Signs, Language and Behavior

Morris, Charles 1949. Signs, Language and Behavior. New York: Prentice-Hall, Inc.

I am grateful to the publishers for permissions to quote from the following books: Harvard University Press: Collected Papers of Charles Sanders Peirce, edited by Charles Hartshorne and Paul Weiss; Susanne K. Langer, Philosophy in a New Key; Rudolf Carnap, An Introduction to Semantics; the University of Chicago Press: George H. Mead, Mind, Self, and Society, edited by Charles Morris; Jules H. Masserman, Behavior and Neurosis; Columbia University Press: Ray Lepley, Verifiability of Value; D. Apletton-Century Company, Inc.: Edward C. Tolman, Purposive Behavior in Animals and Men; E. R. Hilgard and D. G. Marquis, Conditioning and Learning; Clark L. Hull, Principles of Behavior; Harcourt, Brace and Company, Inc., and Kegan Paul, Trench, Trubner & Company, Ltd.: C. K Ogden and I. A. Richards, The Meaning of Meaning; Doubleday and Company, Inc.: Walt Whiteman, "Song of Myself," Leaves of Grass; Yale University Press: Robert M. Yerkes, Chimpanzees, A Laboratory Colony. (Morris 1949: vi)

A decent list of literature to keep an eye out in my future readings. Maybe I should finally read Langer's philosophy in a new key?

The understanding and the effective use of language and other signs has become today an insistent task. Popular and technical literature is filled with discussions of the nature of language, the differences between signs in men and in animals, the differentiation of scientific discourse and such other types of discourse as occur in literature, religion, and philosophy, the consequences of an adequate or inadequate use of signs for personal and social adjustment. (Morris 1949: 1)

A good introduction if there ever was one. Morris has an incredible ability for synthesis - these are the topics that are touched upon throughout the book.

The present study is based on the conviction that a science of signs can be most profitably developed on a biological basis and specifically within the framework of the science of behavior (a field which, following a suggestion of Otto Neurath, may be called behavioristics). Hence I shall constantly suggest connections between signs and the behavior of animals and men in which they occur. (Morris 1949: 2)

Morris wrote this book in the "heyday" of behavioris, so it should come as no surprise that his understanding of "a biological basis" is specifically American type of science of animal behaviour with concrete but very limited implications for humans ("men").

There is wide disagreement as to when something is a sign. Some persons would unhesitantly say that blushing is a sign, others would not. There are mechanical dogs which will come out of their kennels if one claps one's hands loudly in their presence. Is such clapping a sign? Are clothes signs of the personality of those who wear them? Is music a sign of anything? Is a word such as 'Go!' a sign in the same sense as is a green light on a street intersection? Are punctuation marks signs? Are dreams signs? Is the Parthenon a sign of Greek culture? Disagreements are widespread; they show that the term 'sign' is both vague and ambiguous. (Morris 1949: 3)

These questions still determine much of semiotic research. That is, these matters still remain somewhat vague and ambiguous.

At some point the semiotician must say: "Henceforth we will recognize that anything which fulfills certain conditions is a sign. These conditions are selected in the light of current usages of the term 'sign,' but they may not fit in with all such usages. They do not therefore claim to be a statement of the way the term 'sign' is always used, but a statement of the conditions under which we will henceforth admit within semiotics that something is a sign."
Then from such a starting point a behavioral theory of signs will build up step by step a set of terms to talk about signs (taking account of current distinctions but attempting to reduce for scientific purposes their vagueness and ambiguity), and will endeavor to explain and predict sign phenomena on the basis of the general principles of behavior which underlie all behavior, and henge sign-behavior. (Morris 1949: 4)

Mihhail Lotman claimed in his Struktuur ja Vabadus I that Morris dealt not with signs but with the conditions for something being a sign. It seems to me that he took this passage in the introduction way too seriously and didn't notice that these conditions are necessary for building "a behavioral theory of signs". On the other hand his theory, especially the lengthy part about modes of signification are quite difficult to understand (at least I myself will probably have to read this book several times before comprehending it).

The buzzer and the spoken sounds are then signs of food and obstacle because they control the course of behavior with respect to the goals of getting food and getting to a certain place in a way similar to the control which food and obstacle would exercise if they were observed. Whatever exercise this type of control in goal-sooking behavior is a sign. And goal-sooking behavior in which signs exercise control may be called sign-behavior. (Morris 1949: 7)

Quite useful. I like that he associates semiosis with goals and needs, because then I can frame my own nonverbal semiosis in terms of social needs (e.g. orienting in society, in everyday interactions, etc.).

By a stimulus is meant, following Clark L. Hull, any physical energy which acts upon a receptor of a living organism; the source of this energy will be called the stimulus-object. By a response is meant any action of a muscle or gland; hence there are reactions of an organism which are not necessarily responses. (Morris 1949: 8)

This sounds vaguely compatible with Uexküll's biosemiotics. Terms like "energy" and "action of a muscle or gland" but these are not incompatible with what we know from elsewhere, e.g. Peirce's interpretants ("muscle action" is energetic, "gland action" is emotional). The difference between reactions and responses is significant: reaction seems to involve all three interpretants (that is, cognitive as well), but response is bodily.

By this time a mixed feeling of dismay, fear, anger, and resentment may well have overcome the readers of contemporary literature on "semantics." Semiotic is not merely losing its entertaining character, they may say, but is slipping into the abyss of technicalities and hard work. And so it is! And so it must if the purpose be scientific. For science has always in its advance forced us away from the presented surface of familiar things to the laborious discovery of those properties of things which give insight, predictions, and control of this surface. And there is no reason why the scientific advance of semiotic should shun this road. No reason why sign-processes, for all their immediate sense of familiarity, should not be as complex as any chemical structure or biological functioning. And we have already admitted that for other immediate purposes, other less technical analyses may be more useful. (Morris 1949: 11)

I like to think that my own work maintains this "entertaining character" because "literary study" is not as "scientific". It can and will be full of technicalities and hard work, but the subject matter is still "literary". Also, Morris's prediction seems correct - many readers seemed to feel this mixed feeling he describes - JSTOR is chalk full of criticisms of this very book. It seems that after it's publishing every philosopher who was able to do so, went ahead and tried his best to demolish Morris. This may very well be the reason why his behavioristic semiotics is not applied further than repeating the cursory distinction between semantics, syntactics and pragmatics.

...we can, within limits, accept the results of analyses which have already been obtained in these fields and even verbal reports by a person on his own signs. (Morris 1949: 12)

From this point, concourse and "native theories" of nonverbal behaviour are nothing more than "verbal reports" of nonverbal signs.

The preceding formulation of 'sign' is not a definition in the sense that it gives the necessary and sufficient conditions for something to be a sign. It does not say that something is a sign if and only if the conditions laid down are met, but merely that if these conditions are met, then whatever meets them is a sign. (Morris 1949: 12)

Seems perfectly reasonable. The justification for this is explained in the notes (quoted later on): adding the "only if" would exclude types of signs that future research might uncover.

The evidence required may not even be a segment of a response-sequence, may not in fact be a response at all. If, for instance, any state of the organism could be found - say its brain waves - which is such that when this state is present the animal later responds in the way required by the formulation of a sign, then this state of the organism would itself be a sufficient condition for saying that the preparatory-stimulus which occasioned this state was a sign. (Morris 1949: 13)

I see this as opening up the possibility for studiying semiosis with magnetic-resonance imaging.

Finally, among such ways is the possibility of using in some cases the verbal responses of an organism as evidence of whether something is or is not a sign to this organism. By correlating such verbal responses with the rrest of the behavior of the organism it is possible to find out the extent to which they are reliable evidence as to the existence of sign-processes. That they are reliable to some extent is not in doubt, and so semiotics can make use of them as evidence for the existence of signs, and indeed must in its present state make as much use of them as it can. (Morris 1949: 14)

Again, literary descriptions of nonverbal behavior can be viewed as these kinds of "verbal reports" about nonverbal signs. By mapping out the "concourse" of specific literary works and authors, I believe we can finally reach "cultural signs" of nonverbal behaviour - that is, what kinds of nonverbal behaviours do certain (literary) cultures imagine to exist.

This approach also avoids the error of those accounts of signs which neglect their relation to the situation in which behavior occurs. Such accounts often imply that the organism responds to the sign alone. But the fact that behavior takes place within a supporting environment implies that the sign alone does not cause the response evoked, since the sign is merely one condition for a response-sequence in the given situation in which it is a sign. (Morris 1949: 15)

I believe this to be Morris's (seemingly "roundabout) way for approaching either the context of a sign, or the definition of the situation.

Any organism for which something is a sign will be called an interpreter. The disposition in an interpreter to respond, because of the sign, by response-sequence of some behavior-family will be called an interpretant. Anything which would permit the completion of the response-sequences to which the interpreter is disposed because of a sign will be called a denotatum of the sign. A sign will be said to denote a denotatum. Those conditions which arre such that whatever fulfills them is a denotatum will be called a significatum of the sign. A sign will be said to signify a significatum; the phrase "to have signification" may be takes as synonymous with "to signify." (Morris 1949: 17)

This may be the most important contribution that Morris brought along. For example, when I see a facial expression of joy, I am predisposed to interpret it as signifying joy. Yet whether the person is actually happy - if the sign denotes - is out of my hands and only the person him- or herself can tell. I should at some point consider this exact passage with regard to Dewey's criticisms (against the notion of interpreter).

In the case of the driver the words spoken to him are signs; the driver is the interpreter; his disposition to respond by avoiding a landslide at a certain place in the road is the interpretant; the landslide at that place is the denotatum; the conditions of being a landslide at that place is the significatum of the spoken words.
According to this usage of terms, while a sign must signify, it may or may not denote. The buzzer can signify to the dog food at a given place without there being food at the place in question, and the landslide signified by the spoken words may not in fact exist. Usually we start with signs which denote and then attempt to formulate the significatum of a sign by observing the properties of denotata. But it is possible at the higher levels of human sign-behavior to determine by decision the significatum of a sign (to "lay down" the conditions under which the sign will denote), and in this case the problem is not what the sign signifies but whether or not it denotes anything. We encounter cases of this nature frequently in the more complex sign-processes. (Morris 1949: 18)

In this example the interpretant seems to be energetic (action taken). The possibility that there is a landslide is the significatum of the warning, the actual landslide is the denotatum. The import for concursive study is that "descriptions of nonverbal behaviour" certainly signify and the significatum is usually tied to the context of the text, but the objective of the nonverbalist is to decide whether or not the description denotes anything. E.g. is there really a posture that accords to "sitting in a gentlemanly manner"? Does "going green with jealousy" occur in real life and how? etc. (these examples accord to my current distinction between descriptions and "metaforms").

The term 'meaning' is not here included among the basic terms of semiotic. This term, useful enough at the level of everyday analysis, does not have the precision necessary for scientific analysis. Accounts of meaning usually throw a handful of putty at the target of sign phenomena, while a technical semiotic must provide us with words which are sharpened arrows. 'Meaning' signifies any and all phases of sign-processes (the status of being a sign, the interpretant, the fact of denoting, the significatum), and frequently suggests mental and valuational processes as well; hence it is desirable for semiotic to dispense with the term and to introduce special terms for the various factors which 'meaning' fails to discriminate. (Morris 1949: 19)

An iconic passage.

