2. Setting the scene

Hamann (1730–1788) was a few years younger than Kant (1724–1804). He had been a student at the University of Königsberg and knew Kant personally. They remained acquaintances throughout Hamann’s life, although opinions do differ on their precise relationship. Some have seen them as friends, others are less sure.Footnote ¹ Hamann visited Kant’s home intermittently (although the visits were never reciprocated); they would also occasionally dine together with others, and Hamann sporadically accompanied Kant on his well-known daily walks.

Kant helped him both financially and indeed with gaining employment. Hamann remained for many years a minor Prussian civil servant until his resignation in 1787, dying a year later. Hamann throughout his life was an autodidact and remained in regular correspondence with many philosophers and intellectuals of the time – even whilst apparently employed in the Prussian civil service. Most notably, he had a significant aptitude for languages, including Greek, Latin, French, Arabic, and English (Berlin Reference Berlin1994: 89).Footnote ² His language ability also brought him to translation work, one more notable example being Hume’s Dialogues concerning Natural Religion (1781), which was never published but was lent to Kant.

However, what is of most interest is that within Hamann’s circle of acquaintances was the Riga publisher Johann Hartnoch – publisher of Kant’s Critique of Pure Reason (1^st edition 1781, 2^nd edition 1787). Hartnoch, as far as is known without consulting Kant, sent the manuscript of Kant’s first edition of the Critique of Pure Reason to Hamann, presumably to review it. It thus appears that Hamann was the first to ever read and evaluate Kant’s signature work. Hamann’s mixed views on the work led him to draft a philosophical response – ‘The Metacritique on the Purism of Reason’ – completed in January 1784. He remained unhappy with the text and (as far as we know) did not show it to Kant, although he sent it to Herder, who in turn passed it on to Jacobi in manuscript form. Of course, Herder later published his own Metacritique.Footnote ³ However, Hamann’s essay remained in this manuscript format and was only published posthumously within a collection of essays in 1800. It is doubtful that Kant ever read it, although Hamann had been a fairly regular correspondent with Kant from the later 1750s and had taken a continuing critical interest in Kant’s work from that time onwards.

Hamann’s Metacritique is probably his most significant philosophical work although it was preceded by the Socratic Memorabilia in 1759, as well as several short opuscula essays (or as he referred to them ‘Flying leaves’). Some of these latter essays focused critically on Herder’s work (particularly on the issue of language) and are collectively known as the Herderschriften (see Griffiths-Dickson Reference Griffiths-Dickson1995: 150–245). Another Hamann work Golgotha and Scheblimini (1784) was largely an appraisal of Moses Mendelssohn’s Jerusalem, or on the Religious Power and Judaism. This latter work was reckoned by Hegel, in the 1820s, as a significant philosophical piece on a par with Metacritique.Footnote ⁴

3. Where to place Hamann?

The current essay will try not to fix Hamann with a convenient marker. Yet, unavoidably in the diverse literature, he has attracted many tags. At the more affable end, he is often seen as a founding figure in the early European Romantic movement. In the German context, this would be associated with the Sturm und Drang, that is, within poetry, philosophy, literature, and the fine arts. In parallel, he has also been classified, erroneously, with Jacobi as a philosopher of feeling. Feeling is certainly given a place in the Hamannian perspective, but it does not have any outright priority. In line with the more conventional view of Romantism’s negative reaction to the claims of the Enlightenment, Hamann has also been taken as an early philosophical exponent of the Counter-Enlightenment. Analogous to this latter point is a more negative appraisal, in the literature, an assessment often associated with Isaiah Berlin’s text The Magus of the North (1993).Footnote ⁵ This situates Hamann as a nascent exponent of European philosophical irrationalism. Hamann has even been compared in some recent writings with postmodern thought (see Betz Reference Betz2012: 312ff).

An obliquely related judgement to the irrationalist claim, apparently coined by a contemporary German admirer of Hamann, Carl von Moser (who also concocted the title for Hamann as the ‘Magus of the North’), referred to him as the ‘Prussian Burke’. This was due perhaps to Hamann’s use of the concept of tradition. Again, no matter that for Moser this was a compliment, it is unlikely that Hamann would have accepted the moniker. Hamann’s, like Herder’s, overt anti-monarchical demeanour and clear republican sympathies do not rest with Burke’s conservatism. Some have also seen Hamann as a religious mystic, comparable to figures such as Jacob Böhme, within the German mystical tradition (White Beck Reference White Beck1969: 376–7). Unquestionably, the religious theme is obvious in his work, but for Hamann, his ideas had little or nothing mystical about them, if anything there was a curious religious realism (of sorts). Bringing matters closer to the present, the fact that Hamann spends such so much time excoriating rationalism and highlighting existence has led some commentators, initially Walter Lowrie, to identify him as a proto-existentialist (Lowrie Reference Lowrie2016). Lowrie’s judgement links closely to the work of Søren Kierkegaard, often seen as one of the founding figures of philosophical existentialism. Lowrie was deeply involved in editing and translating Kierkegaard’s works into English for Princeton University Press. Further Kierkegaard himself was indubitably an aficionado of Hamann’s work and mentions him positively, for example, in his Concluding Unscientific Postscript.

