes Wort, jede grammatische Form, jedes Morphem gehört hierher. Jede von diesen Einheiten besitzt ihre eigene Bedeutung, so allgemein und lückenhaft sie auch sein mag. Das signatum fungiert dabei stets als ein Inhalt. Auch alle diejenigen Mittel, die zum Appell oder zur Kundgabe dienen, müssen unter diesem Gesichtspunkt in die gleiche Klasse eingereiht werden. Die Affektfülle, die Selbstäusserung des Senders, sein Verhalten dem Empfänger gegenüber, das alles gehört zum Inhalt der Sendung. Die Gesten und die piktographische Schrift, z.B. die Piktographie der Indianer, bezeichnen ebenfalls unmittelbar den Inhalt und gehören somit zu den Zeichenwerten der ersten Klasse.
[First class: the signified acts in the statement, or, more broadly, in the show, as content. Every sentence, every word, every grammatical form, each morpheme belongs here. Each of these units has its own meaning, so general and incomplete they may be. The signatum always acts as a table of contents. Also, all those means which are used to appeal or proclamation must be classified under this point of view in the same class. The affective value, self-expression of the transmitter, his behavior towards the receiver, all parts of the contents of the message. The gestures and the pictographic writing, for example, the pictography the Indians also call directly the content and thus belong to the sign-values ​​of the first class. Google Translation.] (Jakobson 1962a[1939]: 295-296)

There is a three-fold classification of semiotic ("Designation") phenomena, but I can't make out the exact point of it. These passages in foreign languages are not a lot of help.

---

Jakobson, Roman; Colin E. Cherry and Morris Halle 1962[1953a]. Toward the Logical Descriptionof Languages in their Phonemic Aspects. In: Rudy, Stephen (ed.), Selected Writings I: Phonological Studies. s-Gravenhage: Mouton, 449-463.

We leave aside here sound features that perform other functions, namely CONFIGURATIONAL features, which signal the division of the utterance into grammatical units of different degrees of complexity, and expressive (or more precisely PHYSIOGNOMIC) features, which signal solely the emotional attitudes of the speaker. [...] Physiognomic features are illustrated by the different ways of pronouncing the word for "yes" (simply [d'a] when unemphatic) according to the degree and kind of emphasis. These features convey subsidiary information similar to that which is carried by such graphic equivalents of configurational features as spaces or punctuation marks, and such equivalents of physiognomic features as underlining or italicizing. (Jakobson, Cherry & Halle 1962[1953a]: 451)

This is much more like it. "Physiognomic" features carry an emotive function and are definitely comparable to Morris's modors.

Jakobson, Roman and Morris Halle 1962[1956a]. Phonology and Phonetics. In: Rudy, Stephen (ed.), Selected Writings I: Phonological Studies. s-Gravenhage: Mouton, 465-504.

EXPRESSIVE FEATURES (or EMPHATICS) put the relative emphasis on different parts of the utterance or on different utterances and suggest the emotional attitude of the utterer. (Jakobson & Halle 1962[1956a]: 469)

It seems that "physiognomic features" went through a term-change (quite possibly because physiognomy is pseudo-science).

Under such conditions, the extraction of redundant features is, in many instances, laborious but feasible. More difficult is the isolation of the expressive features, but, even in this regard, the record may yeld some information, given the difference between the markedly discrete, oppositional character of distinctive features and the more continuous "grading gamut" characterizing most of the expressive features. (Jakobson & Halle 1962[1956a]: 476)

So not only is it difficult to elucidate expressive features (those parts of language that carry the so-called emotive function) but the distinction between distinctive and expressive features is tantamount to that between discrete and continuous codification (that is, binary oppositions vs grading gamut). For example, the difference between [bIg] and [bUg] consists of a difference in distinctive features (thus, informational, referential), while the difference between [bIg] and [bI:g] depends on the "emotional value" (or intensity or whatever) of the vowel. I am disturbingly cozy with these linguistic terms.

Ellipsis and explicitness. The case of the man faced with family names of people entirely unknown to him was deliberately chosen because neither his vocabulary, nor his previous experience, nor the immediate context of the conversation give him any clues for the recognition of these names. In such a situation the listener can't afford to lose a single phoneme from the message received. Usually, however, the context and the situation permit us to disregard a high percentage of the features, phonemes, and sequences in the incoming message without jeopardizing its comprehension. (Jakobson & Halle 1962[1956a]: 466)

This seems to be the earliest mention (thus far found) of either the context or situation of communication. The example is very down-to-earth and seems to point towards yet another phonemic interpretation of his scheme of language functions: that the referential function may have something to do with names... (I've had vaguer leads on the context/reference aspect.)

The code of features used by the listener does not exhaust the information he receives from the sounds of the incoming message. From its sound shape he extracts clues to identify the sender. By correlating the speaker's code with his own code of features, the listener may infer the origin, educational status, and social environment of the sender. Natural sound properties allow the identification of sex, age, and psycho-physiological type of the sender, and, finally, the recognition of acquaintance. Some ways to ward the exploration of these PHYSIOGNOMIC INDICES were indicated in Sievers' Schallanalyse, but their systematic study still remains on the agenda. (Jakobson & Halle 1962[1956a]: 470-471)

Is it a coincidence that they write this in 1956, the same year that Ruesch dropped his synopsis of the theory of human communication that dealt in detail with such "metacommunicative instructions".

Jakobson, Roman 1962[1926a]. Contributions to the Study of Czech Accent. In: Rudy, Stephen (ed.), Selected Writings I: Phonological Studies. s-Gravenhage: Mouton, 614-625.

Particularly detailed is Trávníček's treatment of various word units beginning with a monosyllabic word. The attention paid to the affective accentual variants makes his study a contribution to Czech stylistics. The reader is somewhat wearied by the excessively detailed sorting, while the basic classification of the data has not been properly pseneted by Trávníček. [...] The choice is determined by syntactic reasons (certain words cannot lose their primary stress while bearing the sentence stress) or by stylistic factors (the affective coloring of words). (Jakobson 1962[1926a]: 617)

Hmm, so here the affective, emotive, expressive, physiological, whatever value is relegated to "style". Since it was written so early it makes sense.

Jakobson, Roman 1962[1928c]. On the Elimination of Long Consonants in Czech. In: Rudy, Stephen (ed.), Selected Writings I: Phonological Studies. s-Gravenhage: Mouton, 626-628.

The lengthening of Czech consonants plays an emphatic role and serves, furthemore, as a secondary component of the stress. (Jakobson 1962[1928c]: 626)

I would not have thought to relate the emotive function and "emphatic role", but I found all of the above passages via the index under "Expressive (emotive) features (emphatica)".

Jakobson, Roman 1962c. Retrospect. In: Rudy, Stephen (ed.), Selected Writings I: Phonological Studies. s-Gravenhage: Mouton, 629-659.

The interrelation of distinctive, configurative (especially demarcative), expressive, and redundant features (cf. above, p. 469) requires precise comparative scrutiny. Such inquiry must particularly avoid any confusion between these essentially heterogeneous sets of features and any effacement of the actual limits between their divergent functions. Equally distorting is a prejudice to confine phonological investigation to the distinctive features alone and to label them quite arbitrarily as the only relevant and pertinent ones. Their discreteness, which sets them apart specifically from the grading gamut of expressive features, does not etitle the linguist to dismiss the latter. (Jakobson 1962c: 647)

So, several remarks. First of all, these different features may be what Jakobson has in mind when he says "a set (Einstellung) toward...". That is, perhaps it is not so much a "setting" (e.g. "orientation", as it has been translated into Estonian, for example) but rather a "group" (a logical "set") of phonological features. Secondly, this classification of features may be related to his famous language functions, perhaps, maybe. And thirdly, the expressive features are the only continuous (grading gamut) type of features, thus reinfocing the standpoints of Joos (1950) and Saporta (1960).

Phonological research faced two new fundamental problems, in accord with the twofold nature of language. Distributional analysis, which has been applied fruitfully to the SYNTAGMATIC relations within language, and to its phonemic structure in particular, but had been confined originally to sequential concatenation, demanded extension to the other dimension of the verbal sign, i.e., to the superposition of its simultaneous components. Henceforth questions of context embrace not only the antecedent and subsequent but also the concurrent factors. (Jakobson 1962c: 636)

This is an invaluable remark, because even if the auto- and synfunction (sensi Marty 1908 and then Gehring 1910) deal with antecedent and subsequent factors, the importance of concurrent factors has been painfully neglected. This aspect goes a long way towards making the so-called "communication model" more operational (though at this point I'm beginning to doubt if there is any point to making it operational).

