The Oxford Handbook of German Philosophy in the Nineteenth Century

by Michael N Forster

  All three issues loom large in nineteenth-century Germanophone philosophy of language. But that tradition divides into three strands. These do not constitute genuine traditions, let alone schools. Nonetheless, they differ in orientation and, to some extent, developed independently.1 For want of better terms, I shall refer to them, respectively, as the hermeneutic strand, the logical strand and the critique of language. For present purposes, the hermeneutic tradition comprises not just the ‘classical’ hermeneuticians of the nineteenth century, namely Schleiermacher, Schlegel, and Dilthey, but also its all-important eighteenth-century roots (Hamann, Herder), as well as associates within Sturm und Drang, Romanticism (Schleiermacher, Schlegel, von Humboldt), and German Idealism (mainly Hegel). The logical strand consists of what has come to be known as the ‘Austrian tradition in philosophy’ (mainly Bolzano) and of Frege, a singular figure as regards both his academic roots and his intellectual accomplishments. The critique of language consists of otherwise diverse figures like Lichtenberg, Gruppe, Nietzsche, and Mauthner. Its interactions, both positively and negatively, are mainly with the hermeneutic strand; yet it shares the quest for clarity with the logical strand.

  The ideas of language as the medium of communication, the glue of society and a driving force of history are clearly more prominent in the hermeneutic tradition. The idea of language as a source of (philosophical) problems and confusions is the defining feature of the critique of language, while also playing a role in the logical strand. And the idea of language as a resource for resolving such problems and for facilitating the quest for knowledge is most evident in the logical strand. Finally, the relationship between thought and language has exercised all three currents in roughly equal measure.

  Nevertheless this chapter will accord more space to the hermeneutic strand. This is not to lend succour to those myopic enough to reduce Germanophone philosophy after Kant to Hegel, Nietzsche, and Heidegger, or to extol German Idealism and Lebensphilosophie as ‘classical’ German philosophy. Rather, the Germanophone roots of analytic philosophy have by now been widely recognized. By comparison, the considerable contributions of the hermeneutic strand to the philosophy of language remain underappreciated in the Anglophone world. Ironically, this holds not just for mainstream analytic philosophers, but also for those interested in so-called ‘continental philosophy’ mainly as an antidote to Enlightenment ideals and as a source of political and cultural ideologies. In any event, in addition to sketching the story of the different strands, there will also be some comparisons and juxtapositions, both between and within them. In this respect, the following highly general oppositions will loom large:

  •lingualism vs. Platonism and mentalism,

  •psychologism and empiricism vs. anti-psychologism and a priorism,

  •individualism vs. communitarianism,

  •universalism vs. pluralism,

  •absolutism vs. relativism,

  •constancy and precision of meaning vs. instability and vagueness,

  •language as a medium of thought vs. language as a tool or calculus,

  •ordinary vs. ideal language.

  19.2 KANT’S LEGACY: THE REFLECTIVE TURN

  There have been sporadic attempts to detect a systematic and novel philosophy of language and even a linguistic turn in Kant.2 This is a case in which the time honoured maxim ‘If it’s in Kant, it must be good!’ has been supplemented by the even more suspect ‘If it’s good, it must be in Kant!’. Kant’s importance for our topic does not lie in his scattered and largely unoriginal remarks about language. Instead, it lies in the general impact of his ‘Copernican Revolution’ on nineteenth-century thought, and in particular on the self-image of philosophy.

  From Kant onwards, one concern has united many otherwise diverse strands of Germanophone philosophy, namely whether philosophy can preserve a distinct role in view of the progress of the empirical sciences.3 This question was linked to a second one, namely whether philosophy is a priori. If philosophy is to be a cognitive discipline, yet distinct from empirical science, this is because, like logic and mathematics, it aspires to knowledge of a non-empirical kind. One explanation of such knowledge is the kind of Platonism favoured by Bolzano and Frege: logic and mathematics are a priori because they deal with entities (numbers, concepts, thoughts, truth-values) beyond the physical realm. Another explanation was provided by Kant. Philosophy is immune to confirmation or refutation by empirical evidence not because it deals with abstract entities beyond the empirical world, but because it reflects on the non-empirical preconditions of empirical knowledge, logical or cognitive structures that antecede contingent matters of empirical fact.