Finally, it is worth underscoring the point that Hamann – the critic of the Enlightenment – has now become for some the post-secular critic of modernity. Let me just mention one example in recent philosophy. Charles Taylor has made a singular point of articulating what he refers to as the expressivist-constitutive account of language, which he distinguishes from a designative-instrumental account (Taylor Reference Taylor and Taylor1985).Footnote ⁶ The designative indicates a conception of language where words and meanings refer, represent, point to, or designate external things or objects, largely from the standpoint of the subject’s private consciousness. In effect, thought (and reason) precedes language. Language is deemed as a mere assistance to pure conceptual thought. The expressive idea, sketchily put, alternatively views language as something integral to social existence, tradition, and history. Words are not simply ciphers, but conversely the thoughts and speech-acts themselves.Footnote ⁷ Hamann is taken by Taylor as a root to the expressive perspective. In fact, in Taylor’s various writings, Hamann is usually linked up with Herder and Humboldt in what Taylor calls the ‘H-H-H’ expressivist tradition, contrasted to the ‘HLC’ tradition (Hobbes, Locke, Condillac) representing the designative-instrumental conception of language.

4. Brief overview of the Metacritique

In this particular essay, Hamann is essentially querying a key assumption of the Enlightenment concerning the central role of reason in human affairs. Kant, in this context, is seen by Hamann to hypostatise reason by assuming a self-sufficient entia rationis (i.e. reason in a virtual Platonic sense) that exists separately from the realms of language, history, and experience. In effect, Kant is seen to purify reason from tradition, custom, experience, and most problematically, language. Yet paradoxically, for Hamann, underlying Kant’s notion of reason is still language. In addition, he sees language as rooted phenomenally in the world of sensibility and experience. Thus, the purported Kantian purity of reason is in fact a delusion. Reason is essentially simply a feature of language and experience. One very rapid way of characterising Hamann’s aim here is that he was trying to replace a critique of reason with a critique of language.

With regard to the ‘Metacritique’, there are a couple of initial points to note: first, it is worth recording that the Critique of Pure Reason is the only Critique that Hamann explicitly knew, outside of his conversations with Kant. Hamann died in 1788 and had been failing in health for a few years; it is therefore unlikely that he knew the B edition of the Critique of Pure Reason. Further, the Critique of Practical Reason was published in 1788, the year of Hamann’s death, followed by the later Critique of the Power of Judgement (1790). Hamann does speak in a letter of anticipating with curiosity the Groundwork of the Metaphysics of Morals (1785), which he does indeed make some random comments on (see Bayer Reference Bayer2012: 16). He also remarks critically upon some of Kant’s essays, particularly ‘What is Enlightenment?’ (see Hamann ‘Letter to Christian Jacob Kraus’ in Schmidt ed.: Reference Hamann and Schmidt1996). Second, what Hamann does write about Kant is often compressed and occasionally elusive. Even Kant beseeched him to write in understandable language, remarking in a letter to Hamann, ‘Poor son of earth that I am, I am not organised to understand the divine language of the intuitive reason’ (Kant, letter 6 April 1774, in Gregor Smith Reference Gregor Smith1960: 47). Paradoxically, however, if Hamann had expanded his Metacritique into, say, a systematic monograph, which some of his friends wanted him to do, he would, in his own terms, have betrayed his own idiosyncratic anti-systemic position.Footnote ⁸

Turning to the Metacritique essay: the term ‘metacritique’ largely encompasses his reaction to Kant. It is Hamann’s neologism and has subsequently, of course, merged into the philosophical lexicon (Bayer Reference Bayer2012: 153). At the time of its drafting, Hamann was also working on his translation of Hume’s Dialogues.Footnote ⁹ If anything, with hindsight, he probably had more overall respect for Hume’s than Kant’s philosophy.Footnote ¹⁰ However, he was also clearly very attentive to Hume’s impact upon Kant. When feeling more considerate, he thus occasionally referred to Kant in letters as the ‘Prussian Hume’.