"The typology of linguistic structures" has emerged as a timely task, and together wit hJ. N. Tynjanov I maintained the thesis that "an analysis of the structural laws which underlie language and its evolution necessarily leads us to ascertain a limited set of actually given structural types" (Novy Lef, No. 12/1928). Although this task cannot yet be considered accomplished, the ground has been cleared for systematic research (cf. above, p. 523ff.). The tentative list of distinctive features so far encountered in the languages of the world (see above, p. 477ff.) is intended just as a preliminary draft, open to additions and rectifications. A framework was traced by the close coöperation of the three authors of Preliminaries, supported by many helpful suggestions of our Harvard and M.I.T. friends; but a further, revised and specified version will undoubtedly bring more precise difinition for the correlates of single distinctive features at the different stages of the speech event. (Jakobson 1962c: 654)

It does seem indeed that the four types of phonetic features may have been "a preliminary draft" for the scheme of language functions.

In the treatment of "practical" language and its history we were still under too strong a pressure from the elaborate, codified, and compulsory tenet of the neogrammarians to venture toward modes of analysis which were to be tentatively christened STRUCTURAL METHOD, in my proposals to the first Congress of Slavists, 7 October 1929 (cf. Indogermanisches Jahrbuch, XIV, p. 386f.). Poetic language, disregarded by neogrammarian doctrine but presenting the most patently deliberate, goal-directed, and integrated linguistic species, was a field that called for a new type of analysis and particularly required us to study the interplay between sound and meaning. (Jakobson 1962c: 633)

Cf. "the "noncasual", purposeful character of poetic language" (Jakobson 1981[1960c]: 19).

Jakobson, Roman 1971[1957c]. Shifters, Verbal Categories, and the Russian Verb. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 130-147.

Both the message (M) and the underlying code (C) are vehicles of linguistic communication, but both of them function in a duplex manner; they may at once be utilized and referred to (= pointed at). Thus a message may refer to the code or to another message, and on the other hand, the general meaning of a code unit may imply a reference (renvoi) to the code or to the message. Accordingly four DUPLEX types must be distinguished; 1) two kinds of CIRCULARITY - message referring to message (M/M) and code referring to code (C/C); 2) two kinds of OVERLAPPING - message referring to code (M/C) and code referring to message (C/M). (Jakobson 1971[1957c]: 130)

Well, "M/C" (message referring to code) is a metalingual operation, and "M/M" is Ruesch's metacommunication, but the other two types, "C/M" and "C/C" don't seem "operational".

Jim told me "flicks" means "movies". This brief utterance includes all four types of duplex structures: reported speech (M/M), the autonymous form of speech (M/C), a proper name (C/C), and shifters (C/M), namely the first person pronoun and the preterit, signaling an event prior to the delivery of the message. In language and in the use of language, duplicity plays a cardinal role. In particular, the classification of grammatical, and especially verbal, categories requires a consistent discrimination of shifters. (Jakobson 1971[1957c]: 133)

Huh, so the "M/M" type concerns "repoted speech" or "speech within speech", and thus also probably "the text within the text" (e.g. Lotman 1994[1981). By "autonymous forms of speech" he probably means "synonymous" forms of speech, but here one may inquire whether the auto- and syn- distinction carries any semantic weight.

There is a multiplex scale of linguistic processes for quoted and quaszquoted speech; oratio recta, obliqua, and various forms of "represented discourse" (style indirect lybre). Certain languages, as for instance Bulgarian (s. Andrejčin), Kwatiutli (s. Boas), and Hopi (s. Whorf), use particular morphological devices to denote events known to the speaker only from the testimony of others. Thus in Tunica all statements made from hearsay (and this covers the majority of sentences in the texts aside from those in direct discourse) are indicated by the presence of /-áni/, a quotative postfix used with a predicative word (Haas). (Jakobson 1971[1957c]: 130-131)

This is actually quite interesting, because it is an example of a strictly linguistic "contextual" feature carrying a "referential" function. It is a linguistic device that designates the type of reference - in this case that it is not ostensible (present in the given situation) but reported from previous interactions.

Every shifter, however, posseses its own general meaning. Thus I means the addresser (and you, the addressee) of the message to which it belongs. For Bertrand Russell, shifters, or in his terms "egocentric particulars", are defined by the fact that they never apply to more than one thing at a time. This, however, is common to all the syncategorematic terms. E.g. the conjunction but each time expresses an adversative relation between two stated concepts and not the generic idea of contrariety. In fact, shifters are distinguished from all other constituents of the linguistic code solely by their compulsory reference to the given message. (Jakobson 1971[1957c]: 132)

Now here it makes perfect sense why shifters are classified above as "C/M" - this kind of constituent of the code does not have "a single, constant, general meaning", but must refer... Well, I would like to say, to the context of the speech act or the communication situation (addresser, addressee). I think this is so because shifters signal "an event prior to the delivery of the message", which would mean that it has to signal something outside of the message itself.

(M/M) "REPORTED SPEECH is speech within speech, a message within a message and at the same time it is also speech about speech, a message about a message," as Vološinov formulated it in his study of this crucial linguistic and stylistic problem. Such "relayed" or "displaced" speech, to use Bloomfield's terms, may prevail in our discourse, since we are far from confininf our speech to events sensed in the present by the speaker himself. We quote others and our own former utteranges, and we are even prone to present some of our current experiences in the form of a self-quotation [...] (Jakobson 1971[1957c]: 130)

The 1986 translation of Marxism and the Philosophy of Language by Ladisrav Matejka and I. R. Titunik reads:

Reported speech is speech within speech, utterance within utterance, and at the same time also speech about speech, utterance about utterance.
Whatever we may talk about is only the content of speech, the themes of our words. Such a theme - and it is only a theme - might be, for instance, "nature," "man," or "subordinate clause" (one of the themes of syntax). A reported utterance, however, is not just a theme of speech: it has the capacity of entering on its own, so to speak, into speech, into its syntactic makeup, as an integral unit of the construction. In so doing, it retains its own constructional and semantic autonomy while leaving the speech texture of the context incorporating it perfectly intact.
What is more, a reported utterance treated solely as a theme of speech may be characterized only superficially at best. If its content is te be had to the full, it must be made part of a speech construction. When limited to the treatment of reported speech in thematic terms, one can answer questions as to "how" and "about what" so-and-so spoke, but "what" he said could be disclosed only by way of reporting his words, if only in the form of indirect discourse.
However, once it becomes a constructional unit in the author's speech, into which it has entered on its own, the reported utterance concurrently becomes a theme of that speech. It enters into the latter's thematic design precisely as reported, an utterance with its own autonomous theme: the autonomous theme thus becomes a theme of a theme. (Voloshinov 1986[1929]: 115)

Quoted from: Voloshinov, V. N. 1986[1929]. Marxism and the philosophy of language. Translated by Ladislav Matejka and I. R. Titunik. Cambridge: Harvard University Press., that is, the 1st page of Chapter 2, "Exposition of the Problem of Reported Speech". The full reference is necessary because Jakobson himself mis-quotes a bit and doesn't cite. // Now I would like to experiment and bring together Jakobson's four duplex structures and the four initial factors participating in linguistic communication:

1. (M/M) - REPORTED SPEECH; QUOTATION; a message within (and about) a message;
2. (C/C) - PROPER NAMES; a patent (circular) reference to the code;
3. (M/C) - CIRCUMLOCUTION; SYNONYMS; a message referring to a code;
4. (C/M) - SHIFTERS; INDEXICAL SYMBOLS; PERSONAL PRONOUNS; a compulsory reference to the given message;

And now compare these to the four basic factors in "Results of a Joint Conference of Anthropologists and Linguists" (Jakobson 1971[1953d]: 556):

1. THE TOPIC OF THE MESSAGE - related to "theme" (cf. Voloshinov); relayed, displaced or quoted speech as-if conflates the poetic function, the context component and the content of the message;
2. THE RECEIVER - the addressee is referred to by proper names (e.g. "Jerry" and "Fido"); the interlocutor's name is a kind of "code in itself"; this replaces the conative function;
3. THE CODE USED - metalingual operations specify the code through circumlocution, synonyms and any other autonymous form of speech;
4. THE SENDER - instead of the emotive function here we have reference to the sender's person; autoreferentiality through personal pronouns, "I" and "me"; talking about oneself without using one's own name;

This conflation is very tentative, but ultimately points to some aspects of the scheme of language functions that have been neglected - namely, that reference involves much more than the context. In actual speech it is possible to refer to other messages by oneself, the other, or from somewhere else (M/M); to refer to the addressee by his name, or to any other person or non-person within or without the immediate context of the speech act (C/C); to refer to the words or phrases themselves and explain aspects of the code used or about language as such (M/C); and to refer to oneself (C/M). Ultimately this is an expansion of the referential function.

