INTRODUCTION

MARTIN GUSTAFSSON AND LARS HERTZBERG

Language, it seems, is a matter of practical employment. Without real-life exchange between concrete human beings, there would be no such thing as linguistic meaning. And, as engaged participants in such linguistic activities, we do not treat people’s utterances as objects for theoretical study. Rather, we act upon what other people say; we provide answers to questions, argue for or against hypotheses, draw conclusions, obey orders, and so on and so forth. Of course, it also happens that things we hear or read evoke contemplation rather than immediate overtaction. But even then, the contemplation is not external to the practice itself. The utterances are still being treated as meaningfully employed expressions. We are concerned with what is being said, rather than with something that can be identified in abstraction from the context of meaningful use.

On the other hand, it is of course possible to treat linguistic phenomena from a detached, theoretical standpoint. Such theoretical treatment is not uncommon today, and is both scientifically and technologically important. Hence, any serious attempt at philosophising about language will necessarily involve reflections on the relation between theory and practice. It will have to deal with questions such as: How are theoretical notions of language related to actual communication? What do theoretical models tell us about real-life language use? And are there any limits to what such theoretical notions and models can reasonably be taken to accomplish?

In discussions of such issues, the significance of the practical dimension of language is often misconstrued or undervalued. Roughly speaking, there is a strong temptation to give undue priority to one’s favourite theoretical perspective, and to think that the descriptive scheme invoked by that theory constitutes a pattern to which all linguistic practices somehow have to conform. The fact that such conformity is not immediately visible in what speakers actually do, is usually not enough to hamper this temptation. Having recognised the gap between theory and practice, the theorist often goes on to postulate a deeper, invisible level at which the mechanisms described by the theory are supposed to be at work; this level is thought somehow to underlie and explain the verbal behaviour at (what now gets to be called) the surface. This sort of move makes it seem as if the actual practice is only of secondary importance – as if what the speakers do is just a superficial symptom or effect of what is really essential about their language.

A recurring theme in this volume is the questioning of such essentialist and explanatory aspirations. From many different perspectives, doubt is cast on attempts to fit the whole of linguistic practice into some general theoretical framework or explanatory model. Importantly, the criticism is not directed at the use of such frameworks and models to fulfil certain intra-scientific or technical purposes of investigation. Nor is it denied that theoretical notions may be used to provide fruitful accounts of certain limited aspects of linguistic activity. Rather, what is questioned is the metaphysical projection of frameworks and models onto the object of study itself, a projection which is based on the supposition that the theoretical apparatus and explanatory principles somehow constitute neutral reflections of features that are inherent in, and essential to, language as such.

As the essays below show, manifestations of such theoretical innocence take many forms, and can be found among theorists of quite different colours. An overall aim of this volume is to make the reader more aware of the various guises in which this metaphysical way of thinking may appear. The multitude of perspectives and themes is intentional, and the contributions should be read as providing examples of how the same kind of philosophical mistake can flourish in quite different intellectual environments. Indeed, it would not be inappropriate to perceive this pluralistic approach as a criticism of a certain narrow-mindedness that is much too common in today’s academia. Such narrow-mindedness makes it nearly impossible to see the fundamental similarities that often exist between conceptual problems arising in different areas of inquiry. It is our hope that the contributions in this volume can help the reader broaden her views.

The volume begins with three historically oriented discussions. Sören Stenlund’s essay provides new perspectives on the so-called linguistic turn in twentieth-century philosophy. Stenlund criticizes the standard kind of historiography which presents this event as if it occurred for “purely philosophical” reasons, independently of external factors. According to Stenlund, to understand fully the linguistic turn, one has to consider why the mentalistic vocabulary employed by thinkers in the seventeenth and eighteenth centuries gradually lost its original vitality and significance, and came to be perceived (by Frege and others) as entailing a dangerous and untenable psychologism. Stenlund offers an account according to which this gradual transformation is tied to overall changes in Western society, such as secularization, the loss of belief in the uniformity of human nature, and the development of empirical psychology. Toward the end of the nineteenth century, such changes had made it quite natural – even seemingly necessary – to conceive language, not as an external instrument for transmitting independently identifiable mental contents (as Locke and others had done), but as the medium in which thoughts and reasons must already be articulated in order to constitute determinate thoughts and reasons at all. Stenlund calls this notion that we cannot (as it were) step outside language, the idea that reason and thought are immanent in language.