Hamann elaborates, in the Metacritique, on what he refers to as three ‘purifications’ of reason within Kant’s perspective. The first is the ‘attempt to make reason independent of all tradition and custom and belief in them’. The second is the sanitisation of reason from all ‘experience and its everyday induction’. The third purification is where reason is cleansed from empirical impurities, particularly the impurities present in ordinary language. This third purification is Kant’s most significant error for Hamann, who indicates that language must, on the contrary, be ‘the first, and last organon and criterion of reason’ (Hamann in Haynes ed.: Reference Hamann and Haynes2007, 207–8).Footnote ¹¹ Kant is thus seen to abstract reason from sense experience, tradition, and most problematically from language. Briefly, I want to expand on Hamann’s three purifications.

5. Autonomous reason

For Hamann, Kant assumes problematically a priority to pure reason.Footnote ¹² For Hamann, the opposite is in fact true, namely, that language is prior to reason. Reason is simply a word within language. A direct consequence of the Kantian priority of reason for Hamann is, on the one hand (what he refers to as), a ‘mystical’ Platonic love of pure abstract form (e.g. in pure mathematics), and, on the other hand, a ‘gnostic hatred’ of sensory materiality, that is, in separating pure reason from sensory experience. Hence, an exemplar of this mistaken priority of pure reason is, for Hamann, the Kantian perspective on mathematics. As Kant, for example, had commented ‘mathematics gives us a shining example of how far, independently of experience, we can progress in a priori knowledge’ (A712/B740).Footnote ¹³ Hamann notes here what he calls Kant’s ‘cold prejudice for mathematics’, indicating that mathematical reason is understood as a model of pure, self-evident, a priori knowledge, independent of experience. The conceptual certainty of mathematics is not, in itself, the problem here for Hamann. It is conversely the potential idealisation of this mathematically orientated pure reason as a model form of abstract universal clarity.

Hamann’s critical argument here rests on two points: first, mathematics, in essence, is still a language. In fact, for Hamann, ‘the entire faculty of thought [is] founded on language’ (Hamann in Haynes ed., Reference Hamann and Haynes2007: 211). Consequently, all the purported certainties of mathematics necessarily remain within the bounds of language and experience. As Hamann commented in a letter in 1785, ‘the whole certainty of mathematics depends upon the nature of its language’; thus, mathematical proof essentially rests upon a particular kind of symbolic language and indeed, for Hamann, upon an imaginative ‘poetic license’, assuming, for example, ‘lines and surfaces which are physically impossible’ (Hamann in Gregor Smith ed. Reference Hamann and Gregor Smith1960: 246). For Hamann, mathematics works with the ‘simplest, most sensible intuition’; it emphasises ‘the ease of proving and representing its synthesis… in obvious constructions or symbolic formulas and equations, by whose sensibility all misunderstanding is excluded of itself’ (Hamann in Haynes ed., Reference Hamann and Haynes2007: 210). The problem here, for Hamann, is that many then suppose, mistakenly, that these ‘simple’ and ‘sensible intuitions’ – constituting mathematical language – are a manifestation of a universal reason, that is, entia rationis, distinct from the experiential realm of language. For Hamann, this is simply a delusion on Kant’s part.

Second, for Hamann, the abstract form of mathematics should not be a model of clarity for language in general. Ordinary language is infinitely more complex and mysterious for Hamann. His argument here is therefore that mathematics misleadingly lays claim to a ‘privilege of nobility because of its universal and necessary reliability’ (Hamann in Haynes ed., Reference Hamann and Haynes2007: 211). For Hamann, mathematical symbols and characters are ‘kyriological’; that is, they merely designate objects by means of conventional signs, hieroglyphs, or symbols. Apodeictic certainty is limited to this particular set of signs and symbols. Hence, mathematical certainties only function effectively within this constrained symbolic domain. As such mathematics, as a language, can never do justice to the richness and sheer diversity of human thought, language, and experience. For Hamann, human life and experience always exceed and surpass such a constrained set of abstract symbols. Thus, mathematical certainty is not the kind we seek in matters of our existence (see Griffiths-Dickson Reference Griffiths-Dickson1995: 290). As O’Flaherty neatly puts Hamann’s point, progress in pure reason, in this mathematical setting, always means ‘egress from life’ (O’Flaherty Reference O’Flaherty1952: 86). For Hamann, if the exactness of mathematics was our highest human achievement, we would have little more than the ‘unerring instincts of insects’ (Hamann in Haynes ed., Reference Hamann and Haynes2007: 211).Footnote ¹⁴

In summary, mathematics for Hamann is an idiosyncratic offshoot of language. It works with simple intuitions, transforming them into ‘hieroglyphs and types of ideal relations’ (Hamann in Haynes ed., Reference Hamann and Haynes2007: 210). However, when such hieroglyphs, ideal relations, and symbols are idealised as the epitome of pure a priori reason and made into a model of universal certainty, independent of experience, then for Hamann, we are being essentially presented with (what he refers to as) a ‘magic shadow play’, peopled by mysterious entia rationis, largely functioning as ‘empty sacks and slogans’, ultimately distorting and corrupting the richness of both language and reality (Hamann in Haynes ed., Reference Hamann and Haynes2007: 210). Any philosophy that therefore takes such an idealised pure view of mathematics creates a body of unreal problems and distorts both human experience and ordinary language.

