To appear in The Scandinavian Journal of Psychology


John Shotter

Department of Communication

University of New Hampshire

Durham, New Hampshire


Abstract: Jan Smedslund's proposal to establish a psychologic as an a priori formal system for use, in this case, to establish a system of axioms in terms of which to analyze the basic concepts used in psychological research into memory, is criticized. He claims to be making explicit parts of an implicit calculus built into language which enables people to predict and understand each other's behavior. However, in working in terms of the selection of regularities upon which there is a consensus, thus to create to create amongst them a single, logical order of connectedness, the rich, multidimensional order of our everyday concepts is ignored. The critique offered is influenced by the work of Wittgenstein and Goethe.
"Only in the stream of thought and life do words have meaning" (Wittgenstein, 1981, no.173).

"The ultimate goal would be to grasp that everything in the realm of fact is already theory... Let us not seek for something beyond the phenomena - they themselves are the theory" (Goethe, 1988, p.307).

In replying to Jan Smedslund's most thoughtful article, I want to draw attention to a number of themes articulated in the work of Wittgenstein and Goethe. The general stance I want to take is indicated in the quotations above. I want to be critical of Smedslund's suggestion (and by implication, of the aims of experimental psychologists also), that we should base our psychological inquiries, especially those that crucially concern our everyday lives, in a system of psychologic. Indeed, as I see it, to base our everyday social lives in any intellectual system, however arrived at, is to ignore how we do in fact use and make sense of our expressions and utterances in unique, particular, and continuously creative ways, in each new concrete situation as we encounter it. It is to base our lives within an external, artificially constructed world, devised to promote a single, preferred order, rather than living within life itself.

    In claiming to focus on "what is consensually self-evident, that is, on what everyone takes for granted as necessarily true, given that the words mean what they are ordinarily taken to mean," Smedslund takes it that embedded in our common sense understanding of everyday psychological issues is "an implicit calculus built into language, that enables people to predict and understand each other's behavior." Psychologic is, he says, his explication of it. Central to it, is a system of propositional statements. Indeed, they play a crucial role. For the professed goal of psychological research, Smedslund notes, is also to arrive at such a propositional system. However, the attempt to discover such a system empirically, seems misconceived. Many of the propositional statements put forward by experimental psychologists for empirical test (mostly in laboratory experiments) are analytic or necessarily true, and are thus not amenable, logically, to empirically testing.

    In making such a claim, both here and elsewhere (Smedslund. 1988), he is enunciating both a negative and a positive project. Negatively, he is critical of current empirical studies in psychology. They are pseudo-empirical: what experimenters in fact are doing in the experiments they devise to test their hypotheses, Smedslund suggests, is merely displaying the orderly nature of the psychological knowledge they already possess. Their experiments are not testing "general principles," but merely "auxiliary hypotheses" - "hypotheses about the conditions of application of the main hypotheses in a given situation at a given time, with given participants, instructions, materials, measuring instruments, and so on." These conclusions set the scene for his positive project. As a system of axioms and theorems, he offers psychologic as a major tool in a scientific psychology: it "relates to psychological realities in the same way as geometry relates to physical realities, that is, geography" (Smedslund, 1988, p.104). Psychological states can be 'mapped' in the same way that "the location and form of an object is characterized by its geometrically expressed relations to other objects and to a coordinate system" (p.104). Indeed, the main point of psychologic is that "[it] is constructed to serve the specific purposes of psychological research and practice. [Where] these purposes include preciseness of communication, and agreement on rules for prediction and for the evaluation of procedures." Indeed, as he sees it, psychologic is crucial to the continuation of psychology as a science, for without the precision in communication it provides, it is impossible to draw an appropriate distinction between the empirical/contingent and the a priori/noncontingent, and thus to achieve agreement as to "what follows and does not follow from given formulations, that is, to use logic." And, if we are unable to use logic in those of our affairs in which psychological matters are at issue, then "what follows from given words and sentences becomes intrinsically contestable, and this engenders endless debates where the provisional outcomes are the result of rhetoric."