One thing that makes Stenlund’s account unusual is his refusal to depict this change simply in terms of intellectual or scientific progress. In fact, he argues that the notion of linguistic immanence, of being “trapped within language”, is no less metaphysical than the view of language as an external instrument for transmitting already identified thoughts. According to Stenlund, both these opposing pictures give undue priority to certain theoretical perspectives on language. Much of his essay aims to uncover what these different ways of theorizing involve. Stenlund connects the earlier picture to the tendency among seventeenth and eighteenth century thinkers to theorize about language in terms of its origin. As Stenlund notices, the withering away of talk about the origin of language goes hand in hand with the gradual decline of the mentalist vocabulary. At the time of the linguistic turn, this talk about origins is definitively replaced by other approaches, in which language is depicted in ahistorical – logical or structuralist – terms.

The other essays in this first section deal with two of the most influential figures in the history of Western thinking about language: Wilhelm von Humboldt and Ferdinand de Saussure. Both these thinkers were deeply concerned with the question of how real-life language use is related to a notion of language as a formally well-defined and generally specifiable whole. By engaging in a close reading of numerous passages in Humboldt’s oeuvre, Olav Gundersen traces Humboldt’s attempt to reconcile what he conceived of as the amorphous character of linguistic practice with the notion of intellectual, normative and systematizable forms of speech (Redeformen). On the one hand, Humboldt regards such forms of speech as given a priori in relation to present speech acts; on the other hand, those intellectual forms can be identified only by means of analysis of, and abstraction from, previous uses of language. Gundersen shows how the apparent tension between these different aspects forces Humboldt to a radical reappraisal of the relation between philosophy and the empirical Sprachwissenschaften.

Michael Gustavsson offers a fresh interpretation of Saussure’s notions of langue, langage, and parole. According to a standard reading, langue denotes an abstract, systematized structure, in principle distinguished from actual linguistic practice. Gustavsson argues that this reading is not just over-simplified but fundamentally mistaken. In fact, Saussure never speaks of langue as something abstract but, on the contrary, as something concretely given in real-life language use. Why? Gustavsson’s proposal is that Saussure’s notion of langue was primarily developed as a way of distinguishing between language as it is concretely given for participants engaged in linguistic practice, and language as studied from the viewpoint of empirical science. Rather than an abstract structure, langue is language as seen from within its meaningful use; and this is contrasted, not with some more “concrete” object, but with linguistic phenomena as conceived from the outside, by empirical linguistics. Gustavsson discusses at length the implications of this reading for the understanding of other aspects of Saussure’s viewpoint, such as his notion of the linguistic sign.

Section II contains essays concerned with notions of language within linguistics and feminist epistemology. Sven Öhman observes that, although spoken language has normally been seen as primary in relation to written language, linguists have nonetheless tended to treat all language as if it were already graphically represented. More precisely, linguists have thought of the standard graphical segmenting and structuring of language into words, grammatical sentences, and so forth, as if it mirrored features already present in speech. Öhman argues that this assumption is confused, and he offers various examples to show that innumerable alternative methods of graphical representation are conceivable, not one of which can sensibly be described as more adequate to the intrinsic nature of speech than any other. Thus, our choice of method of graphical representation depends, not on the character of language itself, but on our specific purposes in giving such a representation. A connected point is that in everyday language, words are spoken about in two different ways: on the one hand we refer to text words (the string of letters separated by two typed spaces), and on the other hand, to spoken words, i.e. to vocal expressions used in linguistic communication. Text words, however, have no counterparts in speech.