6. Reason and experience

A second theme of the Metacritique is the argument that Kant misconstrues the relation between reason and experience. Responding to Kant’s general observation that ‘the chief question is always this – what and how much can the understanding and reason know apart from all experience?’ (Avii), Hamann gives an unexpectedly brisk answer, namely, that language explains the possibility of both experience and reason. This latter point, in passing, also underpins Hamann’s understanding of the Kantian antinomies and paralogisms.Footnote ¹⁵ He takes what looks, prima facie, like a more idiosyncratic empiricist or sensuous approach here.Footnote ¹⁶ Accordingly, space and time are, from one perspective, uncomplicated bodily, sensuous, physiological experiences, concerned primarily with sight and hearing.Footnote ¹⁷ Time finds its oldest source in the ‘temporal measures and intervals of the human body’, that is, the ‘palpable rhythms’ of the human pulse or breathing and indeed eventually music. Space, on the other hand, finds its origin in human sensuous activities of, for example, painting and drawing. If anything, for Hamann, can claim the earliest title of a priori here, it would therefore be poetry or music and the direct physicality of drawing or painting.

However, to be clear here, for Hamann, this initial prima facie sensuous or empirical appearance of space and time still needs to be grasped within the fuller context of language. Thus, on the one hand, there exists the sensuous felt experience of space and time, and, on the other hand, there exists the conceptual understanding of space and time. For Hamann importantly, these are ‘two stems of human knowledge’ derived from ‘One common root’ (Hamann in Haynes ed., Reference Hamann and Haynes2007: 212). This one common root is popular or ordinary language – a root which should not be abstractly fragmented or divorced into independent spheres or faculties, such as sensibility and understanding or reason and experience. Thus, Hamann notes, ‘popular language gives us the finest parable of the hypostatic union of the sensible and intelligible natures, the joint communication of the idiom of their powers’ (Hamann in Haynes ed., Reference Hamann and Haynes2007: 213). In such ordinary language terms, the word is made flesh. Accordingly, there can be, for Hamann, no arbitrary divorce of the sensuous and conceptual features of, for example, space and time. The words of ordinary language (the common root) combine sensible (and aesthetic) and logical aspects, that is, the visible or audible with the conceptual.Footnote ¹⁸ Hamann’s general philosophical point here is that any overly sharp distinction between the rational and sensible, in effect, cuts off reason from language. Thus, for Hamann, ‘words are pure and empirical intuitions as much as pure and empirical concepts’. Meaning arises then from a simultaneous combination, or as Hamann refers to it, a ‘reiterated bond’ of the sensible and conceptual within ordinary language (Hamann in Haynes ed., Reference Hamann and Haynes2007: 215–6).

Thus, for Hamann, one casualty of Reason’s dominance in philosophical discussion is the distinction between sensibility and understanding. This latter distinction is important within Kant’s epistemology yet is described by Hamann as both ‘violent’ and ‘unjustified’. This criticism manifests Hamann’s oft-stated idea of a ‘great divorce’ – a divorce referring once again to the unnecessary and artificial fragmentation of the prior monistic unity of language. The Kantian distinction is thus seen to capriciously hack ordinary language into fragments, via the abstract formal distinctions of reason. Hamann’s solution to this symbolic ‘violence’ is to reiterate, once again, the erstwhile unity of language. Hence both sensibility and understanding have their everyday roots in language use. They therefore have no faculty status outside of language (Hamann in Haynes ed., Reference Hamann and Haynes2007: 213). Consequently, ‘understanding’ and ‘sensibility’ and likewise the recognisable Kantian terms ‘a priori’ and ‘a posteriori’ are and remain for Hamann simply words within ordinary language, nothing more and nothing less.Footnote ¹⁹

In sum, to reiterate the argument, Hamann argued that words amalgamate sensibility – in terms of audible or visible signs and symbols – and conceptual understanding. Words are our closest estimate of reality. A thing or object can only appear to us clothed in words. Both experience and thought are blended and fused in language and word usage. Reason, in this context, is simply just another word. Reason can therefore neither be pure nor autonomous from experience. In Kant, Hamann therefore saw the grandiose artifice of pure reason as a symbolic artifice that essentially divorced humanity from common human sensory experience.