Jakobson, Roman 1971[1955a]. Aphasia as a Linguistic Topic. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 229-238.

In the next illustration the loss of a distinctive feature in aphasia is compensated for by an additional expressive feature. In Czech the opposition of long and short vowels is capable of distinguishing word meanings. Thus, draha means "road" when the first vowel is a long /a:/ and the second shart /a/ - /dra:ha/; but if the distribution of long and short is inverted, then the word /draha;] is the feminine singular of the adjective "dear"; the corresponding neuter form in colloquial Czech is /drahi:/. Hence in Czech, emphasis cannot be rendered by the prolongation of a short vowel, wheeras in Russian, in which the quastity of vowels has no distinctive value, the increase of the length of a vowel plays an important expressive role. (Jakobson 1971[1955a]: 229-230)

Another example of the expressive feature in Czech.

Language in its various aspects deals with both modes of relation [similarity and contiguity]. Whether messages are exchanged or communicated proceeds unilaterally from the addresser to the addressee, there must be some kind of contiguity between the participants of any speech event to assume the transmission of the message. The separation in space, and often in time, between two individuals, the addresser and the addressee, is bridged by an internal relation: there must be a certain equivalence between the symbols used by the addresser and those known and interpreted by the addressee. Without such an equivalence the message is fruitless - even when it reaches the receiver, it does not affect him. (Jakobson 1971[1955a]: 232)

E.g. there must be "a CONTACT, a physical channel and psychological connection between the addresser and the addressee" as well as a way to "check up whether they use the same code" (metalingual operations) (cf. Jakobson 1985[1976c]: 19-20).

Briefly, you must have a common code with the addresser, so that when listening to his message you can identify its constituents with the corresponding code units. We may, then, define the addresser as an encoder and the addressee as a decoder. (Jakobson 1971[1955a]: 233)

Cf. "a CODE fully, or at least partially, common to the addresser and addressee (or in other words, to the encoder and decoder of the message)" (Jakobson 1985[1976c]: 19).

However, it is not enough to know the code in order to grasp the message. When I say "he did", you may be familiar with the words he and did and with the rules of word order, and you will then realize that I speak about some man who performed some action, but in order to learn who this person is and what the action is which he performed, you need to know the context, verbalized or non-verbalized, but verbalizable. Here again we enter the field of contiguity. The components of any message are necessarily linked with the code by an internal relation of equivalence and with the context by an external relation of contiguity. (Jakobson 1971[1955a]: 233)

Now here "non-verbalized" begins to make sense. Specifically, in that while the "meta-language message" (M/C) can specify the "internal relation of equivalence" between, khm, "autonymous forms of speech", the "object-language message" (M/R) [here I replaced (C)ode with (R)eference and yielded the "practical" or "informative" function of language] must refer to something external to language - something in the context that is related to the given code-message "by an external relation of contiguity", a reference to something marked by "separation in space, and often in time" (above, 232), something either verbal or nonverbal, but in any case "verbalizable" in the sense that it can be referred to and spoken about. In this sense you cannot (at least not directly) talk (verbalize) something that is not verbalizable.

When operating with the hierarchy - phoneme, morpheme, word, sentence, utterance - linguists are tempted to see this arrangement as a merely quantitative scale, whereas in fact each of these ranks is also qualitatively, structurally different. (Jakobson 1971[1955a]: 233)

This argument could be used to show that the so-called "emotive function" is in fact very limited - specifically, with the level of phonemes.

The losses and compensations in aphasia throw new light on the interrelationship of the so-called cognitive and expressive elements in language; and aphasia providesa vital topic indeed for a conference on Expressive Language. (Jakobson 1971[1955a]: 230-231)

What?! Apparently, yeah, they had a conference and even published the papers: Werner, Heinze (ed.), 1955. On Expressive Language (Papers presented at the Clark University Conference on Expressive Language Behavior). Worcester, MA: Clark University Press. (GB)

Grading in respect to what quality? The answer is given by the context to which the suffix belongs. The word is the conetxt of morphemes, just as a sentence is the verbal context of words and an utterance the verbal context of sentences, while a morpheme in its turn is the context of phonemes. (Jakobson 1971[1955a]: 233)

This concerns the synfunctional dimension of context - the immediate verbal context of the syntagm.

In any language there exist also coded word-groups called phrasewords. The meaning of the idiom how do you do cannot be derived by adding together the meaning of its lexical constituent; the whole is not equal to the sum of its parts. Word-groups which in this respect behave like single words are a common but nonetheless only marginal case. (Jakobson 1971[1955a]: 242)

A neat addition to the notion of holophrase.

As noted above, it is the external relation of contiguity which unites the constituents of a context, and the internal relation of similarity which underlies the substitution set. (Jakobson 1971[1955a]: 249)

This is perhaps the clearest explanation of the context (and referentia function) I've found yet.

When, before World War II, phonemics was the most controversial area in the science of language, doubts were expressed by some linguists as to whether phonemes really play an autonomous part in our verbal behavior. It was even seggested that the meaningful (significative) units of the linguistic code, such as morphemes or rather words, are the minimal entities with which we actually deal in a speech event, whereas the merely distinctive units, such as phonemes, are an artificial construct to facilitate the scientific description and analysis of a language. (Jakobson 1971[1955a]: 252)

This attitude towards phonology actually helps explain why Jakobson deliberately or unintentionally "hid" the phonological core of his "communication model". In the end, the "language functions" are not so much functions of word-signs, as is often presumed, but functions of phonemes. E.g. the following passage: "When I was in Sweden, B. Collinder, who disliked phonemics, said that he would like me to do a book for the Linguistic Society of Uppsala: "Only, please no phonemics!" I was just completing my book on the phonemics of children's language and aphasia, and I simply eliminated the phonemic terms, upon which he said: "Now it's fine!" The book was, in fact, widely understood, and I, in turn, understood that it was possible, even in discussing totally new problems, to emancipate the work from new terms." (Jakobson 1971[1953d]: 557) This view - that "language functions" are actually "phonemic functions" - is reinforced by Jakobson's view that not the word, but the distinctive feature - his favorite term (Jakobson 1971[1964d]: 593) - is the ultimate verbal sign: "When phonemes are analyzed into elementary quanta termed 'differential elements' by Saussure and 'distinctive features' by Sapir and Bloomfield, the science of language discovers its ultimate units and their various superpositions and sequences." (Jakobson 1971[1953b]: 224) ["Pattern in Linguistics (Contributions to Debates with Anthropologists)", SW 2, pp. 223-228.]

Jakobson, Roman 1971[1953d]. Results of a Joint Conference of Anthropologists and Linguists. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 554-567.

Let us analyse the basic factors participating in linguistic communication: any speech event involves a message and four items connected with it - the sender, the receiver, the topic of the message, and the code used. (Jakobson 1971[1953d]: 556)

So in 1953 Jakobson still had only 4 items instead of 6 as he does in 1956. The timeline is so narrow that I am now tempted if I should read everything he published between these dates to elucidate the more-or-less exact point of birth for every factor and function (or, rather, the new ones). Notice that instead of context/reference we here have "the topic of the message" - a feature that is henceforth neglected (although it is in line with some Bakhtinian models in which there are: genre, topic, common memory, etc.).