The relation between interpretant and significatum is worth noting. The interpretant, as a disposition to respond caused by a sign, answers to the behavioral side of the behavior-environment complex; the significatum, as the set of terminal conditions under which the response-sequences to which the organism is disposed can be completed, connects with the environmental side of the complex. Each therefore involves the other. A complete description of the interpretant would include a description of what the organism is disposed to act toward, and a formulation of the significatum is simply a formulation of what would permit the completion of the response to which the organism is disposed because of a sign. The distinction between behavior and environment need not of course coincide with the distinction between the organism and the non-organic world, since an organism can respond to other organisms and to itself. It is because of this that parts of the organism, dreams, feelings, and even interpretants can be signified. (Morris 1949: 18)

Since something is a sign, significatum, denotatum, interpreter, or interpretant only with respect to its occurrence in sign-behavior, and since such constituents of sign-processes are studied by other sciences in other connections, the basic terms of semiotic are stable in terms drawn from the biological and physical sciences - a point which will prove to be of central significance for understanding the relation of socio-humanistic studies to the natural sciences. Since the factors operative in sign-processes are all either stimulus-objects or organic dispositions or actual responses, the basic terms of semiotic are all formulable in terms applicable to behavior as it occurs in an environment. Semiotic thus becomes a part of the empirical science of behavior, and can utilize whatever principles and predictions the general theory of behavior has attained or can attain. (Morris 1949: 19)

These two paragraphs were damaged in this book. I looked them up in Essential Readings in Biosemiotics and made a sheet to put between the book for future readers so that they won't become as frustrated with the damaged text as I did.

The formulation in terms of other signs of what a sign signifies (the description of the conditions which something must fulfull to be a denotatum of the sign) will be called a formulated significatum. A formulated significatum is designative if it formulates the significatum of an existing sign, and is prescriptive if it formulates the significatum which a sign is henceforth to have - a distinction which the commonly employed term 'semantic rule' fails to make. A sign may, of course, signify without there being a formulation of what it signifies. A recognition of this simple fact cuts the ground out from under the frequent insinuations that a sign is "without a meaning" to a person or animal who cannot "tell what the sign means." For something to signify is a different thing from the often very difficult task of formulating what it signifies. (Morris 1949: 20)

This is relevant for concourse studies, because denotata can be discriminated very seldom. It is rather a formulation of significata that concourse studies must occupy itself with. It is less important to find out if the behaviour occurs in reality than it is to find out what it signifies in the text.

A particular physical event - such as a given sound or mark or movement - which is a sign will be called a sign-vehicle. A set of similar sign-vehicles which for a given interpreter have the same significata will be called a sign-family. (Morris 1949: 20)

These terms may sound mystical, but - as a note later reveals - these are merely synonyms for Peirce's sinsign or token and legisign or type.

A sign-vehicle that does not belong to a sign-family is a unisituational sign since it has signification in only one situation. Such signs rarely if ever occur;; most signs are plurisituational. (Morris 1949: 21)

This unisituational sign is what I would call "unique sign" - because there is no type or rule for it, it cannot occur again in another situation - otherwise it would cease being a unique unisutiational sign and become a sign-family. I don't think these occur "rarely if ever" though - rather, from the first-person perspective, many nonverbal signs are borne first as unisituational signs and then graduate to plurisituational sign. For example, a unique facial expression catches your attention in a specific social situation and you don't know what it "means". Then you notice this same expression on other faces in other situations and perhaps read from experts, that this specific expression is related to a certain emotion - then you come to learn that there's a "rule".

To the degree that a sign has the same signification to a number of interpreters it is an interpersonal sign; to the degree that this is not so the sign is a personal sign. The interpreters for whom a sign is interpersonal may be called an interpreter-family. A given sign may be in principle entirely interpersonal or entirely personal; most signs are neither. Since it is always possible in principle to find out what a sign signifies for a given interpreter, and so make it interpersonal, no sign is inherently personal; but in actual practice many signs are highly personal - the signs of the schizophrenic provide extreme examples. It may be remarked that we should not necessarily classify a note which a person writes to himself for reading at a later time as interpersonal; such a note would be personal by the criterion proposed if the signs were signs to him alone, and interpersonal if this were not the case though no one else read the note. (Morris 1949: 21)

This may be freely related with #autocommunication and the curious phenomena of personal message systems.

A sign is vague to a given interpreter to the degree that its signification does not permit the determination of whether something is or is not a denotatum; to the extent that a sign is not vague it is precise. (Morris 1949: 21)

Sharpening our semiotic with technicalities.

A sign-vehicle is unambiguous when it has only one significatum (that is, belongs to only one sign-family); otherwise ambiguous. (Morris 1949: 21)

The example is "chair", because it belongs to the family "something you sit on" and "an academic position". In modern parlance this is simply "polysemanticity"

A sign is singular when its significatum permits only one denotatum; otherwise it is general. 'The 1944 President of the United States' is a singular sign, since it could not, because of its signification, denote more than one person. So is the sign 'I,' for though this sign is a family of sign-vehicles each member of which has the same significatum, the significatum that in each instance of signifying there can only be one denotatum (the person who produces the sign-vehicle of that instance). (Morris 1949: 22)

I believe this is exactly what Emile Benveniste is known for, at least in the semiotics of names (nimesemiootika või nimetamise semiootika).

The extreme of generality is reached when the significatum of a sign is such that the sign denotes the denotata of any other sign. Such a sign is universal, and is an implicate of every sign; the term 'being' and 'entity' seem to be, in the vocabulary of some philosophers, universal signs. (Morris 1949: 22)

In the vocabulary some semioticians, such terms as "structure" and "modelling" are equally general. For Jelena Grigorjeva, every sign system is "language".

A sign is reliable to the degree that the members of the sign-family to which it belongs denote; otherwise unreliable. The degree of reliability (and so the degree of unreliability) of a sign is capable of quantitative formulation. If the dog obtained food 90 per cent of the times the buzzer sounded, the buzzer sign is 90 per cent reliable; such a statement would not of course insure that the degree of reliability of a sign would continue unchanged in the future. A sign is iconic to the extent to which it itself has properties of its denotata; otherwise it is non-iconic. A portrait of a person is to a considerable extent iconic, but is not completely so since the painted canvas does not have the texture of the skin, or the capacities for speech and motion, which the person portrayed has. The motion picture is more iconic, but again not completely so. A completely iconic sign would always denote, since it would itself be a denotatum. A sign which is to some extent iconic may itself have properties which are not iconic and which are not relevant for its signification. One of the dangers of the use of models in science, for instance, arises out of the temptation to ascribe to the subject matter of a theory properties of the model illustrating the theory which are not involved in the theory itself. (Morris 1949: 23)

This may be useful for semiotics of art.

At the language level examples are more easily recognizable. If the driver in the car has been told to turn to the right at the third intersection, he might have held up three fingers on his right hand until he reached the intersection in question, or might have continued repeating the instructions to himself; such action on his part would be a sign to him signifying what the original spoken words signified, and such a sign would guide his behavior in the absence of the spoken signs.
Generalization from such examples suggests the following distinction: Where an organism provides itself with a sign which is a substitute in the control of its behavior for another sign, signifying what the sign for which it is a substitute signifies, then this sign is a symbol, and the sign-process is a symbol-process; where this is not the case the sign is a signal, and the sign-process is a signal-process. More succinctly, a symbol is a sign produced by its interpreter which acts as a substitute for some other sign with which it is synonymous; all signs not symbols are signals.
The advantage of such symbols is found in the fact that they may occur in the absence of signals provided by the environment.; an action or state of the intpreter itself becomes (or produces) a sign gouding behavior with respect to the environment. (Morris 1949: 25)

#autocommunication and signal-symbol distinction.

A person may interpret his pulse as a sign of his heart condition or certain sensations as a sign that he needs food; such signs are simply signals; his resulting words - when substitutes for such signals - would however be symbols. Nor are all sounds uttered by a person or by others symbols even when they are signs: sounds too may be simply signals. (Morris 1949: 26)

Signals/symbols distinction, "certain sensations" in terms of Sherrington's proprioception.

Some readers must feel strongly that we have "left something out," and perhaps something central to a theory of signs, namely the "ideas" or "thoughts" which the sign causes in its interpreters. At a number of places in our account we will return to this issue; here we must be content to isolate the underlying problem.
It is true that the preceding account has completely and deliberately avoided all use of "mentalistic" terms in building the terminology of semiotic. And it is also true that the mentalistic approcah has dominated the history of semiotic and still seems to many thinkers to be a preferable alternative to the approach here developed.
The defense of a behavioral semiotic must not be misunderstood. We have not contended, and do not believe, that such terms as 'idea,' 'thought,' 'consciousness,' and 'mind' are "meaningless". Nor have we in any sense denied that an individual can observe his feelings, or his thoughts, or his dreams, or his signs in a way not possible to other individuals. Our purpose is simply to advance semiotic as a science, and it is this purpose alone which determines what basic terms are to be accepted for building the terminology of semiotic. The issue is not between "mentalism" and "behaviorism," bit is solely a methodological ploblem: are such terms as 'idea,' 'thought,' 'mind' more or less procise, interpersonal, and unambiguous than such terms as 'organism,' 'stimulus,' 'response-sequence' nad 'disposition to response'? In choosing the latter terms we but express the belief that they are more suitable for scientific advance.
Suppose, for instance, the mentalist should argue - as he often does - that for something to be a sign to some interpreter it must give rise to an "idea" in his mind, must cause him to "think" of something else. (Morris 1949: 28)

My purpose is not to advance a "science" in this sense and I will probably not avoid using such mentalistic terms when I deem it suitable. I am not a "behaviorist". So it is quite likely that I will implement Morris's terms with those of Peirce ("idea") and Mead ("mind").

Daes the "phenomenological description" of sign-processes furnish another alternative to a behavioral semiotic? I think not. For in a wide sense of 'phenomenological' a behavioral semiotic is phenomenological since it includes a description of observed behavior; and a narrower use of the term (the description by an individual of his own sign-processes) is covered by our admission of self-observation, an admission which is compatible with either a behavioral or a mentasitic psychology, and so does not decide between them. (Morris 1949: 30)

I'm hoping for this 'self-observation' to make some leeway into appling Morris's "behavioristic semiotic" on nonverbal behaviour.

Because of this vagueness and ambiguity a number of scientists have proposed to discontinue the use of the term 'sign.' Of course behavioristic can be developed without the use of the term; no sign is indispensable. But since the term is in such frequent usage in ordinary language and in writings of semioticians and behavioristicans it has seemed advisable to continue the term but to use it in a more precise manner than is customary. Whether this usage is to be incorporated into behavioristics itself must be left to specialists in the field. (Morris 1949: 249)

But instead of a handful of putty Morris has crafted arrowheads such as significata and denotata.

I owe the suggestion of giving only sufficient conditions for saying that something is a sign, instead of giving a definition (that is, suffirient and necessary conditions), to Alfred Tarski. This procedure seems advisable at the present state of the argument, since premature definitions may rule out other whenomena which we may later wish to include. The consequence of the present procedure is that for the moment our statements about signs are limited to signs as identified by the criteria here proposed or by any other criteria which imply these criteria. (Morris 1949: 250)

This is really neat. I should do something similar, e.g. provide sufficient conditions, for nonverbal communication. A good idea would be to take the "historical" approch and define it in light of it's first uses in the psychology of attitude of the 1930s and refer back to it's possible origins as Cooley's pre-verbal (and, possibly, other mentions of pre-linguistic communication).

In previous writings (such as Foundations of the Theory of Signs), I took "mediated taking account of" as the primitive term for semiotic, defining a sign-process as a process in which something mideately took account of something else by taking account of something immediately present. The present analysis resolvose this primitive term into the stimulus, response, and organic state terminology of behavioristics. Hence it provides the basis for a behavioral formulation of all terms signifying signs. (Morris 1949: 251)

This primitive term sounds awfully good. That is, it sounds like Peirce's "standing for". I found Foundations of the Theory of Signs on scribd and it is only 60 pages long.