7. Hume, Kant, and Hamann

I want to pause for a moment here on the issue of Hume’s role in this whole debate between Kant and Hamann. Hume had a potent effect on both thinkers. Kant, for example, saw himself as amplifying and correcting Hume’s scepticism via transcendental idealism. Sceptical reason was thus enlisted as a central pillar of the first Critique, largely to help undermine the dogmatic understanding of reason.Footnote ²⁰ Many have indeed taken critical comfort from Kant’s remarks in the 1^st edition Preface that ‘our age is, in especial degree, the age of criticism, and to criticism everything must submit’, and religion, as Kant adds, is no exception (Axii, note a). However, other philosophical contemporaries, such as Moses Mendelssohn, were deeply disturbed by Kant’s deployment of this sense of Humean reason.Footnote ²¹ For Kant, however, Hume – the ‘geographer of human reason’ (A760/B788) – performed an initial vital service to philosophy by showing that, for example, causal necessity could not be justified, that is, as existing substantially within objects and appearances. Hume, for Kant, expelled such questions beyond the ‘horizon of reason’. Hume for Kant was clearly not denying the customary usage of causality (quid facti) but was conversely rejecting the claim that reason could account for it (quid juris).Footnote ²² Habit and custom were thus taken by Hume as prior to reason in explaining the appearances.Footnote ²³

As is well known, Kant’s argument here is that Hume moves the argument away from the dogmatic metaphysics of reason. Yet what Hume failed to grasp, for Kant, was that experience had to be grasped in relation to the faculties (conditions) necessary for the possibility of experience itself. Hence, the appearance of objects relates to the manner in which we experience them. Causation becomes a conditional term embedded within the possibility of experience. Kant saw this argument as both extending Hume’s philosophy and transforming it into a systematic account of reason. Hume’s error for Kant was that he did not grasp the difference between the (transcendental) conditions of our knowledge of appearances, as distinct from knowledge of objects in themselves. Reason, in this context, is not a metaphysical explanation of entities in nature (in themselves); conversely, it shows us our own projections of order (and purpose) in the appearance of objects. Transcendental idealism, with Hume’s sceptical assistance, thus places reason at the centre of human activity, although it is a sceptically tempered and limited conception that has abandoned all dogmatic metaphysical claims. It has accepted what Kant refers to as Hume’s ‘censorship of reason’ but reconceived as a stage on the pathway towards a conception of the ‘power of judgment sharpened by experience’ (A761/B789). Reason cannot for Kant get beyond this critical setting without pitching into either extreme scepticism or alternatively fantasies or antinomies.

For Hamann, like Kant, Hume is also seen as a philosophical ally in the struggle to curtail dogmatic reason. There is also a direct comparison here with Hamann’s view of Socrates (Hamann in Gregor Smith Reference Gregor Smith1960: 175–84). Socratic ignorance was not a propaedeutic to a more complete rational knowledge; it was rather an affirmation of the severe limits of reason. The key hazard of reason is again its over-estimation of its abilities and powers.Footnote ²⁴ Further Hamann, like Kant, embraced Hume’s critique of natural theology and design arguments. Arguably, this is likewise what Kant immediately took away from reading Hamann’s translation of Hume’s Dialogues Concerning Natural Religion. The religious dimension is key here in Hamann, although, once again, it should not divert us from a resonance with Kant. It is true that Hume, in effect, fed Hamann’s extreme fideism. The sceptical assault on reason was thus seen to open the door to religious faith (see Graubner Reference Graubner1989: 383).Footnote ²⁵ Yet we should not imagine, even at this stage, that Hamann is in a different ballpark to Kant. For Kant, recognisably, ‘I cannot even assume God, freedom and immortality for the sake of the necessary practical use of my reason unless I simultaneously deprive speculative reason of its pretension to extravagant insights’; as he continued, thus ‘I had to deny knowledge in order to make room for faith’ (Bxxx).