E. Sapir analysed the linguistic phenomena prevalently from the view of their "cognitive function", which he conceived as the primary function of language. But the emphasis of the message on its topic is far from being the only possible. At present, the emphasis of the message on its other factors begins to attract greater attention among linguists both in this country and aboard, in praticular the emphasis on the communicators - the sender and receiver. Thus we welcome Smith's keen observations of those linguistic components which serve to characterize the speaker and his attitude to what he is speaking about and toward the listener. (Jakobson 1971[1953d]: 556-557)

1. It is now clear that the "set (Einstellung)" is properly "emphasis on". 2. I asked a long time ago, "What is the cognitive function?" (and why doesn't Jakobson talk about it?) - it is now clear that this comes from Sapir and has to do with "the topic of the message". 3. The term "emotive" is attributed to Anton Marty, but it is beginning to look like this mysterious "Smith" could be the initiator (quite possibly with "physiognomic features"). I have a guess at which Smith it may be. Jakobson's reference reads: S. Smith, "Contributions to the solution of problems concerning the Danish stød" Nordisk Tidsskrift for Tale of Stemme, VIII. On the web I found this: Smith, Svend 1944. Contributions to the Solution of Problems Concerning the Danish Stød: Bidrag Til Løsning Af Problemer Vedrørende Stødet i Dansk Rigssprog. Translated by Niels Haislund. The University of Michigan. This Smith's dissertation thesis is about phonetics of Danish language.

Returning now to the linguistic functions - I mentioned emphasis on the topic, on the sender, on the receiver; and we see how many new things we are able to do when analysing this paramount problem of sender and receiver. Moreover, there is the possibility of an emphasis either on the code or on the message. The emphasis of the message on its own self is called the poetic function. [...] The proper subject of inquiry into poetry is precisely language, seen from the point of view of its preponderant function: the emphasis on the message. This poetic function, however, is not confined to poetry. There is only a difference in hierarchy: this function can either be subordinated to other functions or appear as the organizing function. (Jakobson 1971[1953d]: 557-558)

Now I am even more sure that the "set (Einstellung)" is essentially "emphasis on". In this sense the referential function of language, for example, has an emphasis on a contiguous relationship with something in the antecedent, subsequent or concurrent context of the message, whether the referent be verbal or nonverbal (but if is to be referred to then it has to be verbalizable, because we are dealing with the speech act and not at all with nonverbal communication).

We mentioned the factors involved in the speech event, but we did not touch upon their interchangeability, the roles of sender and receiver merging and alterating, the sender or the receiver becoming the topic of the message, and the other interactions of all these factors. The most essential problem for speech analysis is that of the code common to both sender and receiver and underlying the exchange of messages. (Jakobson 1971[1953d]: 558)

This is where, ideally, the emotive and conative functions would step in - here we have at our hands messages with an emphasis on either one of the participants, e.g. ego-centric talk (and what ever might be the opposite of ego-centric talk... alter-centric talk?). But no, either one of the participants becoming the topic of the discussion does not seem to be the "simple solution" to either the emotive or the conative function. Rather than investigating such simplistic aspects of the communication situation, Jakobson focuses on the language, or "code", itself and the general aim of his later "scheme" becomes apparent - it is intended for speech analysis.

There occur attempts to revert to a very, very old, I should say pre-Whitneyan, stage of our science in considering individual speech as the only reality. As I already mentioned, individual speech doesn't exist without an exchange. There is no sender without a receiver - oh, yes, there is, if the sender is drunk or pathological. (Jakobson 1971[1953d]: 559)

Now I understand that I mistinterpreted this bit as Jakobson's early renouncement of autocommunication, to be rectified a few years later in Ruesch's terms as intrapersonal communication. What he actually seems to be talking about is the definition of communication and how it linguistics shouldn't be - as one early (here, contemporaneous) investigator of intrapersal communication put it - a study of "speech making". Whether you talk about "the exchange of messages between sender and receiver, between addresser and addressee, between encoder and decoder" (ibid, 559) in each and every case one implies the other: you cannot "send" if there is no receiver; you cannot "address" if there is no addressee; and you cannot "decode" something that isn't first encoded (in this last case the roles are reversed, though).

Jakobson, Roman 1971[1961b]. Linguistics and Communication Theory. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 570-579.

"The stimuli received from Nature," as Colin Cherry wisely stressed, "are not pictures of reality but are the evidence from which we build our personal models." (Cherry 1957: 62) While the physicist creates his theoretical construct, imposing his own hypothetical system of new symbols upon the extracted indices, the linguist only recodes, translates into symbols of a metalanguage those extant symbols which are used in the language of the given speech community. (Jakobson 1971[1961b]: 573-574)

What Jakobson seems to mean is that the linguist translated the symbols used in the speech community into the metalanguage of linguistic science. In case of "microlinguistics" (phonology), for example, the symbol "big" that the speech community uses is translated into "[bI:g]".

The constituents of the code, for instance, the distinctive features, literally occur and really function in speech communication. Both for the receiver and for the transmitter, as R. M. Fano points out, the operation of selection forms the basis of "information-conveying processes" (Fano 1949: 3f). The set of yes-or-no choiches underlying any bundle of these discrete features is not an arbitrary concoction of the linguist, but is actually made by the addressee of the message, insofar as the need for their recognition is not cancelled by the prompting of the verbal or non-verbalized context. (Jakobson 1971[1961b]: 574)

This is where I'm guessing that the "features" of the scheme (communication models) are sometimes called "features" instead of "components" because they may be related to the typology of distinctive, expressive, etc. features upon which the scheme was built. The part about non-verbalized context cancelling out the need for recognizing discrete features remains cryptic. And "literally" occurring is what I think grounds his scheme in phonology rather than communication theory. That the scheme is really meant to aid the researcher in speech analysis, rather than modelling communication as such.

Although the framework of the linguistic code has been adequately outlined in linguistics, it is still frequently overlooked that the finite ensemble of "standard representations" is limited to lexical symbols, their grammatical and phonological constituents, and the grammatical and phonological rules of combination. Only this portion of communication may be defined as a mere "activity of replicating representations". On the other hand, it is still oportune to recall that the code is not confined to what communication engineers call "the bare intelligence content" of speech, but that likewise the stylistic stratification of the lexical symbols and the allegedly "free" variation, both in their constitution and in their combination rules, are "foreseen and provided for" by the code. (Jakobson 1971[1961b]: 573)

This allows for the "broad" view of code, as something not at all rigid and imposing one-to-one equivalences. (e.g. the Bakhtin vs Lotman "code" problem.)

According to Weaver, the analysis of communication "has so penetratingly cleared the air that one is now, perhaps for the first time, ready for a real theory of meaning", and especially for handling "one of the most significant but difficult aspects of meaning, namely the influence of context" (Shannon and Weaver 1949: 116). Linguists are gradually finding the way of tackling meaning and in particular the relation between general and contextual meaning as an intrinsic linguistic topic, distinctly separate from the ontological problems of reference. (Jakobson 1971[1961b]: 577)

Another statement about the context being the "white spot on the map". By the last bit Jakobson probably distances the linguistic problem of reference from the philosophy of language.

Jakobson, Roman 1971[1966c]. Linguistic Types of Aphasia. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 307-333.

As to Dr. Wepman's important remarks concerning the natural context of aphasic utterances, yes, the ideal would be to have not only tape recordings but also sound films, since gestures may plai a considerable role. In general, in our detailed discussion of aphasia, it is appropriate to recollect a statement of Highlings Jackson's to the effect that aphasia could be labeled "asemasia" because it is not necessarily confined to deficiencies in verbal behavior but can extent to other semiotic activities as well, for example, gestures accompanying speech. (Jakobson 1971[1966c]: 327)

This is quite relevant for the context function. In short, it is a kind of recognition that gestures as well may figure as "the antecedent and subsequent but also the concurrent factors" of a speech event.

Evidently there is an interrelationship between different semiotic patterns, but it does not mean that all of them must necessarily be disturbed. I recall, for instance, once more the remarkable observations of Feuchtwanger, who described the mutual independence of verbal intonation and musical melody. Either of them can be lost without any harm to the other. Aphasic cases of completely monotonous speech, without any syntactic and emotive intonations but with full mastery of complex musical melodies, and vice versa, were observed and recorded. In the question of verbal problems with respect to other semiotic provinces, I am again for autonomy but against any isolationism. When investigating aphasia, we must perform an intrinsically linguistic analysis and at the same time pay due attention to the semiotic whole. (Jakobson 1971[1966c]: 328)

"Emotive intonation" is yet another category under the emotive function. Ultimately it may be possible to make a list of all such variants, from affective expressions and expressive features to physiological features and punctuation (e.g. Frageton).