    Smedslund's project, then, is important. Not only is he attempting to place psychology as an empirical science upon a proper footing, but he is also claiming to offer a basis for settling otherwise endless debates in which psychological matters are in some sense at stake.

    But what is the existential status of the "implicit calculus built into language" that Smedslund claims merely to be explicating? Does it in fact exist? Is it a discovery, or is it simply a possible construction, one among countless others? Does it in fact underlie our common sense reasoning, or is it something that could be used to overlay it? As I see it, in his 'explication' of psychologic, Smedslund follows many of the moves made by Saussure (1959) in his construction of structural linguistics as an intellectual enterprise. In setting the scene for his eventual focusing on language as a system of "differences without positive terms" (p.120), Saussure (1959) notes that: "Other sciences work with objects that are given in advance and can be considered from different viewpoints; but not linguistics... Far from it being the object that antedates the viewpoint, it would seem that it is the viewpoint that creates the object..." (p8). Indeed, linguistic phenomena are so various, so complicated and many-sided, that "the object of linguistics appears to us as a confused mass of heterogeneous and unrelated things" (p.9). Consequently, "as I see it," Saussure continues, "there is only one solution to the foregoing difficulties: from the very outset we must put both feet on the ground of language and use language as the norm of all other manifestations of speech" (p.9). Where the term "language" here, means language considered "as a self-contained whole and principle of classification" (p.9).

    Such a view of language, however, is not one that is given to us naturally; it is a view constructed for us by an intellectual expert, adopting a particular viewpoint. As I see it, Smedslund in his account of psychologic follows a similar set of moves: a system of logic can be found in the messiness of everyday commonsense activities, if we select just those normative elements from it upon which all native informants can agree.

    Clearly, in themselves, such systematic constructions are often of great importance. As Hertz (1894/1956) remarked in physics, as to the then confused question physicists were asking themselves about the nature of force and electricity: "But the answer we want is not really the answer to this question. It is not by finding out more and fresh [empirical] connections and relations that it can be answered; but by removing the contradictions existing between those already known" (pp.7-8). Conceptual clarifications produced by the construction of appropriate theoretical systems, as we know with respect to Einstein's work, have been crucial in the natural sciences. But in everyday life, the situation is different, very different. As Saussure (1959) goes on to remark, after having, technically, defined language as a system of elements known negatively, only in terms of their relations to the other terms of the system: "Whether we take the signified or the signifier, language has neither ideas or sounds that existed before the linguistic system, but only conceptual and phonic differences that have issued from the system" (p.120). Ex post facto, "the system" can easily come to displace, quite inappropriately, the everyday concepts from which it has been derived. In other words, while such systematic constructions may play a crucially important role in the natural sciences, if they are allowed wholly to usurp the role of our much richer, more open and multidimensional everyday concepts, they may exert a very dangerous influence indeed. Rather than opening up a royal road to unconfused scientific research, they might very well block off access to just those initial explorations required at the inception of new intellectual enterprises. Indeed, Saussure's (1959) own work would have been prejudiced.

    If we turn to his initial remarks in which he justifies his focus on just certain aspects of language and not others, we can ask: How could he already know so much about linguistic phenomena? How was it possible for him to argue that speech straddles "several areas simultaneously - physical, physiological, and psychological - it belongs to both individual and society" (p.9), and for us as readers to find no trouble in immediately acknowledging the truth of his claims? Isn't it because as masterful users of language already as speakers of our mother tongue, as well as the languages in a few specialized spheres, we (at least, the knowledgeable adults among us) already possess the only concepts of both a language and of language in general which are worth having? Without an initial, organized, unitary, synoptic grasp of what language as such is, i.e., a conceptual grasp of it, we would not be able to follow Saussure's arguments for his (re)-constitution of it as a normative system of self-identical forms, and thus as an object of study for scientific linguistics. Indeed, this is precisely what I take Goethe to mean in claiming that everything in the realm of fact is already theory. Saussure would not be able to make his claims about the many-sided complexity of language, if we did not all in some sense already know of language as a conceptual unity.