Oddly, so far, Kant and Hamann are roughly on the same page with regard to Hume, with the exception of Hamann’s lively enthusiasm for faith, Kant being altogether more sober. If this Humean view was the end of the matter, then Kant could still have argued that Hamann had not grasped that reason is still eminently salvageable through transcendental idealism. Similarly, he could have further argued that Hamann did not understand the crucial evolution from Humean sceptical reason to Kant’s own unique conception of critical reason. Seemingly, Hamann also possibly confused critical reason with dogmatic reason, the latter, in Kant’s reading, being characteristic of the tradition of classical European rationalism, as in Wolff or Leibniz, something Kant clearly rejected in his first Critique.Footnote ²⁶ Hamann, in essence, therefore failed to see the difference between the (transcendental) conditions of our knowledge of appearances, and the knowledge of objects in themselves (realism or naturalism). The capacity of reason to view objects and appearances in accordance with lawlike regularity (regulatively) is consequently apparently missed by Hamann in his overall appraisal of Kant. It could almost be said that Hamann was blinded to Kant’s critical innovations by taking Humean scepticism and empiricism far too seriously. As Hamann admitted, he preferred ‘to labour…. with doubt… against the dictators of pure reason’ (Hamann quoted in Bayer Reference Bayer2012: 44). Thus, there is an important question here as to whether Hamann really grasped Kant’s transcendental argumentation, namely, that in order to comprehend the possibility of natural science, it is essential to relinquish Hume’s empiricist suppositions and to acknowledge that it is the activity of the a priori transcendental understanding, which provides the crucial framework for experience. It is this transcendental framework that organises our sense of the objective world. Thus, did Hamann’s fundamentally misunderstand Kant’s form of transcendental argument?

8. Transcendental argument, metaphysics, and words

The above question – on transcendental arguments – might look like a potentially neat conclusion to the Kant/Hamann debate, except that it is not. We need to refocus on the next stage in Hamann’s argument which embodies his critique of Kant’s transcendental argument. Hamann’s argument here takes a slightly unexpected shape. Hamann centres on the use of words in Kant’s first Critique text. Thus, are words simply subsumed into concepts and the understanding?

For Hamann, words tend to arise in Kant’s text as (what he calls) ‘spectres, non-words or unwords’. They ‘become determinate objects for the understanding only through their institution and meaning’ (Hamann in Haynes ed., Reference Hamann and Haynes2007: 216). The trouble here, for Hamann, lies in Kant’s transcendental idealism. Thus, the word reason is taken by Kant to be crucial for all genuine critical thinking. Hamann then asks the slightly elusive question: ‘Is it possible from the matter of the word “reason”, from its 7 letters or 2 syllables in German – is it possible from its form which determines the order of these letters and syllables – to elicit something of the concept of the word “reason”?’ – thus moving from the audible and/or visible word “reason” to the concept of reason. Hamann continues in the same vein: ‘From the opposite pole, is it possible “to find from the concept of reason the matter of its name, that is, the 7 letters or 2 syllables in German or any other language?”’ (Hamann in Haynes ed. Reference Hamann and Haynes2007: 216). Thus, can one move from the concept ‘reason’ to the audible and visible materiality of a word of 7 letters and 2 syllables? Or, alternatively, can one pass from the sensuous materiality of the word reason to deduce the concept? Hamann claims here that Kant, in effect, personifies reason or makes it into a mysterious active presence (entia rationis). In Hamann’s terms, he ‘has dreamed the universal character of a philosophical language’. Such a mystical presence (of reason), Hamann suggests, belongs ‘to the categorical and idealist woods’ (Hamann in Haynes ed., Reference Hamann and Haynes2007: 217). In other words, for Hamann, there is something deeply perplexing at the heart of the transcendental argument, and the Metacritique might thus be described as an investigation of Kantian reason in the context of language.

The theme of words thus introduces Hamann’s own idiosyncratic take on the concept of reason and thus transcendental argument. As he noted in a letter to Jacobi: ‘With me it is not so much the question: what is reason? but far more: what is language? and this I suppose to be the ground of all paralogisms and antinomies’ (Hamann quoted in Griffiths-Dickson Reference Griffiths-Dickson1995: 311). In effect, the word reason, in this setting, cannot be extracted from language and made into something else, namely, a critical presence or separate entity. Hamann’s argument is not so much a rejection of reason but rather a firm reminder that reason is just a word and nothing more than a word. This latter argument is the core of Hamann’s critique of Kant’s prosopopoeia, that is, hypostasising reason into a something, or minimally a process that is somehow separated from language and experience. There is clearly something to Hamann’s claim here that reason can appear, at points in Kant’s texts, as a virtually autonomous entity or thing that has the power to command us. That is to say, almost imperceptibly, a verb or an activity (i.e. reasoning) can appear to mutate into an active noun. Reason becomes, as it were, personified. Yet, this entity in Kant remains a fiction for Hamann. There is nothing ‘behind’, ‘before’, ‘above’ or ‘after’ language.Footnote ²⁷ Hamann called his own position here ‘verbalism’.