Jakobson, Roman 1971[1963e]. Implications of Language Universals for Linguistics. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 580-592.

For centirues this field [linguistics] has been a no-man's land, and only a few philosophical contributions - from the medieval treatises on grammatica speculativa, through John Atmos Comenius' Glottologia and the rationalist essays of the 17th and 18th centuries, to Husserl's and Marty's phenomenological meditations, and finally to the modern works in symbolic logic - have ventured to lay the foundations for a universal grammar. [...] Likewise the philosopher's distinction between autocategorematic and syncategorematic signs remains vital for the construction of a universal grammar even if some of its traditional interpretations have proved to be "totally untenable". (Jakobson 1971[1963e]: 590)

As far as I can tell, Jakobson himself associates syncategarematic signs with "shifters".

Jakobson, Roman 1971[1956b]. Two Aspects of Language and Two Types of Aphasic Disturbances. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 239-259.

The bipolar structure of language (or other semiotic systems) and, in aphasia, the fixation of one of these poles to the exclusion of the other require systematic comparative study. The retention of either of these alternatives in the two types of aphasia must be confronted with the predominance of the same pole in certain styles, personal habits, current fashions, etc. A careful analysis and comparison of these phenomena with the whole syndrome of the corresponding type of aphasia is an imperative task for joint research by experts in psychopathology, psychology, linguistics, poetics, and semiotics, the general science of signs. The dichotomy discussed here [metaphor/metonym] appears to be of primal significance and consequence for all verbal behavior and for human behavior in general.²⁸ (Jakobson 1971[1956b]: 256)

I'll first note that this paper is also dated as 1956a, because it appeared as "Part II" in The Fundamentals of Language (Part I, "Phonology and Phonetics" was written with Morris Halle, but Part II is attributed solely to Jakobson). The footnote numbered 28 reads: "For the psychological and sociological aspects of this dichotomy, see Bateson's views on "progressional" and "selective integration" and Parson's on the "conjuction-disjunction dichotomy" in child development: J. Ruesch and G. Bateson, Communication, the Social Matrix of Psychiatry (New York, 1951), pp. 183ff.; T. Parsons and R. F. Bales, Family Socialization and Interaction Process (Grencoe, 1955), pp. 119f." This is the first reference to Ruesch and Bateson's Communication I've discovered thus far in Jakobson's corpus. To find out what exactly he is referring to here, I'll take a small detour (next subheading, Bateson 1951).

To indicate the possibilities of the projected comparative research, we choose an example from a Russian folktale which employs parallelism as a comic device: "Thomas is a bachelor; Jeremiah is unmarrked" Fomá xólost; Erjóma neženát). Here the predicates in the two parallel clauses are associated by similarity: they are in fact synonymous. The subject of both clauses are masculine proper names and hence morphologically similar, while on the other hand they denote two contiguous heroes of the same tale, created to perform identical actions and thus to justify the use of synonymous pairs of predicates. A somewhat modified version of the same construction occurs in a familiar wedding song in which each of the wedding guests is addressed in turn by his first name and patronymic:" Greb is a bachelor; Ivanovič is unmarried." While both predicates here are again synonymous, the relationship beween the two subjects is changed: both are proper names denoting the same man and are normalle used contiguously as a mode of polite address. (Jakobson 1971[1956b]: 256-257)

This illustration falls in line with "equational propositions" discussed under the rubric of metalingual operations in both "Metalanguage as a Linguistic Problem" and "Linguistics and Poetics". Earlier in the same paper:

When repeatedly asked what a bachelor was, the patient did not answer and was "apparently in distress [Goldstein 1948: 270]. A reply like "a bachelor is an unmarried man" or "an unmarried man is a bachelor" would present an equational predication and thus a projection of a substitution set from the lexical code of the English language into the context of the given message. The equivalent terms become two correlated parts of the sentence and consequently are tied by contiguity. The patient was able to select the appropriate term bachelor when it was supported by the context of a customary conversation about "bachelor apartments", but was incapable of utilizing the substitution set bachelor = unmarried man as the topic of the sentence, becaus the ability for autonomous selection and substitution had been affected. The equational sentence vainly demanded from the patient carries as its sole information: "«bachelor» means an unmarried man" or "an unmarried man is called «a bachelor»". (Jakobson 1971[1956b]: 246-247)

Because of the lengthier explanation, this example can be used instead of the sophomore one in the papers about the scheme of language functions.

Obviously such operations, labeled metalinguistic by the logicians, are not their invention: far from being confined to the sphere of science, they prove to be an integral part of our customary linguistic activities. The participants in a dialogue often check whether they are using the same code. "Do you follow me? Do you see what I mean?" the spaker asks, or the listener himself breaks in with "What do you mean?" Then, by replacing the questionable sign with another sign from the same linguistic code, or with a whole group of code signs, the sender of the message seeks to make it more accessible to the decoder. (Jakobson 1971[1956b]: 248)

Here i'm beginning to wonder if the "Do you follow me?" and "Do you see what I mean?" type questions are really aimed at the code component or rather concern the effect component that is missing from Jakobson's discussion. He does indeed sometimes retreat from such pragmatic questions to the more linguistic matters of "code". Here, for example, the breakdown in communication may not be caused by code, but by any number of extraneous factors, such as the receiver's inattention, distractions, "noise in the channel", disinterest (or feigned interest), etc.

Words with an inherent reference to the context, like pronouns and pronominal adverbs, and verbs serving merely to construct the context, such as connectives and auxiliaries, are particularly prone to survine [amnesic aphasia]. (Jakobson 1971[1956b]: 246)

Huh. Earlier I guessed that the context is related to (M/M) type of duplex structure. This was obviously naive. Here it turns out that the context is more related to the (C/M) type of duplex structure, e.g. shifters, indexical symbols and personal pronouns. This makes a lot more sense.

In the theory of language, since the Middle ages, it has repeatedly been aserted that the word out of context has no meaning. The validity of this statement is, however, confined to aphasia, or, more exactly, to one type of aphasia. In the pathological cases under discussion an isolated word means actually nothing but 'blab'. As numerous tests have disclosed, for such patients two occurrences of the same word in two different contexts are merely homonyms. Since distinctive vocables carry a higher amount of information than homonyms, some aphasics of this type tend to supplant the contextual variants of one word by different terms, each of them specific for the given environment. Thus Goldstein's patient never uttered the word knife alone, but, according to its use and surroundings, alternately called the knife pencil-sharpener, apple-parer, bread-knife, knife-and-fork (p. 62); so that the word knife was changed from a free form, capable of occurring alone, into a bound form. (Jakobson 1971[1956b]: 246)

The British comedian Stewart Lee has claimed something like this Medieval view of contextless words. Jakobson is here touching upon auto- and synsemantica (Marty) which can easily elaborate the context component.

In order to delimit the two modes of arrangement which we have described as combination and selection, F. de Saussure states that the former "is in presentia: it is based on two or several terms jointly present in an actual series", whereas the latter "connects terms in absentia as members of a virtual mnemonic series". That is to say, selection (and, correspondingly, substitution) deals with entities conjoined in the code but not in the given message, whereas, in the case of combination, the entities are conjoined in both, or only in the actual message. The addressee perceives that the gives utterance (message) is a combination of constituent parts (sentences, words, phonemes, etc.) selected from the repository of all possible constituent parts (the code). The constituents of a context are in a state of contiguity, while in a substitution set signs are linked by various degrees of similarity which fluctuate between the equivalence of synonyms and the common core of antonyms. (Jakobson 1971[1956b]: 234-244)

That last part is especially significant, because it indicates that context is constituted by contiguity (the referent is, as-if, in presentia) while the code is a virtual core of metalingually autonymous set of signs (and, of course, in absentia).