There is an obvious relation between the term 'behavior-family' and the term 'sign-family': a sign-family is a set of similar events or objects which act as preparatory-stimuli for response-sequences of the same behavior-family. This the term 'sign-family' involves the wider concept of behavior-family. Egon Brinswik has used the term "cue-family" in a way somewhat similar to the present usage of 'sign-family' (see his "Phychology as a Science of Objective Relations," Philosophy of Science, 4, 1937, 233). Peirce distinguishes sign-vehicle and sign-family by the terms 'token' and 'type'; Carnap by the terms 'sign-event' and -sign-design.' (Morris 1949: 251)

This made these notions much clearer.

The term 'language" is highly vague and ambiguous. Beyond the very general agreement that language is a sign phenomenon social in nature, there are many disagreements as to what is necessary for signs to be language signs. (Morris 1949: 32)

And this is why he postulates 5 conditions that must be fulfilled in order to call something truly a language.

The first item to be noted here is that the situation is one of social behavior, that is behavior in which organisms provide other organisms with stimuli. Where two or more organisms provide each other with stimuli, the social behavior may be called reciprocal; otherwise non-reciprocal. Behavior which is not social is individual. The behavior of both the speaker and the driver is reciprocal social behavior, for the words were spoken because of the presence of the driver and the words as spoken by the speaker were stimuli to the driver. An organism which produces a sign that is a stimulus in social behavior will be called a communicator, and an organism which interprets a sign furnished by a communicatior will be called a communicatee. (Morris 1949: 32)

Some general notions. I'd prefer communicator/communicatee to addresser/addresseer, but the latter is more prevalent here (in our local semiotics) because of Roman Jakobson.

A sign which has the same signification to the organism which produces it that it has to other organisms stimulated by it will be called comsigns. A comsign is this a special class of interpersonal signs, since not all interpersonal signs are comsigns. Each of the person in our example is thus potentially a communicator or communicatee of spoken words, and each is the communicatee when he is a communicator (the person giving the warning hears the words which he himself utters to the other person and they have the same or a similar signification to him and to the person addressed). (Morris 1949: 33)

#autocommunication and the distinction between interpersonal signs (signs that are available to others than the sign-producer) and comsigns (signs that are understandable to others than the sign-producer).

Some persons might argue that the "perception" of food or abstacle was itself a signal-process, and that language signs are always substitutes for such signals. But this raises the difficult question as to whether or not perception is to be regarded as a sign-process. (Morris 1949: 34)

Recall the cheese conundrum (Austin 1970: 15; footnote 1).

Comsigns are either activities of the organisms themselves (such as gestures), or the products of such activities (such as sound, traces left on a material medium, or constructed objects). An odor, for instance, might be interpreted in the same way by a number of organisms in a given situation, and hence be interpersonal, and yet would not be a comsign. Odors would be language signs only if in addition to being interpersonal they were producible by their interpreters. (Morris 1949: 35)

Odor is a good example of an unintentional signal - aside from using "hygiene products" and perfumes there are seemingly few actual possibilities for effecting one's odor. I also like that he mentions sounds as products of activities (e.g. phonokinesics).

Since the term 'language' is so vague and ambiguous in current usage, it may be doubted whether it is wise to employ the term at all in semiotic. In this connection it may be pointed out that what we have really done is to define a certain set of signs as a language and then to define language sign as a member of this set. The sign-vehicle 'language" is not important for the analysis. Hence it need not be used. We propose therefore to call sign-sets of the kind in question lansign-systems, and the individual members of these systems lansigns. In what follows we are thus concerned with lansign-systems and lansigns; and while we continue to use the terms 'language' and 'language sign,' for reasons now to be given, the reader who wishes may replace 'language' by 'lansign-system' and 'language sign' by 'lansign,' and then use the terms 'language' and 'language sign' in any way he wishes - or discard them altogether. (Morris 1949: 36)

This is an extremely relevant passage. I'll leave the implications open here.

But if the claim is made that what are commonly called language signs occur, after they have arisen, only in social b ehavior, even in the minimal sense of one organism producing stimuli for another organism, then the claim seems to be in fact doubtful. Suppose someone sitting alone in a room writes a poem which is then destroyed, that is, the poem is not produced in the presence of stimulation from other organisms nor does it act as a stimulus to any other organism. Yet surely it is intelligible to ask whether the poem was written in English or French. (Morris 1949: 37)

#autocommunication and recall "the law of conservation of poetry" (Torop 1999: 104).

As George H. Mead saw with incisive insight, the crucial genetic problem in the origin of comsigns is to explain how an animal that does something that another animal reacts to as a sign can itself react to its own action as a sign with the same significatum. In our terms, only if an organism can react to its own activity (or its products) wwith an interpretation (interpretant) similar ta that made to this activity (or its products) by other organisms can a sign producible by one organism have to that organism a signification in common with that of other interpreting organisms. Mead pointed out that most responses of an organism do not affect the receptor of an organism making the response as they do the receptors of other organisms - the organism, for instance, does not see its facial movements as do other organisms. (Morris 1949: 39)

Very true and important. What I would like to call "nonverbal self-communication" consists of taking note of one's own behavior (e.g. Chris Shea's example of noticing herself protruding her tongue). In this specific case the tongue protrusion became a sign to Shea because she was knowledgeable about the significance of tongue showing (A Facial Display of Humans and Other Primate Species).

Mead's conception of the gesture is clearly related to what we have called a signal, and he himself at times calls it a non-significant symbol to distinguish it from the significant symbol. But Mead sees that such gestures are not necessarily signs to the organism which produces them, since the organism seldom respond to these gestures themselves. He notes, however, that sounds produced in the reaction may themselves serve as gestures (he calles them under these conditions "vocal gestures") and that such sounds are heard by the producing organism itself as well as by other organisms. (Morris 1949: 43)

Even in case of humans nonverbal self-communication is achieved in rare occasions. Nevertheless there are those among us who craft their awareness of their own behaviour meticulously, spending many hours in front of the mirror, practicing their facial expressions and learning to recognize certain muscular sensations (they effectively learn how to use these muscles intentionally).

There is another factor in Mead's analysis that requires attention: the notion of "taking the role of the other." Speaking of the significant symbol Mead makes the following remarks: "We must indicate to ourselves not only the object but also the readiness to respond in certain ways to the object, and this indication must be made in the attitude or role of the other individual to whom it is pointed out or to whom it may be pointed out. If this is not the case it has not that common property which is involved in significance. It is through the ability to be the other at the same time that he is himself that the symbol becomes significant." Now at the level of voluntary language communication it is easy to identify what Mead has in mind: there is clearly some sense in which the person who tells another that a given road is blocked is "putting himself in the place of that other" and giving information regarded as relevant to some purpose of the other individual. There is merely the question whether such phenomena are to be embodied in the definition of the language sign or are to be regarded as necessary conditions for its genesis. Mead seems to vacillate somewhat on this point: at times he talks as if role-taking were a precondition of the significant symbol and at times as if it were made possible by such symbols. (Morris 1949: 45)

This sounds like the concept of semblance in James Mark Baldwin (I'd like to refer to Valsiner in The Social Mind on this point but I don't have the exact reference yet).

"...What the development of language, especially the significant symbol, has rendered possible is just the taking over of this external social situation into the conduct of the individual himself." The individual can then "utilize the conversation of gestures that takes place to determine his own conduct. If he can so act, he can set up a rational control, and thus make possible a far more highly organized society than otherwise." (Morris 1949: 47)

The relationship between self-control and social organization. Recall culture in the sense of display rules, e.g. self-censure.

Thus a sound, when a language sign, can be heard by other organisms than the communicator, but certainly there is some sense in which there is still a sign - a post-language sign - when the organism does not speak aloud and so is not heard by others or by itself. This is the kind of situation which J. B. Watson referred to as subvocal talking, and which many behaviorists have identified with thinking. The point here stressed is that such a sign is not social in operation, though social in origin since it depends on language. And iti s a new sign, since the original sign was a sound, while now no sound is uttered. The driver who repeats "silently to himself" the commands which he first heard from another has substituted for the spoken words (an exteroceptive stimulus) some stimuli within his own body (proprioceptive stimuli). The latter are different signs and not themselves language signs; since they are substitute sign synonyms with language signs they are properly called post-language symbols (and in this case, personal post-lnaguage symbols). (Morris 1949: 47)

Remember the behaviorist's identification of thought with speech. Also, internal dialogue. And this substitution or synonymity can be related with Derrida's grammatology. I'm not sure if Sherrington's notions (extero- and proprioception) are correctly used here, because he talks of physiological phenomena not sign-phenomena, so this move (which Ruesch also performs) is merely metaphorical/analogical.

It is not hard to understand in behavioral terms how personal post-language signs can arise. For since a language sign, such as a sound, is produced by an action of an organism, any other stimulus within (or at times without) the organism which is produced by the action (or connected with the action) may become a substitute for the sound. This substitute sign is by definition a symbol; it is not a language sign in that it is not a stimulus to other organisms; it has the advantage of making available to the individual organism with the minimum of time and effort the significance of the language sign for which it is a substitute. Different organisms will in general have different post-language symbols as substitutes for the same language sign. These post-language symbols will then be synonymous (or highly similar in signification) but they will not be interpersonal since they will beling to different sign-families of different interpreters.
It is not necessary, though it is the most frequent case, that the personal post-language sign be a stimulus within the organism. Ii might be a mark on poper made by the organism (as in the case of many personal systems of notation, which differ from "writing" in that the latter is interpersonal); it might be some stimulus-object in the environment which has ocurred along with language signs and so become a substitute for them (as a stone which one was fingering during a treasured conversation and which one then keeps on one's desk). (Morris 1949: 147-148)

Personal post-language symbols is an extremely interesting topic, but very rarely is something like this discussed. Juri Lotman studied Pushkin's marginalia and discussed some acronymic personal messages.

In case the stimuli in question are accessible to other organisms they may become interpersonal post-language signs, perhaps comsigns, and perhaps even elements in a language. In this way language extends its complexity to the natural and constructed objects which form the environment of a seciety. A development of semiotic in this direction offers great promise for a better understanding of the sense in which, and the degree to which, human culture is itself a sign phenomena. (Morris 1949: 48)

Indeed a lot of common technical words have at some point been a "personal" word for whoever happened to coin it. The successful signs became part of language, unsuccessful ones remain "personal" to whoever happens to be the only person who used them. Nonverbalism, semiophrenia and concursivity are my personal post-language symbols and I cannot know if others will ever use them (or use them to signify same concepts as I do). The note about culture here deserves more attention, but I'll have to return to it once I have read Powys's The Meaning of Culture - I expect him to relate his philosophy of solitude to the meaning of culture..

It may be fruitful to regard any interpretant as an "idea" (with 'concept' possibly being limited to the interpretant of a general symbol), and to regard the difference between "talking out loud" and "thinking to oneself" as the difference between the presence of language signs and personal post-language symbols. It is then a matter of decision whether to regard every sign-process as a "mental" process or whether to limit "mind" (as Mead does) to sign-behavior in whcih language signs or post-language symbols occur. That the distinctive characteristics of human mentality are closely related to language signs and personal post-language symbols seems certain in the light of Mead's discussion of reflective thinking. Whether the psychology of the future will, however, choose to define such terms as 'thought,' 'idea,' 'concept,' and 'mind' on the basis of semiotical distinctions must be left to psychologists. (Morris 1949: 49)

Generally the answer seems to be no, but then again there is cultural psychology (e.g. Valsiner), which does use semiotic notions (not to mention explicit psychosemiotics).