In passing, it should be here noted that Hamann took a very dim view of metaphysics, in whatever form it appeared. He considered the discipline to be nothing but the ‘lukewarm hieroglyphs’ of mysticism, and thus, it had to be excluded from serious discussion (see Hamann in Haynes ed., Reference Hamann and Haynes2007: 210). Like Hume, he was more than happy for all such books of metaphysics to be committed to the flames. Further, it is important to note that he did not regard his own suppositions as in any way metaphysical.Footnote ²⁸ His core focus on idiosyncratic linguistic realism essentially denied the need for any metaphysical assumptions. As indicated earlier, the basic idea for Hamann is that language and thought are deeply interwoven and constitute what we think of as reality.Footnote ²⁹ Further, language (as a common root) is still a phenomenal and sensuous activity. However, as indicated earlier, there is an important distinction to acknowledge here, namely, between ordinary and more abstracted rational language. Ordinary language, in this reading, is phenomenal and moulded into reality as a ‘reiterated bond’. Hamann saw himself as a non-metaphysical linguistic realist on this issue. Sense, feeling, imagination, and emotion were all blended within ordinary language usage. Further, there were empirically real objects, which ordinary language embodied or encapsulated in words. These real objects, as encountered within word signs, constituted what Hamann referred to as sense units. Hamann thus sees reality in the mould of immediately intuited objects and relations – as encased simultaneously in ordinary sign usage.Footnote ³⁰

The implicit distinction here is therefore between a symbolism premised on ordinary word usage (in ordinary language) and a symbolism that is artificed via abstract reason. Hence, for Hamann, it is empirically the case that both objects and their manifold relations do exist. Such relations between objects are not however the product of reason; they rather sensuously exist in ordinary language.Footnote ³¹ Further, only natural or ordinary languages embody such sense units. Intuitive empirical relations are existentially concrete; they only become unreal as they become rational abstractions. As such, they comprise what Hamann referred to as naïve ‘impromptus of language’.

One other consequence of Hamann’s idiosyncratic linguistic realism is his denunciation of abstractly or logically motivated language and spelling reform. In essays such as ‘Miscellaneous Remarks concerning Word Order in the French Language’ (1760) and later a ‘New Apology of the Letter h’ (1773) Hamann castigates, for example, the likes of the French Academy’s attempt to reform word order via the canons of reason and then later with regard to spelling in the German and Latin languages (see Hamann in Haynes ed., Reference Hamann and Haynes2007: 20–32 and 146–63). Pure reason, for Hamann, strikes again here in a different setting, in this case mutilating the lush complex diversity of ordinary language, word order, and spelling. In consequence, Hamann sees ordinary language impoverished once again by the intervention of reason. As he commented, ‘The purity of a language diminishes its riches; a too strict correctness diminishes its strength’ (Hamann quoted in O’Flaherty Reference O’Flaherty1952: 20).

It is worth noting here, very briefly, the following summary points on Hamann’s position: philosophical problems are caused either by a rational misuse of language or alternatively trying to ignore the place of language in cognition. To acknowledge the core role of ordinary language in cognition is then to dissolve philosophical problems. Language is always public, and there can be no such thing as a purely private language. The human mind always subsists in public linguistic symbolism, and such symbolism is embedded in multifaceted ordinary usage. Meaning is always intertwined with ordinary use. One cannot therefore get behind the back of language; there is thus no super-rationality standing outside of language. We might, of course, immediately recognise some of these philosophical themes in Wittgenstein’s later philosophy of language, and some more recent scholarship has indeed uncovered Wittgenstein’s own close and very sympathetic interest in Hamann’s work, particularly in the context of late works such as On Certainty (see, e.g. Snellman Reference Snellman2018 and Halpern Reference Halpern2019).

9. Kant, reason, and language

How accurate is Hamann’s account here of the priority of reason in Kant? Further, to what extent does Kant fail to grasp the significance of language? On one count, Hamann is clearly correct, namely, that Kant does undoubtedly give reason a primary role in his system. Speaking of his Critique of Pure Reason, in his Prolegomena, Kant remarks that it is a difficult work ‘and requires a resolute reader to think himself little by little into a system that takes no foundation as given - except reason itself’ (P, 4:274). This is also the gist of his essay ‘What does it mean to orient oneself in thinking?’ (1786). The faculty of reason he claims, in this latter work, is ‘implanted by nature and made habitual through frequent practice’ (WOT, 8:135). A little mysteriously, Kant also speaks here of reason having a ‘feeling of its own need’ (WOT, 8:136). Reason orientates us – analogous Kant suggests to a compass – even with regard to the supersensible, although he admits that there is more than enough to deal with in the realm of sense. In terms of practical reason, it further provides the groundwork for freedom, partly because practical reason sets its own laws, although again he adds, ‘reason alone can command validly for everyone’ (WOT, 8:145). The universality of reason needs however to be kept distinct, for Kant, from the diversity and plurality of languages. Reason is viewed as the very ‘touchstone of truth’, and failing it makes you ‘unworthy of freedom’ (WOT, 8:146). Making use of our reason implies for Kant asking oneself ‘whenever one is supposed to assume something, whether one could find it feasible to make the ground or the rule on which one assumes it into a universal principle… The test is one that everyone can apply to himself; and with this examination he will see superstition and enthusiasm disappear’ (WOT, 8:146 note).Footnote ³²