These two operations provide each linguistic sign with two sets of interpretants, to utilize the effective concept introduced by Charles Sanders Peirce: there are two references which serve to interpret the sign - one to the code, and the other to the context, whether coded or free, and in each of these ways the sign is related to another set of linguistic signs, through an alternation in the former case and through an alignment in the latter. A given significative unit may be replaced by other, more explicit signs of the same code, whereby its general meaning is revealed, while its contextual meaning is determined by its connection with other signs within the same sequence. (Jakobson 1971[1956b]: 244)

This is actually a workable definition of "contextual meaning" and can help explain the referential function (moreover, in terms of infinite semiosis).

Bateson, Gregory 1951a. Information and Codification: A Philosophical Approach. In: Ruesch, Jurgen and Gregory Bateson, Communication: The Social Matrix of Psychiatry. New York: Norton, 168-211.

Certain characteristics of the total codification-evaluation process will now be considered, in such a way as to pose questions about changes in this process - such changes being, according to our hypothesis, essential to therapy.
Broadly, there seem to be two sorts of process within the general area of codification-evaluation. These may be contrasted by considering two extreme examples. The first process we shall call decision by selective integration, and we shall exemplify it by a man's making a choice among a number of objects. To make this choice, he recognizes the specific objects as apples, oranges, pears, etc., and he knows from past experience which he likes and what actions and gratifications will be involved in eating the various sorts. If there is an unknown fruit among them, this too will be categorized as "unknown," and this category will have positive or negative value, determined by past experience. In this process of selective integration, the man categorizes and evaluates alternatives according to impressions derived from past experience, equating and differentiating elements of the unique present according to his experience with other elements in his unique past.
In contrast, an entirely different process of decision seems to occur in, for example, an extemporizing dancer. For any given movement within a sequence of movements, it is evident that some type of selection occurs which is different from the choice of a fruit of a given species. The dancer's choice is influenced to a much greater extent by the ongoing characteristics of his sequence of actions, and even, perhaps, by the ongoing dancing of a partner. This second type of decision we shall call decision by progressional integration, and we shall amplify the example by saying that the phenomenon is not confined to activities involving rapid physical movement, though the movement of the dancer is a convenient model to characterize the state of any person whose actions involve relatively rapid complex movement in a "psychological space." It seems that this type of progressional integration is especially characteristic for action sequences in which the component acts are imperfectly differentiated and categorized, and in which speed of decision is important. (Ruesch 1951a: 183-184)

This subchapter is titled "Selective and Progressional Integration" so it must be what Jakobson was referring to. The connection with the dichotomy of metaphor and metonymy seems contingent on Bateson's "Selective Integration" manifesting "a patently metaphorical attitude" (Jakobson 1971[1956b]: 256), and "Progressional Integration" having more to do with contiguity (temporal succession, spatial relations). Personally I see here more of a distinction between sender and receiver. The sender is engaged in Progressional Integration, involving relatively rapid complex movement in a "psychological space" and the receiver in turn engaging in Selective Integration, categorizing and evaluating alternatives according to impressions derived from past experience. This analogy makes more sense for me because whatever the speaker says is a continuous stream of "psychological movement" through mental images and thought content, but the listener takes in only a portion of what he hears, mainly that which is familiar to him, stuff he can recognize as significant to him, personally. In case of a book, for example, the stream of text covers page after page after page, but for me as a reader only a small portion of it is actually relevant for present purposes.

Both the selective and progressional processes are probably present in some degree in every human decision. The man who is choosing fruit is in part influenced by the ongoing sequences of his own metabolism, by his preference for certain sequences of taste, and by the intricacies of ongoing courtesy between himself and any other person present. To this extent he acts on a progressional integration. Correspondingly, a dancer may envisage alternatives of action (including the alternative of ceasing to dance), and he may introspectively believe that he is choosing among these categories. In general, it seems that the selective and progressional phenomena may occur each within a frame defined by the other: After he has decided to eat a certain fruit, the details of the act of eating may be progressively determined within the farmework of the selective decision. And conversely, in ongoing decisions involving long spans of time it is common for an individual to act selectively at every step and to discover that he has gradually made a major decision (e.g., the choice of a profession) by some progressional process. (Ruesch 1951a: 184-185)

Selective and Progressional Integration by themselves are of course abstractions. Actual life-processes follow both types and complex combinations of these processes. I still don't see what any of this has to do with metaphor and metonym. It rather seems that Jakobson identified this discussion with his dichotomy because of the key terms: selective integration may sound like the paradigmatic selection between semantically equivalent word-signs (synonyms); and progressive integration may sound like syntagmatic combination of word-signs into a string (a phrase). I will surely have to return to this paper by Bateson sometime in the future, but for now I can leave it be. Just in case I'll check up the reference to Parsons as well.

Parsons, Talcott and Robert F. Bales 1955. Family, Socialization and Interaction Process. In collaboration with James OLds, Morris Zelditch, Jr. and Philip E. Slater. Glencoe, Illinois: The Free Press.

It will be remembered that we called attention above to the presumption that the earlier phase of language learning was dependent on successful establishment of the love-dependency relationship with the mother. A second phase of intellectual development seems to be clearly associated with the completion of the oedipal phase and growth of the object-systems which are universalistically categorized. Such authorities as Piaget [Language and Thought of the Child, 1926) and Wallon [Les origines de la pensée chez l'enfant, 1934] strongly emphasize that the development of logical thinking occurs, with individual variations of course, essentially during this period, not getting notably under way before the ages of five to six and continuing until about twelve. There seems to be an important connection between this timing and the fact that in social object terms it is in this phase that universalistically categorized social objects, specifically as differentiated from those which are particularistically categorized, are first clearly internalized. It is presumed not altogether fortuitous that in Western countries formal education begins about the age of six. It may also very well be that the higher levels of abstraction involved in the use of written as distinguished from spoken language, perhaps particularly the elimination of the expressive cues given with tone and cadence of voice, pronounciation, facial expression and gesture, can only be attained when the kind of elaboration of personality structure we refer to has occurred. In particular a spoken communication is relatively inseparable from who says it in what specific context. A written communication focuses attention much more specifically on the meaning-content itself. This is surely more universalistic. (Parsons & Bales 1955: 118-119)

This precedes the passage Jakobson was referring to, but is worth quoting because it points out something about the emotive/expressive function of spoken language - its intimate connection with nonverbal forms of expression and the speaker's person (the "physiological" features).

In an important sense, logical thinking always involves both conjunction and disjunction of objects as many modern logicians put it [Ralph E. Eaton, General Logic, 1931]. And perhaps the latter is most characteristically logical. Thus, for example, in adding together four different things to a sum of four, the child is reasoning by conjunction: 1 plus 1 plus 1 plus 1 makes 4. But at the same time he is reasoning by disjunction in the sense that each one of the things is identical to each other in the abstract sense that it is one more thing to be counted. (Parsons & Bales 1955: 119)

This should somehow relate to Jakobson's metaphor/metonymy dichotomy. The words identical and equivalent are perhaps indeed synonymous, but beyond that I'm not sure how these dichotomies (metaphor/metonymy and conjunction/disjunction) are related.

Disjunction denotes this "or" relation between objects which makes A or B substitutable, i.e. identical, not in the concrete, but for some abstract purpose like counting. Disjunction implies a cognizance of the criteria of similarity between the two concrete objects which are disjoined. (As the teacher often says to a schoolchild, you cannot add oranges and apples ond come out with a meaningful sum unless you make explicit some other criterian of classification.) But, because disjunction implies this cognizance of criteria of similarity, it implies that there be an abstract concept instantiated equally by all the disjoined objects. (Parsons & Bales 1955: 119-120)

More set theory? It would appear that metaphor is related to disjunction and metynomy to conjunction. It does make some little sense. But I wouldn't say that these are "psychological and sociological aspects of this dichotomy" as Jakobson did. Bateson's aspect seems perceptual (rather than generally "psychological") and Parsons' and Bales' aspect mathematical (rather than "socialogical"). Okay, back to Jakobson.

Jakobson, Roman 1971[1963c]. Parts and Wholes in Language. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 280-284.