Signals, on this view, "announce their objects," while symbols lead their interpreters to "conceive their objects." How is this difference to be explained on the behavioral approach? (Morris 1949: 50)

This is how Morris concludes the signal/symbol distinction in Susan K Langer's Philosophy in a New Key.

If the interpreter has other signs at his disposal, as in the case of beings with a language, he may under these conditions pay attention to the sign in question, attempt to formulate its signification, raise questions as to its reliability, assess the bearings of this reliability upon whatever goals he may have, and even make observations upon his own disposition to respond which is produced by the appearance of the sign. And since signs are used for many purposes other than description and prediction, signs low in reliability may still be attended to, produced, as cherished - as in the case of a work of fiction. (Morris 1949: 51)

This extremely relevant for my work - if I'm formulating the significance of descriptions of nonverbal behavior, then I'll certainly meet some forms of behavior which have no basis in reality (or, which are impossible in reality), or which are way too ambiguous to signify anything (not te mention denotation) - but which are still used in fictional discourse exactly because it does not have to accord to reality but can creatively invent behavioural signs and their significations.

There have been many proposals for a unique differentiation of human signs: men have symbols while animals have only signals; men alone have signs of signs; human beings are unique in the transmission of signs by social heredity; human beings use signs voluntarily while animals do not; animal signs occur only in perception while human beings are capable of inferences based on signs; men alone become interested in signs as goal-objects; men alone have language. (Morris 1949: 53)

These proposals are still prevalent in semiotic discourse. For example, the "signs as signs" aspect has been remarked by John Deely, and in semiotics of art this contention is related to the image "This is not a pipe." and Juri Lotman's "formula of art": I see what it depects but I know that it is not what it depicts (but merely a sign of what it depicts).

It should be remarked that on our account not every response-sequence provoked by a stimulus-object is sign-behavior. A person reaching for a glass of water is not prepared to act in a certain way because of a sign, but simply is acting in a certain way to an object which is a source of stimulation; the fact that events (such as an image on the retina) mediate such action does not necessarily make such mediating events signs. The glass of water may of course itself become a sign; this it might prepare response-sequence of a certain behavior-family with respect to something else (say to a person who filled the glass) and so could become a sign of a certain person's kindness; the semiotician, however, must avoid making all response-sequences cases of sign-behavior if he seeks to formulate a behavioral criterion for signs in terms of response-sequences. (Morris 1949: 252-253)

Referring back to Austin's example, seeing the cheese on the table in front of your nose is not a sign, but if, for example, you live alone and haven't bought cheese in a long while, seeing a piece of cheese on the table in front of your nose is a sure sign that something is up - someone has been there and left cheese!

Robert M. Yerkes, Chimpanzees: a Laboratory Colony. Yerkes differs from Kroeber in believing that there is some evidence of slight cultural heredity in chimpanzees, but agrees that "there is no single system of signs - vocal, gestural, or postural - which may properly called a chimpanzee language" (op. cit. p. 51). (Morris 1949: 254)

The case is similar with human nonverbal communication - there seem to be "cultural heredity" of nonverbal behavior and there are different systems of vocal, gestural and postural signs, but they do not constitute a "language" because they do not accord to Morris's definition of a lansign-system (e.g. the five conditions for calling some sign phenomena a language).

The discussion up to this point has aimed to lay the foundations for semiotic within behavioristics. The intention has been to show how the basic terms of a science of signs can be formulated in terms descriptive of behavioral processes. The question now arises as to what light this approach can throw upon the issues with which inquirers in this field are occupied. The first of these is the central problem of the differentiation of the major modes of signifying. (Morris 1949: 60)

This transcends behaviorism, which in probably why it is not called semiotic behaviorism but behavioristic semiotics.

It is perhaps desirable to have special terms for the special kinds of significata involved in signs in the various modes of signifying. We will use locatum, discriminatum, valuatum, and obligatum as signs signifying the significata of identifiors, designators, appraisors, and prescriptors. The reader who fears that we are in this way peopling the world with questionable "entities" need only be reminded that these terms refer only to the properties something must have to be denoted by a sign, that is, to permit the actualization of the response-sequences to which the interpreter of the sign is disposed. So while the terminology is clumsy, and need not often be used, it is harmless enough and still behavioral in formulation. (Morris 1949: 66-67)

I quite like these terms, but I don't know how to use them and Morris himself doesn't seem to use them very often (or I simply miss them).

Emotions may be signified, and emotions may be correlated in various ways with signs as produced or as interpreted. But the signs which signify emotions need not be appraisors, and appraisors may but need not signify emotions. 'Emotion' is not an emotional term; and one can correctly interpret the signification of a poem praising dofs without oneself liking dogs, without feeling emotional when reading the poem, or without even regarding the poem as signifying that its author liked dogs. (Morris 1949: 68)

I bet this is related to the "art as expression" discourse.

This usage of 'expressive' is an application of Leibniz' more general usage to the case of signs. He writes: "One thing expresses another ... when there is a constant and regulated relation between what can be said of the one and of the other" (Montgomery, Leibniz Selections, p. 212). Expressiveness is not restricted to signs. And there can of course be signs of the state of an organism which are not expressive signs: this a blush may be a signal of something about the person who blushes, but it is then simply a signal, even if the interpreter is himself the person in question, since it is not a sign furnished by the fact of production of a sign. One must be very careful not to attempt an explanation of the nature of non-scientific signs by dragging in uncritically the term 'expressive.' It seems to me that many semioticians are guilty at this point. (Morris 1949: 257)

Very pertinent, because I am also guilty of this. But then one has to reconsider what is expressive. And should one view it in conjunction with "impression"?

Besides symbolizing a reference, our words are also signs of emotions, attitudes, moods, the temper, interest or set of the mind in which the references occur. THey are signs in this fashion because they are grouped with these attitudes and interests in certain looser and highter contexts. Thus in speaking a sentence we are giving rise to, as in hearing it we are confronted by, at least two sign-situations. One is interpreted from symbols to reference and so to referent; the other is interpreted from verbal signs to the attitude, mood, interest, purpose, desire, and so forth of the speaker, and thence to the situation, circumstances and conditions in which the utterance is made. The first of these is a symbol situation as this has been described above, the second is merely a verbal sign-situation like the sign-situations involved in all ordinary perception, weather prediction, etc. Confusion between the two must be avoided, though they are often hard to distinguish. Thus we may interpret from a symbol to a reference and then take this reference as a sign of an attitude in the speaker, either the same or not the same as that to which we should interpret directly from his utterance as a verbal sign. (Ogden and Richards 1938. The Meaning of Meaning. 5th edition. Pp. 223-224)

A sign, according to this account, is "emotive" if someone takes the fact of its production as itself a sign of some state of the producer which often accompanies the production of the sign. This as we have seen may and often does happen; we have defined the expressiveness of a sign in precisely these terms. Bit since any sign may be so considered, and since a sign based on taking the production of a sign as itself a sign of the producer is no less "referential" than any other sign, no distinction of emotive from referential meaning really issues from the analysis. (Morris 1949: 71)

Here Morris doesn't "get it". The two sign-situations that Ogden and Richards here describe seem to be the very same as (in my judment) the distinction between "conversation of gestures" and "conversation of attitudes" in Mead.

Indicators are identifiors which are non-language signals. The pointing gesture and the weather-vane may serve as examples. (Morris 1949: 76)

It's beginning to seem that Morris tried to give easier labels to Peirce's categories of signs. It does make a whole lot of sense, because pointing is indicating so the gesture may be called an indicator rather than an index after Peirce. Pointing with eyebrows, as happens in Zamyatin's We, is also a neat indicator.

By a discriminatum is meant a characteristic of an object or situation which discriminates it from other things. A discriminatum is a characteristic in that it distinguishes some object as a stimulus-object of a certain kind; it is not necessary that the stimulus-object actually affects a sense-organ or even that it can directly affect a sense-organ, but it must be such that it can have direct causal effects. A book which completely designates an object (say Africa) would signify all its discriminata, but this complete designation would not include the book itself, that is, it is not a characteristic of Africa that books are written about it. An object then may have properties which are not discriminata. Africa may be written about, liked, disliked, damned; these are not characteristic of Africa but of someone's response to Africa. Discriminata belongs to the stimulus side of a stimulus-response situation; what an organism can observe about an object or situation belongs to (though it does not exhaust) the discriminata of the object or situation. (Morris 1949: 77)

My first thought was that this might be what Donald Favareou calls "kalevis", but that doesn't seem to be the case. "Distinctive feature" hits closer to home, but this is a notion of phonetics. In broad terms, I imagine discriminata to be whatever small difference that distinguishes one gesture, facial expression or body movement or posture from another to the degree that it would call forth a different signification.

Organisms, given certain needs, prefer certain object to others. Such preferential behavior is a widespread and perhaps universal characteristic of living systems. So it is natural that it should be reflected in sign-behavior. It is believed that such preferential behavior gives the behavioral clue for the interpretation of appraisive signs. We have previously defined an appraisor as a sign which signifies to its interpreter a preferential status for something or other, that is, which disposes its interpreter to favor or react unfavorably toward this something or other. Therefore the test whether a sign is or is not an appraisor is given by determining whether or not the sign disposes its interpreter to preferential behavior to something or other. (Morris 1949: 79)

I cannot think of a nonverbal example, but it seems that descriptions of music (e.g. short reviews) for albums that are shared on the internet are such appraisors, because they "praise" the music in some way, try to convince you that it is worth downloading and listening to.

Now if we take 'preferable' to have this designative signification, then an animal may prefer objects which are not preferable, that is, do not as fully satisfy its needs as would other objects. And this is often the case. For while there is a general tendency for organisms to come through trial and error to prefer what is preferable with respect to the satisfaction of their needs, at any given time there may be a wide discrepancy between what is preferred and what is preferable; in the case of psychotic behavior the tendency to correct preferences in terms of preferability is almost completely absent. (Morris 1949: 80)

Yes, that seems to be the case for many things (from unhealthy food to many forms of criminal, unlawful or unallowed behaviour).

An object is preferred because of some of its characteristics, bit its preferential status is not itself another characteristic; the valuata of an object are not additional characteristics of the object, some among others, acting as stimuli to behavior, but are determined by the status which this object has in behavior; its preferential statuses are its valuata, and appraisors signify such a status. (Morris 1949: 81)

Hmm. I thought about the case of modern music again. E.g. some people (like myself) do not listen to certain bands or artists because they are listened to by millions and praised by too many to count. The question is, whether this is a characteristic of the band or artist or is it something else. In this case Morris suggests that it is something else - value. I believe Morris's categories make perfect sense, but it will take time and effort to understand their perfect sense.

The systemic use of signs is to use of signs to systematize (organize) behavior which other signs tend to provoke. The limitation to sign-behavior distinguishes the systemic use of signs from other attempts to organize behavior. One may, for instance, attempt to inculcate in a person by religious discourse a certain need as basic in his system of needs, and in this way to organize or systematize the needs and behavior of the individual, but such a use of signs would be incitive and not systemic as the term is here used. In the systemic use of signs the aim is simply to organize sign-produced behavior, that is, to organize the interpretants of other signs. This may be done with respect to all kinds and combinations of signs, and by use of signs in the various modes of signifying. (Morris 1949: 104)

That is, signs that organize other signs.