  Traditional metaphysics sought insights into either a reality beyond space and time (Platonism) or the most abstract and general features of reality (Aristotelianism). Kant brought about a fundamental re-orientation by turning the metaphysical quest for a priori knowledge of a substantive kind into a ‘critique of pure reason’, that is, a second-order reflection on the non-empirical aspects of the way in which ordinary reality is given to us in experience, or, more generally, of the way in which we represent reality. This reflective turn4 lends itself to a linguistic transformation. If the representation of reality or human thought in general are essentially tied to language, the reflection on the ‘transcendental’ preconditions of thought turns into a reflection on the nature and structure of the linguistic expression of thought.

  At a grand-strategic level, the achievement of Kant’s contemporaries Hamann and Herder consists in having effected such a transformation, treating language rather than reason as the precondition of thought. The point is evident from the fact that both published self-proclaimed metacritiques of Kant’s critical philosophy—respectively, Metakritik über den Purismus der Vernunft 17845 and Metakritik zur Kritik der Vernunft 1799.6 Thus Hamann accuses Kant of a misguided ‘purification of philosophy’, which consists of ignoring the three-fold dependence of reason on ‘transmission, tradition and faith therein’, on ‘experience and its everyday induction’, and on ‘language, the first and ultimate organ and criterion of reason, without any foundation other than transmission and custom’.7

  Hamann and Herder stand in an ambivalent relation to the Enlightenment more generally. Motivated by his Protestant faith, Hamann passionately repudiated the ideals of progress, rationality, and a common human nature. Herder was less of an irrationalist and Christian apologetic. Yet he was equally sceptical of the idea of a universal though progressing human nature, and instead stressed the contingencies of human history and the diversity of human societies. Furthermore, like Hamann he tried to undermine the orthodox distinctions of faculty psychology, notably those between reason, will, desire, and affection. Rather, a human being is a unity infused by a Kraft, a spiritual or vital force.

  On the other hand, Hamann’s and Herder’s preoccupation with language reflects a central Enlightenment topos. For epistemological reasons, the Enlightenment was interested in the connection between thought and language; for anthropological reasons, it was concerned with the difference between human and animal communication and with the origins of language and its relation to the progress of human reason. In his Abhandlung über den Ursprung der Sprache8 Herder propounded the progressive view that language is a human edifice; it is divine only to the extent to which humans are God’s creation. He also recognized that language resulted from gradual processes rather than a spontaneous invention. Finally, the striking first sentence ‘Already as an animal, man possesses language’9 anticipates a profound anthropological insight, namely that the capacity for language is part of a specifically human yet nonetheless completely natural form of life. Hamann condemned Herder’s secular explanation as incoherent and defended the divine origins of language.10 In this context, he invoked a pet idea of his that proved to be subliminally influential, namely that God speaks to us not just through the scriptures, but also through nature and through history.11 Then again, in tying thought closely to language both Hamann and
Herder were indebted to the Enlightenment thinkers Leibniz and Wolff.12 Finally, it is the aforementioned Kantian background which explains why for the hermeneutic strand language is not just one part of reality among others to be investigated by metaphysics or science, but central to the very task of philosophy.