In another type of whole-part relation, the word whole "refers to some temporal period, whose parts are temporal intervals in it", and, as Nagel (1-b) stresses, neither wholes nor parts need to be temporally continuous. The verbal message, for example, a sentence, is a temporal period, and its parts are temporal intervals within this whole. The parsing of a sentence, like linguistic analysis in general, must necessarily follow the principle of "immediate constituents", defined by Husserl and exhaustively elaborated by American linguistics.
These constituents are a clear example of virtually discontinuous parts, such as, for instance, the subject at the beginning and the verbal predicate at the end of the Mongolian sentence. On the other hand, every message may be and must be dealt with as a temporal interval within a verbalized or nonverbalized, continuous or discontinuous temporal context; and we stand before the nearly unexplored question of the interrelation between message and context. In particular, the structural laws of ellipsis have not yet been subjected to a thorough analysis. (Jakobson 1971[1963c]: 281-282)

He is very correct in stating that the context component has been poorly dealt with. He brings the notion of ellipsis into play because the latter is defined as "the omission from speech or writing of a word or words that are superfluous or able to be understood from contextual clues". // I'll also note that this paper, "Parts and Wholes in Language", seems to be a response to this: Nagel, Ernest 1952. Wholes, Sums, and Organic Unities. Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition 3(2): 17-32. I considered reading it now, but there seems to be no point. Rather, I should read it when I'm about to write a paper on the semiosphere.

The artificial treatment of messages without reference to the superposed context once more exemplifies the illicit conversion of a mere part into a seemingly self-sufficient whole.
A contiguous question is the dependence of the message on the simultaneous environmental situation. Here the speech event is "spatially included" in a whole "with a spatial extension" (Nagel, 1-2). The spatio-temporal framework of the message becomes one of the crucial problems for any objective approach to language. From a realistic standpoint, language cannot be interpreted as a whole; isolated and hermetically sealed, but must be simultaneously viewed both as a whole and as a part. (Jakobson 1971[1963c]: 282)

The message (at least a spoken language message) is indeed not a self-sufficient whole but in fact a part of the overall - to use Ruesch's term - communication situation. The "spatio-temporal framework of the message" (or the messages' chronotope, for short), can lead to "The Factual Space and Time of Behavior" (Bentley 1941).

A systematic consideration of multiform whole-part relations broadly extends the scope of our science; it allows a systematic analysis of verbal messages with respect both to the code and to the context; it uncovers the complex interaction of the various levels of language, from the largest to the smallest units, and the constant interplay of diverse verbal functions. It introduces time and space factors into descriptive linguistics [...] (Jakobson 1971[1963c]: 284)

Indeed it exceeds the scope of linguistics, because a full consideration of the context of verbal communication should consider nonverbal behaviour as well. In any case, since this paper was based on a lecture held at M.I.T. in 1960, it could very well be the first intance of Jakobson seriously considering the context component.

Jakobson, Roman 1971[1964b]. Toward a Linguistic Classification of Aphasic Impairments. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 690-692.

Another variant of the same syndrome has been described by Luria and his collaborators. Luria defines this variant as the "dissolution of the regulative function of speech" (1959; 1962, p. 214). Viewed in its linguistic aspect, this symptom, however, may be interpreted as an inability to transpose a verbal dialogue into a nonverbal, artificial system of signs or to carry on a dialogue combining verbal utterances with utterances transposed into the latter system. Such semiotic activities would again go beyond the combinations stipulated and regulated by the habitual verbal code. The patient, as Luria (1962, p. 244) pointed out, "constantly slips toward accustomed verbal clichés". (Jakobson 1971[1964b]: 297-298)

Damn. The regulative function of speech has been taken - hopefully I can at least reserve the regulative function of nonverbal communication for my purposes.

In general, the transition from verbal stimuli to responses in non-verbal sign systems belongs, among the most interesting linguistic and semiotic problems. The inhibition of visual dreams connected with encoding disorders of language (Anan'ev, 1960, p. 336) has been rightly interpreted as a breakd-down of that code which provides the transition from verbal to visual signs (Žinkin, 1959, p. 475). (Jakobson 1971[1964b]: 298)

Cf. (Lotman 2009: 143) for a verification of how dreams indeed constitute an interesting semiotic problem.

Three dichotomies have been found to underlie six types of aphasic impairments (see Fig. 1). Speech devoid of any cognitive function and reduced to mere emotive, interjectional exclamations remains out of scope of this survey. (Jakobson 1971[1964b]: 301)

Cf. "The purely emotive stratum in language is presented by the interjections."

Jakobson, Roman 1971[1960b]. The Kazan's School of Polish Linguistics and Its Place in the International Development of Phonology. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 394-428.

For the young Baudouin the phoneme, like any other linguistic category, belongs to langue. This code is by no means merely an invention of scholar; it forms the fundamental basis for every spoken message. "Linguistic categories, based on the people's insight into their language", such as "sound, syllable, root, stem (theme), ending, word, sentence, various categories of words" really do "live in the language". "The people's insight into their language, i.e. everything that we would now call the metalinguistic operations of the language community, "is not a fiction, is not a subjective delusion, but", as Badouin observed in 1870, "a genuine, positive function which can be defined by its properties and acts, confirmed objectively and factualy proved", since in general for Baudouin, the realist, "all science basically constitute one single science whose subject matter is reality". (Jakobson 1971[1960b]: 411-412)

It would appear that Baudouin was knowledgeable of what we now call metalingual operations, but I can't make out if it's actually so or if Jakobson wished it were so.

Saussure carefully studied the theory of language elaborated by the two linguists, whom history has linked forever, and in his lectures, which were refashioned by Bally and Sechehaye into a posthumous book, he took from the teachings of Baudouin and Kruszewski and eloquently discussed such fundamental dichotomies as linguistic statics nad dynamics (or in Baudouin's and likewise Saussure's favorite formulation, kinematics); constancy and changeability (immutabilité and mutabilité), and correspondingly "the eternal antagonism between a conservative force, based on associations by contiguity, and a progressive force based on associations by similarity" (solidarité avec le passé and infidélité au passé); language and speech (langue and parole; centrifugal and centripetal forces in language (force particulatrice and force unifiante); the coherent whole of the system and its parts; association by similarity, viz. "the bonds of kinship" (solidarité associative or groupement par familles), as opposed to "bonds of contiguity with concomitants" (solidarité syntagmatique); finally the "inseparable pair" of signans and signatum - oboznačajuščee and oboznačaemoe (signifiant and signifié). (Jakobson 1971[1960b]: 421)

"Kinematics" sounds like something to my liking (for discussing "permanent dynamic synchrony"). I especially like that Baudouin was behind centrifugal and centripetal forces, because even since reading Märt Väljataga's afterword in his edited Kirjandus kui selline, I thought that Bakhtin was behind these terms.

Jakobson, Roman 1971[1964d]. Results of the Ninth International Congress of Linguists. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 593-602.

In the forties, with their sectarian parochialism in linguistic life, Giuliano Bonfante had great difficulty in his efforts toward finding an Italo-American lingua franca for our science. Yet since that time both he and his American opponents have matured and changed, and there is prospect of a common parlance. His creed, which he presented here, is in perfect agreement with the drift toward integration manifested by this Congress. In particular, as regards his statement that "the Crocean or esthetic theory of language can and must be integrated with the structural theory" and that "special attention must be devoted to the 'peripheral' zone of language - slang, jargon, affective and expressive terms, child language, onomatopoeia, interjections", we observe at our meetings how strongly linguists today are absorbed precisely with the structure of all these "peripheral" phenomena. Let us quote just a few topics of the papers presented: "affective linguistic signs", expressive and appellative phonology with particular reference to the manifold function of pitch, the non-intellectual "spheres of communication", "emphasis as a grammatical category", "the emotion in a sentence", sound-symbolism, "the development of grammar in child language". All these problems are being gradually incorporated into the structural analysis of language. (Jakobson 1971[1964d]: 590)

Some of the most interesting - and for me, central - issues are relegated to the periphery by Giuliano Bonfante.

Jakobson, Roman 1971[1969c]. Linguistics in Relation to Other Sciences. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 655-696.

The question of presence and hierarchy upon those basic functions which we observe in language - fixation upon the referent, code, addresser, addressee, their contact or, finally, upon the message itself - must be applied also to the other semiotic systems. In particular, a comparative analysis of structures determined by a predominant fixation upon the message (artistic function) or, in other words, a parallel investigation of verbal, musical, pictorial, choreographic, theatrical, and filmic arts belongs to the most imperative and fruitful duties of the semiotic science. (Jakobson 1971[1969c]: 661-662)

Some of these terms are a little different from how they are presented elsewhere (it's like a miniature bizarro-version). Note that applying the scheme of language functions on other semiotic systems us one of "the most imperative and fruitful duties" of semiotics.