    This is not to say that we all have the same concept of it, nor is to say that in having such a concept we know fully how to describe its structure verbally, nor is it to claim that we are wholly unconfused as to its structure. But it is to say that - like all those who have dwelt in a city of some kind or other for a long enough period of time - that we can all come to know our 'way around' both inside a language or languages to a sufficient degree, to be able to converse in a worthwhile manner with others about its character. Indeed, it is precisely in such conversations with others that we can extend our knowledge of it and clarify our confusions. To do this, we do not need a psychologic. Like living in a city, we cannot get a more accurately detailed picture of a language by drawing back from it, as if to get a more overall view. Trying to regularize or systematize language by imposing an external 'map' upon it is no help to us. If our aim is to get to know our way around inside it better, we must enter into it more actively and explore it in yet more detail, close to. This is Goethe's and Wittgenstein's point. If we observantly follow all the changing details of a living whole, sufficiently closely and continuously through all its interconnected movements, we can come to a synoptic grasp of its 'inner nature'; we can arrive at that kind of understanding which, as Wittgenstein (1953) puts it, "consists in 'seeing connections'" (no.122). We can find our 'way around' inside our language entwined forms of life in an unconfused manner, move around inside their complicated landscapes without coming up against dead-ends; we can be 'at home' within them without maps.

    Jan Smedslund's project is very different: he does want to systematize and regularize language; he is in search of maps. He wants to capitalize "on the high degree of semantic constraint characterizing all languages." Following Wittgenstein (1953), I too want to claim - as Jan Smedslund also seems to suggest - that what we see in the world around us, and how we make sense of and respond to what we see, depends on the different ways in which we find ourselves related to it, linguistically, and thus as a result go on to engage ourselves with it. Associated with a way of talking is a form of life, a certain way of perceiving, thinking, acting, talking, and evaluating states of affairs. However, in opposition to Smedslund's claim that there is an implicit logical calculus built into language, I want to argue that whatever logical systems that we may fashion as representing orderly aspects of our relations to our world and to each other, are just that, i.e., systems of our own devising. Rather than us all speaking in an ordinary language, which can be considered as a single, unitary, implicit linguistic system. Following Wittgenstein, I want to suggest that in the many different spheres of our daily lives, we understand each other from within many different "language-games," which are themselves embedded in and a part of different "forms of life." Where, "the origin and the primitive form of the language game," says Wittgenstein (1980), "is a reaction; only from this can more complicated forms develop. Language - I want to say - is refinement, 'in the beginning was the deed' [quoting Goethe]" (p.31).

    This comment of Wittgenstein's highlights, I think, what I think is missing from Smedslund's whole account of psychologic (and from the conceptual world of experimental psychologist's also, for that matter): the importance, and the relational character, of our living, unpremeditated (thoughtless), spontaneous, bodily activities. On an enormous number of occasions in daily life, we act without having first to 'work out' how to act. We hear a groan from a pile of rags in the street, and we turn with pity toward it; we see what we think is a frog jumping in the road while driving, and we turn to avoid it (later we realize it was a leaf turning over in the wind); a mother says to a child "look" and points at something, and the child notices a rabbit running across the field; we read a sentence in a book or paper, and we gain a sense of its meaning directly; and so on. Without these immediate and spontaneous understandings of our world, we would stand totally befuddled like beings from another planet, completely confused as to how next to act. It is an initial, spontaneous living involvement of some kind with an other or otherness in our surroundings which 'sets the scene', so to speak, for how we go on in the rest of what we do in relation to the otherness concerned. And it is this - the initial way in which our spontaneous responses relate us to our surroundings - which ensures that everything else which follows from that reaction, occurs within the confines of the initial relation it establishes. Indeed, it is by being developed from such beginnings by internal articulation, that all our (new) forms of life, and the language-games which are internally related to them, come to have their character as intrinsically unitary wholes, i.e., as a set of interrelated parts which do not need an imposed external framework, a third thing, to hold them together as a 'systematic' whole.