  19.3 THE LINGUISTIC TURN OF HAMANN AND HERDER

  Hamann and Herder anticipated the linguistic turn of twentieth-century analytic philosophy. For them, language is not primarily a means of reference to non-linguistic objects and a means of conveying non-linguistic thoughts; rather, language expresses and constitutes thought. According to Forster, they advanced two more specific fundamental theses: (1) ‘thought is essentially dependent on and bound by language’ and (2) ‘meaning consists in word-usage’, that is, the meanings of words (the concepts expressed by them) are not non-linguistic entities they refer to (whether material, abstract, or mental), but rather ‘usages of words’.13 There is a ‘priority dispute’ over who actually founded this ‘Herder-Hamann tradition’,14 and thereby my hermeneutic strand. The received picture is that the wayward yet profound Hamann inspired Herder, who publicized and popularized his views.15 And there is no doubt that Hamann, the elder of the two, nurtured Herder’s interest in language, and taught him foreign languages. However, Forster has argued convincingly that Herder propounded (1) and (2) already in the 1760s,16 when Hamann was still wedded to a more orthodox view that at most concedes a mutual dependence of thought and language.17

  As regards the lingualist thesis (1), Herder claimed in 1767–8 that language ‘is the form of cognition, not merely in which but also in accordance with which thoughts take shape, where in all parts of literature thought sticks to expression, and forms itself in accordance with this. …Language sets limits and contour for all human cognition’.18 A later summary has it that ‘language is the character of our reason’.19 As regards the use theory (2), Herder maintained that in determining the concept aka meaning of an expression the question is not how it ‘can etymologically be derived and analytically determined, but how it is used’.20 ‘Let us seek a word’s concept not from etymologies, which are always uncertain, but according to the clear use [Gebrauch] of the name in its various times’.21

  Nevertheless, there are noteworthy complications. First, in several places, including his most famous contribution to the philosophy of language, the Abhandlung, Herder’s lingualism reduces to the contention that thought depends on language because it amounts to a form of inner speech, with concepts/meanings being the words of this language of thought.22 This doctrine is far less original; it boasts illustrious advocates from Plato through Occam to Fodor. Hamann censured Herder for adopting it.23 Rightly so, his failure to adduce arguments notwithstanding. As the anti-psychologism of the logical tradition culminating in Wittgenstein has shown, inner speech is no more necessary or sufficient for thinking than mental images; and the idea of a language of thought beyond the reaches of the subject’s consciousness (notably one implemented by neural firings) is incoherent. For this reason, what one might call process lingualism is untenable. Thinking is not constituted by a process of inner speech accompanying outer discourse. Capacity lingualism, according to which only creatures capable of expressing them linguistically can have thoughts, is more plausible. Yet the actual kernel of truth in lingualism, or so I have argued, is this. Ascribing thoughts makes sense only in cases where we have criteria for identifying thoughts. Something must count as thinking that p rather than that q. Accordingly thoughts, although they need not actually be verbalized either externally or internally, must be capable of being manifested. And only a very restricted range of thoughts can be expressed in non-linguistic behavior.24

  By the same token, however, it is wrong simply to identify thought and language, as Hamann seems to: ‘reason is language, logos’.25 Leaving aside whether Hamann took reason to be equivalent to thought, this purple passage cannot be defended by insisting that logos includes inner speech. But there are mitigating considerations. In other passages Hamann treats language merely as an essential precondition of thought, thereby implying that they are not identical.26 Furthermore, his conception of language includes forms of expression beyond linguistic symbols (understood in Peirce’s sense as signs governed by convention), such as music and visual art. Such a ‘broad expressivism’ was later adopted by Hegel, Schlegel, and Dilthey.27 It is correct in so far as non-linguistic sign systems and media—whether iconic or not—can have a meaning in the sense of expressing concepts and thoughts. It is more contentious whether they can express thoughts that transcend linguistic explanation. Mysticism notwithstanding, any bona fide thinking must allow of being specified as a case of thinking that such-and-such, and hence of linguistic specification. And in the last instance, the established idiom of natural languages remains the ultimate medium for the disambiguation and explication of what precisely is being expressed by other types of signs. In any event, if broad expressivism is tenable, it makes identifying thought with language more palatable.