Among relevant questions [that require psychological experimentation and elucidation], partly discussed by psychologists and partly awaiting an answer, one may cite speech programming and speech perception, the perceiver's attention and fatigue, redundancy as an antidote against psychological noise, immediate memory and simultaneous synthesis, retention and oblivion of verbal information, generative and perceptive memory for the verbal code, interiorization of speech, the role of different mental types in language learning, the interconnection of speechless status and language acquisition with different grades of intellectual development, and, on the other hand, relations between verbal impairments and intellectual deficits, or, finally, the significance of language for cognitive operations as compared with the prelingual status. (Jakobson 1971[1969c]: 671)

E.g. receiver's inattention, distractions, "noise in the channel", disinterest (or feigned interest), etc.

Jakobson, Roman 1971d. Retrospect. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 711-724.

The Saussurian inner duality of langue and parole (which mirrors the synonymous distinction of jazyk and reč launched by Baudouin de Courtenay in 1870: see above, p. 411) or, to use a modern, less ambiguous terminology, 'code' (Saussure's code de la langue) and 'message' - alias 'competence' and 'performance' - gives rise to two divergent approaches within the same section of the Cours: "Sans doute, ces deub objets sont étroitement liées et se supposent l'un l'autre", and, on the other hand, the author claims the impossibility of grasping "le tout global du language", insists on a strict bifurcation of the inquiry into langue and parole, and this restrictive program still finds its theoretical adherents, in fact the absolute separation of the two aspects turns into a recognition of twe different hierarchical relations: an analysis of the code with due regard for the message, and vice versa. Without a confrontation of the code with the messages, no insight into creative power of language can be achieved. Saussure's definition of langue as "la partie sociale du langaga, extérieure à l'individu" in opposition to parole as a mere individual act does not consider the existence of a personal code which removes the temporal discontinuity of the single speech events and which confirms the preservation of the individual, the permanence and identity of his ego; nor does he take into account the interpersonal, social, mutually adaptive nature of the "circuit de la parole" which implies the participation of at least two individuals. (Jakobson 1971d: 718-719)

A personal code? You mean a personal message system? Score for autocommunication!

When one-sided concentration on the cognitive, referential function of language gave way to an examination of its other, likewise primordial, underivable functions, the problems of the code-message relationship showed much greater subtlety and multivalence. (Jakobson 1971d: 719)

I should really look up what Sapir wrote about the "cognitive" function. // Apparently he writes about it in Language: An Introduction to the Study of Speech (1921), Chapter 2, "The Elements of Speech". Also, don't forget to read Ogden and Richards for the expressive function and Karl Bühler for the conative function!

Jakobson, Roman 1971[1959b]. On Linguistic Aspects of Translation. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 260-266.

In its cognitive function, language is minimally dependent on the grammatical pattern, because the definition of our experience stands in complementary relation to metalinguistic operations - the cognitive level of language not only admits but directly requires recoding interpretation, i.e. translation. Any assumption of ineffable or untranslatable cognitive data would be a contradiction in terms. (Jakobson 1971[1959b]: 265)

I can verify this only after reading Sapir.

Jakobson, Roman 1971[1956d]. Sergej Karcevskij: August 28, 1884 - November 7, 1955. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 517-521.

The intersection of morphology and syntax attracts him to the captivating problem of the common semantic value of preverbs and corresponding prepositions, and the interplay of two rival aspects of language, the cognitive and the emotive, inspires his ingenious insight into interjections and especially into the curious link between them and conjunctions. (Jakobson 1971[1956d]: 520)

I was not aware that these functions are "rivals". Though, an indication can be found here: "A number of students from various fields concerned with language activity have recently challenged the modern logistic view that would divide language from its beginnings into two separate spheres: the conceptual and emotive." (Langer 1955: 4) [Langer, Susanne 1955. Expressive Language and the expressive function of poetry. In: Werner, Heinze (ed.), On Expressive Language (Papers presented at the Clark University Conference on Expressive Language Behavior). Worcester, MA: Clark University Press, 3-9.]

Jakobson, Roman 1971[1970d]. Language in Relation to Other Communication Systems. In: Rudy, Stephen (ed.), Selected Writings II: Word and Language. The Hague; Paris: Mouton, 697-708.

The cardinal functions of language - referential, emotive, conative, phatic, poetic, and metalingual - and their different hierarchy in the diverse types of messages have been outlined and repeatedly discussed. This pragmatic approach to language must lead mutatis mutandis to an analogous study of the other semiotic systems: with which of these or other functions are they endowed, in what combinations and in what hierarchical order? Semiotic structures with a dominant poetic function or (to avoid the term relating chiefly to verbal art) with a dominant esthetic, artistic function present a particularly gratifying domain for comparative typological investigation. (Jakobson 1971[1970d]: 703-704)

I would not have guessed that Jakobson himself thinks of his scheme as being "pragmatic". But I do agree that if a similar endeavour would be undertaken on nonverbal communication, quite different functions may emerge.

Nicolas Ruwet, who combines a keen sense for language, especially verbal art, with a rare scientific insight into music, states that musical syntax is a syntax of equivalences: the diverse units stand in mutual relations of multiform equivalence. This statement prompts a spontaneous answer to the intricate question of musical semiosis: instead of aiming at some extrinsic object, music appears to be un langage qui se signifie soi-même. (Jakobson 1971[1970d]: 704)

Music does not refer to any extrinsic objects (ideally), but language (mostly) does. Unless of course we're dealing with metalanguage, in which case it refers to object-language, which in turn refers to some extrinsic object. In a similar vain one could inquire about each and every aspect of Jakobson's scheme: to what intrinsic or extrinsic object does it refer to? (or does it refer at all?)

The introversive semiosis, a message which signifies itself, is indissolubly linked with the esthetic function of sign systems and dominates not only music but also glossolalic poetry and nonrepresentational painting and sculpture where, as Dora Vallier states in her monograph L'Art Abstrait (1967), chaque élément n'existe qu'en fonction du reste. But elsewhere in poetry and in the bulk of representational visual art the introversive semiosis, always playing a cardinal role, coexists and coacts nonetheless with an extroversive semiosis, whereas the referential component is either absent or minimal in musical messages, even in so-called program music. What has been said here about the absence or scantiness of the referential, conceptual component does not discard the emotive connotation carrued by music or by glossolalia and nonrepresentational visual art. (Jakobson 1971[1970d]: 704-705)

For elucidating aspects of the functions the dichotomy of introversive and extroversive may indeed come handy. And I now see why Montana may have thought that introversive semiosis has something to do with autocommunication - the words "signifies itself" may have been misleading.

The uniqueness of natural language among all other semiotic systems is manifest in its fundamentals. The properly generic meaning of verbal signs become particularized and individualized under the pressure of changeable contexts or of nonverbalized but verbalizable situations. (Jakobson 1971[1970d]: 706)

I suspect that particularization here has to do with force particulatrice.

We analyze the messages wit hreference to all the factors involved, namely, to the inherent propertios of the message itself, its addresser and addressee, whether actually receiving the message or merely meant by the addresser as its virtual recipient. We study the character of the contact between these two participants in the speech event, we seek to elicit the code common to the sender and to the receiver, and we try to determine the convergent traits and the differences between the encoding operations of the addresser and the decoding competence of the addressee. Finally, we look for the place occupied by the given messages within the context of surrounding messages, which pertain either to the same exchange of utterances or to the recollected past and to the anticipated future, and we raise the crucial questions concerning the relation of the given message to the universe of discourse. (Jakobson 1971[1970d]: 697)

Another bizarro-explication of the model. Notice that Jakobson here verifies my guess that the "addressee" is a virtual, potential, meant receiver. The context aspect here is the most elaborate - both the auto- and synfunctional dimensions are represented.

If among messages used in human communication the verbal ones play a dominant role, still we have to take into account also all further kinds of messages employed in human society and to investigate their structural nad functional particularities [...] (Jakobson 1971[1970d]: 698)

It should at some point "click" that components are structural and functions are functional particularities (of the verbal speech act/event).