Through signs the individual directs his behavior with reference to things and situations which he may never have encountered and never can encounter, and yet the evidence which gives the ultimate control of knowledge must always be found in situations in which he himself behaves.
A distinction is to be made between the locus of signifying, the locus signified, and the locus of confirmation. A sign-process always involves an organism which at a given time and place interprets something. But what is signified by the sign may be signified as at a time and place other than that of the occurrence of the sign, and the confirmation or disconfirmation may be at a still other time and place. (Morris 1949: 111)

Signs can denote without the interpreter knowing whether it denotes. Things distant and time and space can be signified and knows just as the immediate environment.

The term 'communication,' when widely used, covers any instance of the establishment of a commonage, that is, the making common of some property to a number of things. In this sense a radiator "communicates" its heat to surrounding bodies, and whatever medium serves this process of making common is a means of communication (the air, a read, a telegraph system, a language). For our purposes 'communication' will be limited to the use of signs to establish a commonage of signification; the establishment of a commonage other than that of signification - whether by signs or other means - will be called communization. A person who is angry may be the occasion for another person becoming angry, and signs may or may not be the means of establishing the commonage: this is an instance of communization. Or a person who signifies anger may by the use of signs cause another person to signify anger without necessarily becoming angry: this is a case of communication. The user of signs who effects communication is the communicator and the organism in which the sign-process is aroused by the signs of the communicator is the communicatee. The communicatee may be the same organism which is the communicator, as when one writes a note to oneself to be read at a later time. The signs used are the means of communication and the signification made common by these means in [sic] the content of communication. (Morris 1949: 118)

The communication/communization distinction is essentially the distinction between conversation of gestures and conversation of attitudes, but Morris probably made the replacement so that it can apply to "organisms" rather than only humans.

It is obvious that these four comprehensive usages of signs are closerly related to the four modes of signifying. The primary use of designators is informative, the primary use of appraisors is valuative, the primary use of prescriptive is incitive, and the primary use of formators is systemic. So close in fact is this relation that doubt may be raised as to whether the mode of signifying of a sign can be distinguished from its corresponding primary usage. Thus if a sign is for a particular person at a given moment a designator, it is informing him (correctly or incorrectly) about something, and if a sign is for a particular person at a given moment a designator, it is informing him (correctly or incorrectly) about something, and if a sign is for a particular person at a given moment an appraisor it tends to confer a preferential place in his behavior to something. Hence it may seem, for instance, that the designative mode of signifying is not distinguishable from the informative use of signs, and that the modes of signifying may perhaps even be defined in terms of the primary uses to which signs are put. (Morris 1949: 96)

I would very much like to one day figure out Morris's modes andd uses of signifying and give them nonverbal examples, but for some reason the exact signification of each eludes me still.

Communication may be used to make people different as well as to make them similar, and communization or differentiation between persons may be established by other means than communication. Since sign-behavior is itself a phase of behavior, to control the sign-behavior of other persons is a powerful means of controlling their total behavior, but the control intended may be for any purpose, moral or immoral, divisive or unificative, differentiative or communizative. (Morris 1949: 119)

This could actually be a basis for a semiotic approach to power.

And in case an individual is using signs to control the behavior of someone else, signs which he believes or knows to be false or unreliable may still be more adequate for the accomplishment of his purposes than signs which he believes or knows to be true and reliable. But even here the general dependency of the adequacy of signs upon their reliability is evident in the fact that the producer of signs in such a case must use signs which the person he in influencing takes to be reliable and bust carefully geard against the communication of his own belief in their unreliability. (Morris 1949: 122)

Pretty solid contention, but I can't think of any actual examples (at least not in the novels I'm studying) at the moment.

Examples of the Major Types of Discourse

Mode ↓ / Use →	Informative	Valuative	Incitive	Systemic
Designative	Scientific	Fictive	Legal	Cosmological
Appraisive	Mythical	Poetic	Moral	Critical
Prescriptive	Technological	Political	Grammatical	Metaphysical
Formative	Logico-mathematical	Rhetorical	Grammatical	Metaphysical

(Morris 1949: 125)

Some of these make sense, even. For example, Metaphysical makes sense insofar as it forms a systemic image of the whole world, as Lotman writes.

As science advances, its statements become more purely designative, more general, better confirmed, and better systematized. Scientific discourse is therefore made up of those statements which constitute the best knowledge of a given time, that is, those statements for which the evidence is highest that the statements are true. Science is especially concerned with the search for reliable signs. The goal to which science moves is a systematized body of true statements about everything which has occurred or will occur. But since at each moment the selection of statements for admission to science is dependent upon evidence that the statements are true, the selection will vary as new evidence is obtained, and so the scientific discourse of one time may differ greatly from the scientific discourse of another. (Morris 1949: 126)

I recall Colin Cherry discussing something similar.

A person may be interested in scientific statements for their own sake (interested in collecting them as a person may be interested in collecting butterflies); a person may have knowledge and the increase of knowledge as his goal. Nevertheless, it follows from the very nature of knowledge that expectations based on statements known to be true will not be disappointed, and genetically this significance of scientific knowledge has played a large part in the development of science. It is certainly no accident that scientists at a given time are markedly concerned with getting knowledge relevant to the problems of that time; while science does not appraise or command a particular act, the knowledge that it seeks is significant knowledge, that is, information revelant to the accomplishment of various acts. (Morris 1949: 128)

I think I'm this kind of "collector". Added this to the sidebar of this blog.

Fiction in its various forms may be taken as an instance of designative-valuative discourse. In fictive discourse some sequence of happenings is designated, but whether the events happened as narrated is not, as in scientific discourse, the central issue; fiction explores an imagined universe rather than delineating the actial universe. And in fiction there is likewise a minimization of appraisal or prescreption; though the fictious characters (if the fiction be of persons) may themselves apprause and prescribe, their appraisals and prescriptions occur within the work as additional fictional elements rather than as characteristics of the dominant mode of signifying of the work itself. (Morris 1949: 128-129)

This should be kept in mind when stuidying some phenomena (whether nonverbal behavior or economic rules) in fiction novels.

Fiction abounds with heroes and villains even when it does not label them as such. It aims to induce preferential attitudes to what it designates (and often to the discourse itself) even though it does not itself specifically appraise itself or the environment imagined. The telling of the tale is to be approved and the events narrated are to be found significant; if neither result is attained, the work has failed of its purpose. (Morris 1949: 129)

Pretty harsh.

The interest in the description of a utopia may be dependent upon the belief that the imagined society is no only physically possible (compatible with known empirical laws) but that it is probably (that eventts are moving in a direction such that the imagined society will one day be actual). Insofar as the events depicted are incompatible with existing knowledge, fictive discourse becomes more and more fantastic and in time becomes absurd. This suggests that fictive discourse remains related, though often tenuously, to the actual world and actual behavior, supplying as it were an imaginative exploration of possible evnironments congenial to such behavior. Utopias are projected by persons interested in a world different from the one they inhabit, but a world different only in certain respects, namely, in those respects which would provide a more congenial environment for their actual needs. In general the reader of fiction reads those works which designate the kind of environment in which he is interested and which present this environment in a way which supports his own valuations. (Morris 1949: 130)

I'm not sure about the part about utopias, but the relatedness to actual behavior is significant for concourse.

A sign is metaphorical if in a particular instance of its occurrence it is used to denote an object which it does not literally denote in virtue of its signification, but which has some of the properties which it genuine denotata have. (Morris 1949: 136)

Now that I consider metaphoricity a part of the concursive project, I should collect these kinds of statements as well.

Signs in general serve to control behavior in the way something else would exercise control if it were present. To attain its goals the organism must take account of the environment in which it operates, select for its concern certain features of this environment, respond by response-sequences which will attain an environment suitable to its needs, and organize its sign-provoked responses into some pattern or other. Each of these stages of its activity may be facilitated by the use of signs, and the four primary usages of signs correspond to these four aspects of behavior.
Signs accordingly may be used to inform the organism about something, to aid it in its preferential selection of objects, to incite response-sequences of some behavior family, and to organize sign-produced behavior (interpretants) into a determinate whole. These usages may be called in order the informative, the valuative, the incitive, and the systemic uses of signs. These are the most general sign usages; other usages are subdivisions and specializations of these four. They are the purposes for which an individual produces signs as means-objects in the guidance of his own behavior or in the guidance of the behavior of others. They may be employed with respect to things other than signs or to signs themselves.
An individual may use signs to inform himself or others about what has been or is or will be, with respect to signs or non-semiosical events. He may use signs to confer for himself or others a preferential status upon something - upon things, persons, needs, or even signs (as where he wants the signs he himself produces to be approved as "fine writing" or "fine speech"). He may use signs to incite a particular response in himself or in others to objects or signs, or to call out submission in some one else, or to get the reply to a question which bothers him, or to provoke co-operative or disruptive behavior in the members of some community. And he may use signs to further influence behavior already called out by signs, whether this behavior be to signs themselves or to something other than signs. (Morris 1949: 95)

I have made myself rewrite so many passages from this incredible book that some post-it marks have gotten misplaced and quotes skipped because of this. Luckily I noticed this. These explications make his types of discourse much more understandable. They also suggest that building a semiotic theory of power in terms of "behavior control" may be feasible.

Appraisive-incitive discourse may be illustrated by the language of morality. For discourse that appraises actions as favorable (or unfavorable) from the standpoint of some group, and aims to incite (or inhibit) these actions, is certainly close to what is commonly recognized as moral in character. (Morris 1949: 138)

Seems right. "Let me tell you how you have to live your life..." etc.

From this point of view moral discourse is linked with the sphere of social behavior. We have seen that groups of organisms may have goals in whose realization individual organisms play certain specialized roles. Further, these individuals may, in ways suggested by George Mead, come to symbolize the roles of other individuals in social behavior and the needs of the group, and to appraise their own behavior or the behavior of other individuals with respect to the needs of the group. In terms of the perspective of social behavior the individual in this way gains a basis for appraisals in which he can even approve or disapprove of certain of his own preferences and actions. Appraisals made of oneself or others in terms of what is conductive to group welfare, when incitive in aim, constitute moral discourse. The term 'ought,' so common in moral discourse, is (when a moral 'ought') evidence of the dual appraisive and incitive character of moral discourse, for 'ought' signifies some action that is positively appraised, and is used in contexts which make clear that the user of the sign is endeavoring to incite the action in question. (Morris 1949: 139)

Hm. It seems that because there are no tenses in nonverbal behaviour, this might be one category which cannot be easily reconstructed for my purposes.

Moral discourse is of importance in the social control which it exercises over the conduct of individuals. To the extent that it is effective it couses individuals to act in the light of the effects of their actions upon other individuals of the social group to which they belong. To the degree that moral appraisals are built upon reliable knowledge of group needs and ways of meeting these needs, moral discourse is an important agency in advancing the consummation of group interests and the interests of individuals in a group. (Morris 1949: 140)

Relevant for sociosemiotics.

The adequacy of scientific statement is tested by finding further data which inclease or decrease its confirmation; the adequacy of criticism is tested by additional criticism. The situation is complicated by the fact that criticism may be carried on for a great variety of secondary purposes (informative, valuative, and incitive), and in this case further questions of adequacy arise with respect to the way it accomplishes these purposes. (Morris 1949: 142)

One of such secondary purposes was proposed by Oscar Wilde - that all criticism is autobiographical.

Political discourse, in common with most types of discourse, is both an agency for social conservatism and for social reconstitution. It is a tool through which social behavior in its widest recahes aims to induce approval of certain institutional forms for the realization of its widest social goals. Its adequacy is to be measured in terms of its effectiveness for advancing the purposes for which it is used. (Morris 1949: 146)

From this standpoint very few political discourses are adequate.