  Secondly, Herder only propounds the ‘epistemological’ claim that the meaning of a word is to be gathered from its use. Hamann advances the stronger, ‘metaphysical’ claim that meaning is determined by use: words turn from mere objects of the senses into ‘understanding and concepts’ through the ‘spirit of their employment’.28 As far as I can tell, however, neither Herder nor Hamann simply identifies the meaning of a word with its use. Good for them! The thesis that meaning is identical with use founders, if only because there are significant differences between the use of ‘use of (way of using) an expression’ and that of ‘meaning of an expression’. By contrast, the claim that meaning is determined by and hence evident in use is defensible, provided that it recognizes the normative dimension of use.29

  Thirdly, in addition to broad expressivism, Hamann rather than Herder introduced two other seminal ideas into the hermeneutic tradition. First, semantic holism: ‘Like numbers, words receive their value from the place they occupy, and their concepts, in their determinations and relations, are like coins, mutable depending on location and time’.30 Secondly, a communitarian perspective, which stresses the social character of language. Although Herder occasionally suggested such a view, he vehemently rejected it in the Abhandlung. The origin of language lies in the solitary reflection—Besonnenheit—of individuals. To impose order on the sensory input, they engage in a process of abstraction and thereby isolate specific properties (Merkmal) of objects, which are assigned an inner label (Merkwort) that is subsequently vocalized. ‘The very last thing’ responsible for the genesis of language was ‘consensus, arbitrary convention of society. The savage, alone by himself in the forest, would have had to invent language, even if he had never used it. It was consensus of his soul with itself, a consensus so inevitable as humans being humans’.31 For this individualism Herder was taken to task by Hamann in Philologische Einfälle und Zweifel,32 yet it later endeared him to Chomsky.33

  19.4 CLASSICAL HERMENEUTICS: SCHLEIERMACHER, SCHLEGEL, AND DILTHEY

  What one might call ‘classical’ German hermeneutics developed the aforementioned ideas of the Herder-Hamann tradition, primarily through applying them to problems concerning interpretation and translation. Schleiermacher and Schlegel took on board Herder’s lingualism, often in the version that identifies thought with inner speech. In Schleiermacher we also encounter intimations of the Wittgensteinian view that linguistic meaning depends not on all features of linguistic use (for instance, on the effects of specific utterances on the hearer or on socio-linguistic conditions), but only on rules of correct use. In his hermeneutics lectures he maintained that ‘the meaning of a word is to be derived from the unity of the word-sphere and from the rules governing the presupposition of this unity’.34 This passage also indicates Schleiermacher’s development of semantic holism. The prima facie distinct senses of certain words nonetheless form part of a larger ‘word-sphere’. Furthermore, any specific word or concept is partly defined by its place within a
‘system of concepts’,35 an idea which came to fruition in what Strawson calls ‘connective analysis’. Finally, the grammar of a particular language contributes to the concepts it expresses. By implication, the syntactic features of specific lexical items are partly constitutive of their meaning. This is inimical to the strict separation of syntax and semantics advocated by Chomsky’s generative grammar, yet congenial to the project of a ‘logical grammar’ launched by Husserl and Wittgenstein.

  Schleiermacher rightly perceives holism as an obstacle to linguistic understanding. ‘Any speech’ is intelligible only ‘within the totality of language’, indeed, ‘only in the context of the whole life, to which it belongs’,36 and ultimately within a broader cultural context. When it comes to interpretation, for instance, the reader faces a pervasive and unavoidable circularity. The individual parts of a text need to be understood in the light of the whole; conversely, the whole text can only be understood by understanding its constituent parts. Schleiermacher’s solution consists in recognizing that understanding, far from being an all-or-nothing affair, comes in degrees and can therefore be achieved by a gradual process. Having worked through parts of the text, we arrive at a preliminary interpretation of the whole, which in turn is applied to modifying the understanding of certain parts, and so on. This commonsensical procedure achieved more fame than it deserves under the label ‘hermeneutic circle’ (Dilthey) and ‘circle of understanding’ (Gadamer).