    This distinction between a living whole, which is articulated from within itself into a more complexly structured but still living whole, and an abstract systematic whole, which in exhibiting a single, logical order of connectedness is structured as a closed system of parts, is crucial. The parts of a living unity are indivisible from the whole within which they have their being. They are all internally related in that they owe, not just their character but their very existence at every moment, not only to their relations to the other parts within the whole, but also to its earlier parts from which they have developed - thus, as well as their momentary spatial relations, their temporal (historical, developmental, or genetic) relations are of importance also. Further, the living whole itself owes aspects of its character to its (dynamic) relations to its surroundings. Psychologic and common sense, however, exist independently of each other; they both have their own self-contained natures irrespective of whether they are a part of a larger whole or not. So, although it is true that the elements of Smedslund's psychologic are all internally related to each other - his account here and elsewhere (Smedslund, 1988, 1997) is of it as an axiomatic system of primitive terms defined, not in terms of their positive content but only negatively in terms of their relations to each other within the system - psychologic itself is not linked to common sense internally. In being linked by a third thing - a 'mapping' relation, a relation of correspondence - they are in an external relation to each other.

    Here, then, is what I think is being ignored in such projects as Smedslund's (and Saussure's, and by experimental psychologists also): the degree to which all that is fact for us in our human lives is already theory. Or, to put it in other words, we are ignoring the degree to which, from within their embedding in the ceaseless flow of spontaneously responsive activity continuously flowing between us, our actions and expressions already exhibit conceptual meanings in the reactions they elicit. Thus, to impose a rational system externally upon them, is to ignore all the internal relations existing both within this flow of activity, as well as between it and its surroundings. Seen in this light, in its actual use, language is only ever a minor part of the much more comprehensive flow of interactive activity between people living in spontaneously responsive contact with each other. In its ordinary uses in everyday life, it is never a self-contained linguistic system. Without the reflective effort, without the principle of selection - the focus "on what everyone takes for granted as necessarily true" - and the imaginative effort required to explore the possible systematic relations of meaning between them, the system of psychologic could not have emerged. That Jan Smedslund has brought it into being is no mean achievement. But, I want to claim, his presentation of it to us as here, as an explication of some 'thing' or 'structure' that is already in existence, implicitly, in our language, is tantamount to asking us to see our world in a way very similar to the way the empiricists Smudslund is criticizing see it: as an external world, a world that exists 'out there' independently of us and of our lives. Whereas, if the view from Wittgenstein and Goethe that I have put forward here is accepted, then everything which makes sense to us, makes sense as an internal elaboration or refinement of relationships initially established in our living, spontaneous, reactions to each other. To impose a single, external, theoretical order of connectedness upon such relations is to ignore all the rich and varied connections present to us in such phenomena, as if we had actually to live in our maps rather than in our cities.


Goethe, J.W. (1988) Scientific Studies (Ed. and trans. D. Miller). New York: Suhrkamp.

Hertz, H.H. (1956) The Principles of Mechanics. New York: Dover (orig. German pub. 1894).

Saussure, F. de (1959/1966) Course in General Linguistics (Eds. C. Bally and A. Sechehaye). New York: McGraw-Hill.

Smedslund, J. (1988) Psycho-Logic. Berlin and New York: Springer-Verlag.

Smedslund, J. (1997) The Structure of Psychological Common Sense. Mahwah, NJ: Lawrence Erlbaum.

Wittgenstein, L. (1953) Philosophical Investigations. Oxford: Blackwell.

Wittgenstein, L. (1980) Culture and Value, introduction by G. Von Wright, and translated by P. Winch. Oxford: Blackwell.

Wittgenstein, L. (1981) Zettel, (2nd. Ed.), G.E.M. Anscombe and G.H.V. Wright (Eds.). Oxford: Blackwell.