The term 'function' is often employed in place of the term 'usage,' but - aside from the fact that the term has another signification in logic and mathematics - is wider than the present intended signification of 'use.' It is often said that one of the "functions' of signs is to make possible the development of the complex human personality; but signs may have this and other consequences without being used, in the present sense of the term, as means to attain these consequences. To "use" signs is to employ them as means-objects. (Morris 1949: 261)

Very relevant, because lotmanian semioticians often use the word "function" in a too wide sense. I think the contention about "means-objects" is again related to intention - signs are "used" only intentionally. I'm not at all sure about this, but it seems to be the case.

See T. C. Pollock, The Nature of Literature; C. A. Mace, 'Representation and Expression,' Analysis 1, 1934, 33-38... (Morris 1949: 261)

I will see.

We have considered at some length the conditions which a stimulus has to meet to be an identifior, designator, appraisor, or prescriptor. Let us henceforth call all such signs lexicative signs, or lexicators. Then in terms of our previous discussion of the term 'formator' we may say that if a stimulus is not a lexicative sign, but nevertheless influences in a uniform way the total signification of the particular sign combinations in which it appears, the stimulus in question is a formator. This formulation, however, leaves open the questions whether formators are signs, whether and what they signify, why they occur, and what - if any - feature of language are formators. (Morris 1949: 153-154)

Indeed I still can't make out what are lexicators and formators, exactly.

In the face of such diversity it is reasonable to surmise that the differences are not factual but arise in part from the acceptance of different terminologies in which to talk about signs. Here the semiotician must be careful to take upon himself the advice he likes to give others: Do not confuse terminological and factual problems! (Morris 1949: 154)

Sounds like good advice.

...that the term 'table' in French requires certain endings in certain sign combinations shows nothing about the characteristics of tables. (Morris 1949: 160)

An example of this advice.

Modors are formators which establish the ascriptive mode of signifying of the sign combination in which they occur (that is, determine whether the total interpretant composed of the interpretants of the other signs in the given sign combination is to be the interpretant of a designative, appraisive, or prescriptive ascriptor). 'He is coming' may be spoken in such a way that the ascriptor is a statement, an appraisal, or a prescriptor - signifying merely that an event does occur, or signifying that his coming is good or bad, or signifying that his coming is inquired about or demanded. The intonations and speech melodies which effect these differentiations are modors. In writing, the corresponding modors are the punctuation devices '.' '!' '?' '!!,' though there is no sharp standardization here of appraisive and prescriptive devices. (Morris 1949: 161)

Here it seems that Morris is merely inventing exact terms for phenomena that already have names.

...'is' comes from a word which signified "to grow"... (Morris 1949: 161)

If this etymology is true then it opens up a whole new world for frenchy philosophers to usurp (if they have not already done so).

If there was nothing existent there would be no denotation - and no truth, factual or formal. If by some miracle there wele still ascriptors it would be true that a given ascriptor was formative, and true that it would be true if its antecedent ascriptors were to denote. But true it would not be. And if there was no evidence, no ascriptor, formative or lexicative, would be known to be true. But there is a world and it furnishes evidence as to whether our signs denote. And if there was no such world, then there would be no signs, and no knowledge, and no truth. Not even the truth that there was nothing. The reader is free at this point to utter any appraisor he desires. I am satisfied to smile. (Morris 1949: 168)

Sweet. Smile is an appraisor!

The ascriptor as a whole is accordingly an analytic formative ascriptor, differing only in complexity from such an ascriptor as 'That red thing is colored.' The same is true of '2+2=4.' The analysis of this ascriptor shows that in a certain language the significations of the signs in the sign combination are such that any situation which meets the conditions of denotation of '2+2' thereby meets the conditions of denotation of '4,' and vice versa. '2+2=4' is, therefore, in the language in question an analytic formative ascriptor. (Morris 1949: 169)

Seeing as this sign is present in Dostoyevsky and Orwell, I find it quite pleasant.

These examples are intended to reinforce the point already made that formative ascriptors are subject to the same variety of uses as are other ascriptors. They may contain designative, appraisive, and prescriptive signs along with the formators which give them their unque character, and thy may be used to inform, provoke valuations, and incite actions both respect to linguistic and non-linguistic matters. A developed and comprehensive semiotic must shake off the prevalent tendency to concentrate undue attentiot upon the designative mode of signifying and the informative use of signs. (Morris 1949: 175)

I wonder if I could supplement discussion of regulators with the category of incitive signs.

A distinction is current today between an "object language" and a "metalanguage," an object language being any language which is an object for investigation, and a metalanguage being any language signifying some other language; in this usage the terms do not refer to two kinds of language but to a relation between two languages in a given investigation. So if we talked about French in English, French would be the object language and English the metalanguage. Now this distinction itself, as often interpreted, has certain defects. The first of these, the restriction to language, can easily be avoided by recognizing that there may be metasigns (signs about signs) which denote object signs which are not language signs. Hence an object language and a metalanguage are then simply special cases of objectsigns and metasigns (that is, where these are language signs). The second defect is more serious, and springs not from the distinction itself but from the almost universal neglect of the great variety of types of discourse: it is uncritically assumed that a metalanguage is either scientific discourse or logico-mathematical discourse. A metalanguage, however, might be of any type of discourse and its object language of any type of discourse: there can be poetic discourse about poetic discourse, scientific discourse about poetic discourse, legal discourse about poetic discourse, poetic discourse about scientific discourse, logico-mathematical discourse about scientific discourse, scientific discourse about logico-mathematical discourse, and so on. Therefore the term 'about' in the phrase 'signs about signs,' is as wide in signification as the term 'signifies,' and signs in any mode of signifying can signify ("be about") signs in any mode of signifying. So we either have to recognize explicitly that any type of discourse can be metalinguistic or restrict the term 'metalanguage' to some specific type (or types) of discourse; we propose the former usage. (Morris 1949: 179-180)

Actually, the category of metasigns might become useful when we consider concursivity or "verbal signs about nonverbal signs" (presuming that concourse is about nonverbal signs that in themselves are about nonverbal behaviour).

Semiotic as an empirical science eventuates in designative-informative discourse about language and non-language signs; semiotic as logic eventuates in formative-informative discourse about languages. Semiotic, then, has its scientific discourse and its logic, nad these are no more in opposition than experimental and mathematical physics. (Morris 1949: 182)

The trouble is, modern semioticians rarely uses logic, and "empirical semiotics" is still rarer.

And now a word of explanation. I should feel somewhat disappointed if the reader does not himself experience a sense of disappointment with this chapter, and indeed with the entire treatment of the various types of discourse. Such misgivings, however, are but witnesses to the astonishing complexity of sign phenomena and to the corresponding difficulty of divising a language in which to talk about such phenomena. Our account at least shows that semiotic cannot simply take over the current terminology employed in talking about signs. For these terms are riddled with ambiguities and inconsistencies, as the incomplete analyses of 'science,' 'logic,' 'mathematics,' 'grammar,' 'rhetoric,' and 'metaphysics' has shown. The inconclusiveness which most readers will feel in the analysis comes from the fact that different analyses are possible, each of which is defensible in terms of an appeal to some sement of historical tradition. (Morris 1949: 185)

Some self-criticism. Indeed, for me, this chapter was the most difficult. Thankfully he returns to discussing communication and more down-to-earth sign-phenomena.

EVERY sign involves behavior, for a sign must have an interpretant, and an interpretant is a disposition to a response. But the behavior which is involved in the sign as sign itself occurs within the context of the more inclusive system of behavior of the interpreter of the sign, and frequently within the context of a behavioral system involving a number of organisms. There arises in this way the problem of the relation of signs to the individuals and societies in which signs appear and operate. This topic has been touched upon already in a fragmentary way in the discussions of the genesis of language and the uses of signs. But there the problem was incidental and not focal. Nothing has yet been said in any comprehensive way as to the individual and social conditions for the appearance of signs or the effect of signs upon the individual personality and upon society. And here a very extensive field for semiotic opens up. Not merely do signs have a certain signification at a given moment, but they have this signification only within the particular life history of their interpreters; and their appearance affects for good or ill the further life history of these interpreters. (Morris 1949: 187)

An important topic. I caught myself thinking about how all of this comes together with my plans to tie Morris's behavioristic semiotic with Powys's self-culture and literature - I am, after all, studying nonverbal behaviour in dystopian literature. The model I imagine at this moment goes from: (1) actual nonverbal behaviour of actual people; to (2) signs of nonverbal behaviour mediated by language, e.g. concourse; to (3) culture as the overarching system of sign systems; which (4) bears again back on actual nonverbal behaviour of actual people.

Every sign involves behavior, for a sign must have an interpretant, and an interpretant is a disposition to a response. But the behavior which is involved in the sign as sign itself occurs within the context of the more inclusive system of behavior of the interpreter of the sign, and frequently within the context of a behavioral system involving a number of organisms. There arises in this way the problem of the relation of signs to the individuals and societies in which signs appear and operate. This topic has been touched upon already in a fragmentary way in the discussions of the genesis of language and the uses of signs. But there the problem was incidental and not focal. Nothing has yet been said in any comprehensive way as to the individual and social conditions for the appearance of signs or the effect of signs upon the individual personality and upon society. And here a very extensive field for semiotics opens up. (Morris 1949: 187)

I believe it does. It is revelant that Dewey dismisses Morris's personal signs and post-language symbols on the basis that "only error individualizes".

A proper understanding of the way signs operate in individual and social behavior demands an explicit recognition of the importance of signs other than those produced by the vocal chords and heard by the ear. Spoken-heard signs are so central in the life of man, and offer such opportunities for study, that they have almost dominated the interest of semioticians. Such signs alone are often called "language," and since man has been regarded as the linguistic animal, it has seemed natural to concentrate on them. But this, if carried too far, is a great error, and one from which semiotic must now free itself. For not only does the spoken-heard language depend in crucual places on other signs, and itself give rise to signs which are not spoken or heard (post-language signs), but such language is not coextensive with all signs drawn from sounds and has never supplanted its great rival - visual signs. (Morris 1949: 190)

Morris's version of going against logocentrism.

A light which signals food to a dog is as "primitive" as a sound signal; and interpersonal relations are as much determined by the signs gained by the sight of other persons (manner of dress, gestures, facial movements, physical appearance) as by the sounds he utters. I recall once in Poland an elaborate communication in which not a sound was uttered: a young boy and his father stood near me on a crowded train; the father moved down the car without the boy noticing this movement; the boy then looked for his father, did not see him, became panicky; I touched the boy on his arm, pointed to his father, and the boy ran off, happy; his father smiled, and tipped his hat, and I smiled back. The situation was "clear" to all of us, replete with signs, and soundless. Although what is "perceived" in such institutions is certainly in part due to the prior occurrence of spoken language, such facts do not warrant the general conclusion that visual signs are less primitive than auditory signs. (Morris 1949: 191)

This is the passage where Morris reflects on nonverbal communication most explicitly. The note for the last bold part goes as follows: "Ernst Cassirer has stressed this point. G. H. Mead argues that the perception of objects as enduring is not possible without language ("Concerning Animal Perception," Psychological Review, 14, 1907, 383-90). For material on social perception, see F. Heider, "Social Perception and Phenomenal Causality," Psychological Review, 51, 1944, 358-74." I'll take these into consideration.