Saara Haapamäki provides an illustration of the tendency to mistakenly project one’s principles of theoretical representation onto the represented object itself. Her essay consists of a critical study of how Swedish grammarians have tried to implement Noam Chomsky’s generative linguistics in Swedish grammar-textbooks. According to Haapamäki, these grammarians have naively taken it for granted that Chomsky’s theoretical notions can be used to give a pure description of how the Swedish language functions – as if technical terms such as “deep structure” and “transformation” denoted phenomena present in the Swedish language as it is already known in practice by its speakers. Haapamäki shows that this involves a considerable distortion of Chomsky’s own viewpoint. Whereas Chomsky is quite aware that his theoretical framework involves far-reaching idealisation and abstraction, his Swedish followers proceed as if that framework constitutes a more or less immediate reflection of concrete linguistic phenomena.

Confusions of a similar kind can also be found in other areas, and among theorists of quite different colours. In a critical examination of feminist approaches in epistemology, Sharon Rider argues that feminist philosophers of science have in common with more traditional epistemologists the tendency to conflate their own theoretical perspective with the perspective of working scientists. Feminist philosophers often fail to see that their claims about the contingency and historical and cultural situatedness of scientific practices do not reveal fundamental features of these practices themselves, but define an outsider’s way of looking at them. Of course, such an outside perspective may be fruitful for various purposes, but the claims made from that theoretical standpoint in no way undermine notions such as that of scientific objectivity or the existence of conceptual necessities in scientific language, as those notions are used in scientific practice. Rider provides a detailed examination of Evelyn Fox Keller’s philosophy of science as a sophisticated example of this sort of conflation.

Section III in this volume contains two essays dealing with conceptions of meaning, truth and linguistic practice in the works of Hilary Putnam and Charles Travis, respectively. Frederick Stoutland provides a detailed account of how Putnam’s view of truth has changed over the years. Stoutland describes both the early “metaphysical realist” position, as well as Putnam’s own criticism of that early view, a criticism which eventually resulted in an antithetical, “internal realist” doctrine. The emphasis, however, is on the latest development in Putnam’s thinking: the turn to a Wittgenstein-inspired, “common-sense realist” conception of truth. Stoutland subscribes to the basic features of this common-sense conception. There are a number of points, however, on which he disagrees with Putnam, and he clarifies in detail what those points are. In particular, he argues, pace Putnam, that common-sense realism can be nothing other than a version of a so-called “deflationist” conception of truth.

There is a temptation to think that there must be central aspects of linguistic meaning which precede, and can be separated from, what statements actually accomplish in concrete situations. In particular, there is an inclination to think that those aspects of linguistic meaning which pertain to the truth and falsity of statements can be systematically captured on a general, theoretical level; that, typically, such things as reference and truth-conditions belong to expressions qua mere expressions, independently of the particular context of employment. This idea, which is problematised to some extent in Stoutland’s discussion of Putnam, constitutes a focus of critical attention in Charles Travis’s philosophy of language. In his essay, Martin Gustafsson explores Travis’s criticism of received views of meaning and truth. According to Travis, any attempt to identify the truth-conditions of statements in abstraction from the particular situations in which the statements are made, will founder on the “occasion-sensitivity” of linguistic expressions. Gustafsson clarifies this notion, and then argues that Travis’s criticism can and should be radicalised even further than some of Travis’s formulations seem to suggest.

The final section of the book contains five essays which discuss themes from Wittgenstein. Wittgenstein often said that philosophy is purely descriptive, and he contrasted philosophical description with the construction of theories. What does this notion of a descriptive, non-theoretical philosophy involve? Pär Segerdahl tries to shed light on this issue by relating it to Wittgenstein’s observation that the basic difficulty of philosophy is not a difficulty of the intellect but of the will. To remain at a purely descriptive level is to say no more than one already knows. In particular, it means to resist the usual expectations concerning the kinds of thing a philosopher should be able to say. Philosophers are expected to develop new and exciting hypotheses – not to restrict themselves to the production of seemingly trivial reminders. To illustrate what kind of difficulty this involves, Segerdahl draws a parallel between the descriptive philosophical spirit and the attitude of the main character in Albert Camus’ novel The Outsider.