In the case of "realistic" painting and "program" music at any rate it seems clear that recognizable objects (such as painted chairs or persons, or the tonally "painted" portrait of an object or person) furnishes a vocabulary of signs which are then combined "grammatically" in various ways according to the style of a particular school or artist. It is true that such icons may become very general, as in the case of the "formal" or "automorphic" kinds of painting and music, but generality of signification is not the absence of signification. I have asked many persons, for instance, what kind of situation Stravinsky's Rite of Spring might denote (that is, what is its signification). The answers are various: a herd of wild elephants in panic, a Dionysian orgy, mountains being formed by geological processes, dinosaurs in conflict. But there is no suggestion that it might denote a quite brook, or lovers in the moonlight, or the self's tranquility. "Primitive forces in elemental conflict" - such is the approximate signification of the music, and such conflict is presented iconically in the music itself. (Morris 1949: 193)

This is the only western author so far that I have seen mentioning "program music". I thought this was a notion only in soviet musicology. Turns out it's a notion of early twentieth century musicology.

Iconicity itself is not an absolute criterion of the fine arts, for a nover may be as much a work of art as a painting, and a painting of an object may be scientifically reliable and yet esthetically bad. The common feature of the fine arts of various linguistic media would seem to lie primarily in their valuative use of signs which signify goal-objects, with the additional requirement that the way the signs are employed must awaken a positive valuation of themselves as goal-objects (that is, be at least a part, and perhaps in the limiting case the whole part, of their valuative use). No sign is as such "esthetic" and the attempt to isolate the fine arts by isolating a special class of esthetic signs seems now an error. (Morris 1949: 195)

Very relevant for the semiotics of art.

Speech, when attained, itself makes possible a vast extension of sign-processes: it may, as we have seen, give to perceived objects a signification first attained by language, and it may give rise to languages (such as that of the deaf and dumb, or the highly devoted fine arts) which would not otherwise appear. All such signs are in a large sense of the term "post-linguistic" in that they are dependent upon language for their appearance. (Morris 1949: 196)

This implies that secondary modelling systems in the sense of cultural semiotics are nothing more than post-linguistic systems.

If we mean by 'freedom' the ability of an organism to direct its behavior by signs, then the highest degree of freedom is found in those organisms in which post-language symbols have attained the highest level of development. Mead stresses in a number of places that the significant symbol (which, as we have seen, covers our language signs and post-language symbols) allows man to become a "self-conditioning" being, a fact which he thought the existing doctrines of the "conditioned response" failed to account for. The present account can incorporate and explain such self-conditioning. For the person who is planning a shopping tour can by language and post-language symbols signify to himself the consequences, say, of going first to the store rather than another, and through such signifying "condition" himself to respond to the door of the place in question, determining whether he will enter it or pass it by when he encounters it. (Morris 1949: 197)

Very much related to autocommunication. If it be taken for granted that personality develops through communication, then the notion of self-communication implies that the person can craft his or her own personality (and that is exactly what we humans do). The shopping example is also eerily familiar from Mihhail Lotman's interview on autocommunication. His example was only slightly different: that in making a shopping list one self-conditions (he did not use this term) to select items purposefully rather than buy needless junk.

The "totalitarian" societies of today are highly integrated, and in their extreme form attempt to suppress any form of individual behavior that does not support the goals of the society; (Morris 1949: 204)

Orwell's 1984 appeared approximately at the same time as this book. In Orwell's totalitarian nightmare, even sexual relations are suppressed, because they form loyalties outside of the party.

A sense relevant to our inquiry would be one in which the sphere of the cultural is less wide than that of the social. In all societies, for instance, sexual interactions occur, but they occur differently in different societies, and it is in such differences that differences of culture consist. Not sex relations but the form of sex relations of a given society is cultural; (Morris 1949: 205)

This sounds familiar. It is biological/universal that peaple have sex, but the manner in which they have sex is social/cultural.

But if cultural phenomena are not necessarily sign phenomena, it may still be true that signs play a very important role in culture and in its transmission. And on this point no doubt it possible. For the language in which people talk, the rites the perform, the monuments they erect, the works of art they make, the devices they utilize to indicate social prestige are all cultural phenomena, and all sign phenomena. Fer this reason it is possible and revealing, as Chapple and Coon have done in their Principles of Anthropology, to interpret much of the data with which the social scientist deals from a semiotic point of view. In Parts IV and V of this book (entitled "Symbols and Human Relations") they consider rites of passage, rites of intensification, magic, religion, art, money, and laws as "series of symbols and their associated techniques which have meanings held in common by a number of persons," and attempt to differentiate these various cultural phenomena in terms of the signs employed (regarding a law, for instance, as "any rule symbolizing a pattern of interaction which applios to all the members of a group, regardless of institutions"; and a ritual as "a symbolic congifuration used to restore equilibrium after a crisis"). This treatment goes far to show the important place which signs have in cultural phenomena, and to suggest the power of semiotic as an organon for the study of man. The semiotic used by these authors is much more adequate than the vague terminology in this field still employed by most social scientists, but though behavioral in its approach, their semiotic is still far from adequate for the purposes of a study of human culture (a sign, for instance, is regarded by them as anything "which sets off a conditioned response"). It is believed that the refinements and distinctions in the science of signs which have been made in this and previous chapters will be useful to the study of man as a cultural being, and that as they are so used in this study the results obtained will in turn greatly help to advance semiotic beyond its present formulation. (Morris 1949: 206)

If I intend to unite Morris with Powys, I'll have to consider this passage in detail.

The socially derived techniques by which the individual carries over his self-control the forms of social control are so important that every society is especially concerned with the determination of the signs of its individual members. As Mark May has suggested in conversation, such concern is the distinctive feature of propaganda in all its forms; in the widest sense of the word it is the mark of educational institutions. Through its control over what can be said by school and print and film and stage, society attempts to secure for itself the ultimate control over the sign-processes of the individual, and in this way to control its individual members through the signs which will operate in their behavior. The enormous importance in the modern world as to who will control tho agencies of mass communication becomes evident. For even if explicit appraisals and prescriptions are not made through such agencies, the mere control of what information is made available to individuals will go a long way toward determining the nature of their own appraisals and prescriptions and hence their behavior. (Morris 1949: 208)

This may be one of the reasons Powys felt negatively inclined towards television. Also, compare this with Kecskemeti.

Social control of the individual through the control of his sign-processes is not therefore complete. Nevertheless it can be carried very far, and the extent to which it is attempted and felt desirable will differ with various cultures. A totalitarian society pushes this process to its limit; a loosely organized society will reveal a plurality of various centers competing for control of as many individuals as they can reach; a truly democratic society would aim, as a matter of principle, to enlarge and diversify the sign capacities and resources of its members, assuring them access to a vide area of information and encouraging them to test and improve current appraisals and prescriptions in terms of their reliability and adequacy. (Morris 1949: 209-210)

An individual still retains at least some semiotic freedom. Winston and D-503 write a diary, Montag reads books, etc.

In a healthy society signs in all modes of signifying would fulfill the various uses to which signs can be put, and each specialized type of discourse would lend its support to the other types and be in turn controlled by them. Such a society would ground its basic appraisals and prescriptions upon its actual needs and keep them linked to advance in knowledge of itself and its environment. In this way it would seek to reconstruct its preferences and techniques in terms of its knowledge and its needs, and would support all investigations which gave it knowledge relevant to its needs; its sign-processes would flexibly express its needs and aid in their constant reconstruction and redirection; need, knowledge, appraisals, and prescriptions would mutually support and influence each other. At no one time would its signs be completely reliable or adequate, but conditions would exist such that the way was open for the progressive improvement of their reliability and adequacy. (Morris 1949: 210)

Now look who's being utopic!

That communication is essential to complex social behavior, and that it primarily serves to increase integration and socialization, is not to be denied; it is well to remind ourselves, however, that it does not always do so.
One indication of this is in the special form of communication in which an individual communicates with himself (that is, the self of one moment communicates with the self at another moment). This occurs not merely in the writing of diaries or the devices by which the present self acts to remind the future self of something, but takes place in a peculialy important form in the production of a work of art. For in such a production the artist throughout the process is stimulating himself by the stimuli he produces, and at the end of the process in particular he stands over against his work as a member of his audience. There is self-communication in so far as the self is interpreter of what it signified as sign producer; there is social communication insofar as the communication involves interpreters other than the artist.
Now in both these cases the fact of communication may play an integrative role in the selves involved, but need not do so. Let us borrow a term from Abraham Kaplan and call a sign "self-expressive" if the producer of the sign interprets the fact of its production as expressive of himself. The artist may then through the work of art come to a knowledge of the factors in himself that led to the production of the work of art, and by so signifying these factors be better able to incorporate them into a more adequate integrated self. But it is also certain that this may not take place: he may not interpret his products as self-expressive signs but simply enjoy their signification at successive moments of self-expression. (Morris 1949: 213)

Communication may sound as if it was only about making something common, but it can also be used for making something uncommon. This is one of the primary functions of self-communication, wherein one "increases integration" with oneself. Also, the Jakobsonian example of autocommunication through time (communicating with that "other self" in the future) reaches back to Peirce, who claims that we give signs to ourselves in the future (his signs "grow"). Also, the bit about self-expression related to self-observation in dreams.

The increase of communication may not only fail to give agreement in valuations and modes of conduct but may actually be used to increase conflict, competitiveness, and slavery. For sharing a language with other persons provides the subtlest and most powerful of all tools for controlling the behavior of these other persons to one's advantage - for stirring up rivalries, advancing one's own goals, exploiting others. Modern propaganda is the witness to this within existing nations; a world language would make the same phenomena possible over the earth as a whole. And semiotic itself, as it develops, will be subject to the same kind of utilization by individuals and groups for the control of other individuals and groups in terms of self-interest.
Such conclusions are in themselves no cause for pessimism. To see danger is to be forewarned; the common and uncritical optimism that any increase in communication is necessarily good is to court the very dangers in question. A world language, the internationalizing of the arts, the reading of the same books - these things are under way and will not be stopped. If they may be powerful forces for the liberation of more and more individuals and for the integration of diverse communities into an earth community, they may also be powerful forces for the mechanization of the individual and the enslavement of society. Which they will be, and how much one and how much the other, will depend upon by whom they are used and for what ends. (Morris 1949: 214)

This sounds like Jakobson's contention that language is the highest (subtlest) means of conveying information. I think English has become such word language.

Cf. John Dewey's remark that "probably a time will come when it will be universally recognized that the differences between coherent logical schemes and artistic structures in poetry, music and the plastics are technical and specialized, rather than deep-seated," (Philosophy and Civilization, pp. 120-121).
In my article "Esthetics and the Theory of Signs," Journal of Unified Science, 8, 1939, 131-150, I tried to differentiate the esthetic sign as an iconic appraisor. The present position is more general, since the approach to the arts in terms of the valuative use of signs does not require that the signs are nevertheless of great importance in the arts, and the article mentioned is still relevant to the general problem of art considered as a sign phenomena. It was discussed by the literary critics, Allen Tate (Reason in Madness) and John Crowe Ransom (The New Criticism). For supplementations as to the relation of the arts to signs see G. Kepes, Language of Vision, and Ernst Kris "Appreaches to Art" (in Psychoanalysis Today). Kepes aims to show that modern painting is a discovery of new ways of representation; Kris (as does Dewey) interprets art as a form of communication. (Morris 1949: 274)

"Esthetics and the Theory of Signs" is republished in Morris's Writings....

See on the general question my article "The Mechanism of Freedom" in Freedom, It's Meaning (ed. Ruth Anshen); also G. H. Mead The Philosophy of the Act. A semiotical formulation of the term 'freedom' overcomes the opposition of freedom and causation by phrasing the problem in terms of the extent to which an organism causes its own behavior through the operation of signs. One who takes the position of the present volume is not surprised, for instance, that a person can cause the dilation of the pupils of his eyes by producing language or post-language signs which signify the objects which originally caused this dilation. (See the article by C. V. Hudgins, "Conditioning and the Voluntary Control of the Pupillary Light Reflex," Journal of General Psychology, 8, 1933, 3-51.) (Morris 1949: 275)

I quite like this definition of freedom, because it hinges on what I think of as autocommunication. Also, I cannot find Hudgins's paper, but it's good to know that the voluntary control of pupil dilation was not discovered in relation with sexual arousal, but in the 1930 and in relation with behavioristic conditioning.

What the Freudian has done is to propose a theory as to why the individual finds it difficult to formulate the signification of certain of his signs and indeed why he actively resists such a formulation by himself or others. This theory, in so far as it is sound, is a contribution to the understanding of pathic signs; there is nothing in it which in principle cannot be translated into the terminology of a behavioral semiotic. The Freudian symbols are essentially general icons and as such capable of denoting objects which are like themselves in certain respects only 9dreams of flying as symbolic of the erect penis, dreams of open books as symbolic of the female genitalia, etc.); they are a special case of metaphorical signs, namely, where certain processes in the individual prevent or make difficult the recognition that the metaphorical signification provides partial satisfaction of a frustrated desire. (Morris 1949: 276)

Somewhat relevant for somatoception.

See especially his [Mead's] paper "The Genesis of the Self and Social Control" (International Journal of Ethics, 35, 1925, 251-77). (Morris 1949: 277)

In relation with role-taking.

A table may be brown, weigh ten pounds, and be inflammable; an organism, in addition to characteristics similar to these, may be angry, have after-images, feel anxious, dream, respond to events in itself an in the world as signs. (Morris 1949: 228)

Yeah, a living organism is different from an inanimate object. But, actually, this is Morris's way to approach the semiotic treshold.

Since an organism is itself a complex interrelated set of structures and processes, its characteristics may in turn be stimuli to its own responses: a person may look at his own foot, try to relieve his anxiety, talk about his dreams. (Morris 1949: 228)

This is actually relevant: although animals can also dream, they cannot talk about their dreams. And with "stimuli to its own responses" he once again affirms self-communication/autocommunication.

A third problem that is central to any program for the systematization of knowledge is presented by such humanistic material as literature, art, morality, and religion. Practically all the work of teachers and students in a "Division of the Humanities" in a university consits in talking about discourse of various kinds: about effective speech and writing, about paintings, about musical compositions, about works of literature, about systems of morality, about religious documents. Let us call such objects signs (together with the activities which produce them) the humanities; let us call the metalanguage about the humanities humanistics. Once this simple distinction is made we are well on our way to resolve the problem of the relation of the "Geistwissenschaften" to science. (Morris 1949: 230)

I can retain the "entertaining character" in my work because it is not scientific but humanistic.

The term "philosophical" might not distinguish a type of discourse but apply - like such a term as 'witty' - to any discourse, perhaps in proportion to its generality or to its comprehensiveness; (Morris 1949: 234)

This seems valid. Instead of wittiness we can also use the sign "wisdom".

Only the individual who utilizes the signs of the artists, the prophets, and the philosophers, as well as the information given to him by the scientist, is living at the level of a complex individual. To show that signs other than scientific play a basic role in life, and yet to do this in a way other than scientific play a basic role in life, and yet to do this in a way which does not minimize in the least the unique importance of science, is perhaps the most important single task which semiotic can today perform. (Morris 1949: 240)

Morris himself also wrote poetry. And this contention - that a "cultured" individual is well versed in both science and arts - seems compatible with Powys's The Meaning of Culture.

From the cradle to the grave, from awakening until sleep, the contemporary individual is subjected to an unending barrage of signs through which other persons seek to advance their goals. He is told what to believe, what to approve and disapprove, what to do and not to do. If he is not alert, he becomes a veritable robot manipulated by signs, passive in his beliefs, his valuations, his activities. Through post-hypnotic suggestion an individual can be caused to perform actions suggested to him without realizing the source of his actions, and with the feeling that he is acting as a free agent. (Morris 1949: 240)

Links up with advertising semiotics, I think.

When an individual meets the signs with which he is confronted witha knowledge of how signs work, he is better able to co-operate with others when co-operation is justified. If he asks himself the kind of sign he encounters, the purpose for which that sign is used, and the evidence of its truth and adequacy, his behavior shifts from automatic response to critical and intelligent behavior in which he himself acts as a responsible and spontaneous center. He becomes autonomous human being, neither unduly suspicious nor unduly gullible, a center of life and not a hypnotized animal. (Morris 1949: 240)

I wonder how this suggestion pans out alongside with Powys's self-culture.

An organized human society rests on a common body of beliefs, preferences, and modes of action. Through the signs which reflect this commonage the society gairs its main control over its individual members, insuring at crucial points their participation in the social behavior characteristic of the society. In crisis situations there arises a focal concern for strengthening this control over the individual so that the society may become more powerful and efficient. This process appears in its extreme form in the totalitarian regimes of the present. (Morris 1949: 242)

And in some dystopic works of literature this reaches the absolite, where everything is under social control, arranged by a time-table and monitored at every step (I'm thinking of Zamjatin's We).

To affect the signs of individuals is to bind them by the most powerful chains which men have devised or to place in their hands the most powerful of all instruments for individual liberation and social reconstitution. Social control of individuals through their sign-processes is inevitable, and the possibilities for such control will become ever greater as the knowledge of signs and the techniques of communication develop. The portentous question is how such control is to be exercised. (Morris 1949: 244)

At this point it is almost saddening that my field of expertise in this regard is limited with nonverbal behaviour, because I believe that the social control of the future lies mainly in information technology.

The totalitarian society will give no widespread attention to semiotic in its educational plans for the total population, for knowledge of sign phenomena makes it less easy to manipulate by signs those who have this knowledge. But precisely because of this fact semiotic should have a prominent place in the educational system of a democratic society. (Morris 1949: 244)

I've heard this argument from Katre Pärn. She also emphasized that semiotics will also ruin some enjoyable forms of manipulation.

On its descriptive side semiotic must make far more use than it has fet done of the data on signs which are already available. This material is widely scattered in the studies of psychologists, psychiatrists, social scientists, linguists, and estheticians. It is often in a somewhat hidden form, gathered incidentally in research which was not focused on sign phenomena. But present developments of semiotic furnish a language into which this data can be translated and organized, and wisdom suggests that this be done. Such data require supplementation, however, by more careful descriptive studies directed specifically to securing knowledge of sign-phenomena. (Morris 1949: 246-247)

Very likely the source for Umberto Eco's distinction between explicit and implicit semiotics.

...the Sceptics questioned the whole edifice of metaphysics on the ground that signs could refer only to that which was observable, serving to recall (as "commemorative signs") that which had been observed even though it was not at the moment of reference directly observable. (Morris 1949: 286)

I wonder if "commemorative signs" could be used to tie Kantor's theory of memory with semiotics?

Or again: "As representation is that character of a thing by virtue of which, for the production of a certain mental effect, it may stand in place of another thing. The thing having this character I term a representamen, the mental effect, or thought, its interpretant, the thing for which it stands, its object"; "A sign is a representamen of which some interpretant is a condition of a mind. Signs are theonly representamens that have been much studied."
This material is quoted to illustrate the difficulties which appear when we leave the ground of behavior situations in attempting to define 'sign.' For if 'sign' is defined in terms of mind or thought, then we cannot furnish an empirical criterion for determining whether a certain thing isor is not a sign until we have a satisfactory criterion for the occurrence of mind or thought. It seems that Peirce does not give such a criteruon in a way usable by science. It should be noted that Peirce's own various formulations differ considerably. At times he writes as if "every thought is a sign"; at times merely that thinking never occurs without the presence of something which serves as a sign. And if 'interpretant' is at times defined in terms of "mental effect or thought," the fullest accounts of it are couched in terms of habit. The present treatment follows Peirce's emptasis upon behavior rather than his more mentalistic formulations. Hence it not only avoids the extension of sign-process to inorganic nature, but does not require that all behavior involves sign phenomena. (Morris 1949: 289)

I get a warm and fuzzy feeling whenever Peirce is quoted. Mostly because every time it appears as something different, even if the quote is so familiar you may know it by heart. Here the latter statement, that "Signs are the only representamens that have been much studied." implies that there are other representamens aside from signs, andd yet we are often suggested to think of "representamen" whenever we meet the word "sign" in Peirce. So what are the other representamens? I am not bothered one bit that Peirce doesn't seem to give Morris a criterion for scientificity. The seemingly conflicting statements he has chosen to represent Peirce's flopping don't seem to be mutually exclusive. That is, every thought may be a sign AND thinking never occurs without the presence of something that serves as a sign; similarly, interpretants have three varieties and "mental effect or thought" is one of them. Just as we can talk about habits of thought we can also talk about habits of action and habits of feeling.

There is another point to be mentioned. Peirce almost always defines 'sign' in such a way that the intpretant of a sign is itself a sign, and so ad infinitum. One such formulations is the following: "A Representamen is a subject of a triadic relation to a second, called its object, for a third, called its Interpretant, this triadic relation being such that the Representamen determines its interpretant to stand in the same triadic relation to the same object for some interpretant." Or more succinctly: "It is essential to a representamen that it should contribute to the determination of another representamen distinct from itself." This seems to confuse the problem of the definition of 'sign' with the empirical question as to whether signs always generate new signs. Unless there are distinguished a circularity is introduced into the definition, for a sign is defined as something which generates a sign which generates a sign, etc., and this is certainly objectionable as a form of definition of 'sign' itself. Signs, at least at the human level, do frequently generate a series of sign-processes, but I see no reason why this fact about signs should be incorporated into the definition of 'sign' itself. (Morris 1949: 290)

Apparently Morris also didn't understand the notions of infinite semiosis and sign-growth. Thankfully Jakobson can fill this in.

Similarly, few behavioristicians would deny that a person can observe himself, or that some experiences are accessible to self-observation in a way not accessible to other observers - after-images, pain, and dreams would be cases in point. And certainly the semiotician is not called upon to reject observation of his own sign-process or the reports of others of their observation of their sign-processes. The theoretical question is how what is observed in self-observation is to be described. To enter into this conflict is to assume the burden of the psychologist. (Morris 1949: 300)

Again relevant for self-communication, because there are private experiences that are accessible to self-observation and those that aren't. Dreams are suspended on the edge of this distinction, because we can remember some dreams and not others, or remember them partially or even invent parts of them in waking state. Subjectivism in semiotics is a tough nut to crack.

The fact that the theory of signs has for thousands of years been couched in such terms without attaining a scientific status should raise strong doubts about their continued use as primitive terms for semiotic. On the other hand, the study of behavior has been developing cumulatively in recent decodes and is becoming an advanced experimental science. Hence the naturalness of seeking in it the definition of the basic terms of semiotic. Whether behavior theory is or is not psychology is not in this respect an essential problem for the semiotician.
It is for these reasons, and these alone, that we have sought to link semiotic with the science of behavior. Hence we have interpreted its basic primitive terms in words used to describe behavior situations. Others can if they wish not define these terms at all, or define them in mentalistic terms; in the first case no natural science results, in the second case one will have to see what significant results are obtained - and history does not suggest much ground for optimism. (Morris 1949: 300-301)

Alas, Morris developed behavioristic semiotic because behaviorism was considered the cutting edge of science at his day.