Thorsten Johansson investigates the similarities and differences between the conception of logic in the Tractatus and the later Wittgenstein’s notion of philosophical grammar. According to Johansson, logic in the Tractatus is meant to function as a transcendental “frame” of language, a frame presupposed by any meaningful description of the world. In contrast to the logic of Principia Mathematica, as well as to later varieties of formal logic, the logic of the Tractatus is impossible to capture from a meta-perspective. Indeed, according to Johansson, the early Wittgenstein denies that logic can be meaningfully spoken of at all. And even if Wittgenstein later came to allow for the possibility of describing the logic of language by means of so-called grammatical remarks, Johansson argues, such grammatical descriptions are still very different from the sort of accounts characteristic of the meta-mathematical tradition. Wittgenstein’s grammatical remarks involve no meta-perspective, but are made from within the linguistic practice described. By means of various examples, Johansson looks more closely at what is involved in this central aspect of Wittgenstein’s thought.

Juan Wilhelmi discusses one of the most intensely debated topics in Wittgenstein’s philosophy, namely, that of rule-following. According to Wilhelmi, many contemporary discussions conflate two types of rule that should be kept separate: governing and descriptive rules. Wilhelmi takes up David Bloor’s meaning finitist reading of Wittgenstein as a clear example of such conflation. According to Wilhelmi, Bloor takes it for granted that grammatical rules are on a par with traffic rules and other dicta that are explicitly introduced to regulate and govern an already existing activity. By contrast, Wilhelmi argues that grammatical rules are of quite a different kind: they play no governing role, but remind us of aspects of our language use, aspects with which we are already practically acquainted. Wilhelmi also insists that the point of such reminders is closely bound up with their role in the clarification of particular philosophical problems. If they are made with no such problem in sight, the reminders are, at best, empty.

Gudmundur Steingrímsson criticizes the idea, sometimes attributed to Wittgenstein, that every human being is subject to philosophical confusion merely in virtue of speaking a language. Steingrímsson urges his readers to keep in mind that phrases which a philosophically trained audience might be inclined to classify as “nonsensical” or as involving some “philosophical myth” – say, talk about the mind in mechanical terms – may actually have all kinds of fully understandable, everyday uses. Indeed, the claim that such ways of talking necessarily manifest the speaker’s being held captive by some confused philosophical picture actually belies the real insight behind slogans such as “meaning is use” or “words are tools”. According to Steingrímsson, the important thing is not what phrases are used, but the work to which those phrases are put in particular cases. Premature accusations of nonsensicality bring obscurity rather than clarification.

Finally, Lars Hertzberg’s essay is a commentary on the debate concerning the role of the language community in Wittgenstein’s discussion of language and privacy. In that debate, it has widely been taken for granted that the issue turns on the importance of community standards in giving content to the distinction between correct and incorrect applications of a rule. Hertzberg argues, however, that the notion of a community standard is itself elusive and unable to perform the task it is called upon to perform by some defenders of the community view. He contends that the emphasis on words and meanings being connected with human interaction is nevertheless crucial to the philosophical perspective Wittgenstein was advancing: the crucial lack in the case of the solitary speaker, however, is not community standards but the relation to a listener. Drawing attention to the role of the listener was part of Wittgenstein’s effort to turn the philosopher’s attention away from the preoccupation with objects of discourse, urging her to focus instead on the kinds of interchange we engage in, in speaking about things.

This volume is the fruit of long-standing collaboration in the philosophy of language between various Nordic departments of philosophy, especially those at the University of Uppsala and at Åbo Akademi University. Most of the contributions are the result of work in the Uppsala-based project “Language and Human Action” and the Inter-Nordic project “Language and Practice”.