So where does the above eulogy on reason leave language in Kant’s overall system? Kant, despite the obsessive attention of the likes of Hamann and others, seems oddly to pay little or no overt attention to it. As Ernst Cassirer commented in the 1920s, on Kant’s Critique of Pure Reason, ‘the attempt to critique reason without providing a critique of language appears to be Kant’s real, incomprehensible oversight’ (Cassirer Reference Cassirer2013: 105). Aside from the idiosyncratic work of Herder, Humboldt, and later Cassirer, it remains an anomalous fact of mainstream Kant scholarship, even to the present day, that very little is written on the question of language.Footnote ³³ I suspect that this is largely because Kant himself (as Cassirer noted), in the main, ignored or simply skirted around the question. It would, however, have been intrinsically deeply interesting if Kant had tried to address the question, but the fact remains that he did not and we are consequently left with a void.

Nonetheless, despite the lack of any systematic responses in Kant, there are still some random, slightly indirect, examples where we do get a vague sense of what Kant thought about language.Footnote ³⁴ One rather odd reference appears in Anthropology on the question of the use of signs and symbols. Kant indicates that there is a ‘faculty’ of using signs. Signs he suggests ‘signify something through association with intuitions and then leading through them to concepts’ (Anth, 7:191). Symbolic cognition and the use of signs are still, for Kant, very different from discursive cognition. He concludes that ‘symbols are merely means that understanding uses to provide the concept with meaning’ (Anth, 7:191). The sign and symbol are thus embedded in language, but he continues, something significant is lacking if the agent ‘can only express himself symbolically’; this lack is seen as conceptual. Hence, those who dwell only in this symbolic or sign-based mode show nothing but a ‘poverty in concepts and, therefore, also in the words to express them’ (Anth, 7:191). Recognisably – and returning to the theme of an early work Dreams of a Spirit Seer (1766) – Kant remarks that those who remain simply with signs and symbols of the language are ‘like Swedenborg’ and ‘dwell upon a hidden world behind appearance’, and the best one can say about them is that they are ‘enthusiasts’ (Anth, 7:192). Whether this latter critical view was – in small part – addressed indirectly to Hamann remains unknown.

Thus, language in Kant, in essence, is simply a constellation of signs and symbols. It is a means of signification (or ‘signalling’), and language is ‘the best way of signifying thought’. It is, he suggests, ‘the greatest instrument for understanding ourselves and others’ (Anth, 7:193). But language, as such, simply remains a collection of vacuous signs and symbols without concepts and ultimately the independent activity of Reason. But Kant adds here, ‘even those who can speak and hear do not always understand themselves or others’. This he suggests, ‘is due [either] to the lack of the faculty of signification, or its faulty use (where signs are taken for things, and vice versa)’. This, for Kant, is particularly important ‘in matters of reason’, because ‘human beings who are united in language are as distant from earth in concepts’. For Kant, therefore, ‘it is enlightenment to distinguish the symbolic from the intellectual’ (Anth, 7:193). Thus, speaking and sharing a common language, that is, sharing and using its symbols and signs, is a very different world for Kant from grasping concepts and reason. In essence, Kant appears to be arguing here that despite a plurality of different languages (as symbolic systems), reason remains a constant universal over and above such a symbolic plurality of languages.

There is another oddly illuminating, if passing, reference to language in the Prolegomena to any Future Metaphysics. This provides a slightly different perspective again. In speaking about the categories, which are, as he puts it, simply ‘present in all cognition from experience’, Kant suggests that they are comparable to the rules of grammar in language. They form a necessary underpinning for meaningful communication and experience of the world.Footnote ³⁵ There are thus ‘language rules for the actual use of words in general’, which can be compiled for meaningful usage. Kant proposes therefore that the role such categories and grammatical rules play are ‘very closely related to one another’ (P, 4:323–4). Categories and rules underpin communication.Footnote ³⁶ The a priori is viewed, in this latter context, as a logical underwriting of meaningful language, a language that has moved beyond the basic symbolic aspect. Such a logical underpinning is then perceived as a necessary aspect of the schematism for all meaningful language. Such logical conditions, embedded in language, can then be publicly cross-referenced and analysed.Footnote ³⁷ For Kant, this is the manner in which we organise epistemically our intuitions and sensible experiences. Kant is neither speaking about ontology nor any supersensible aspect at this point. He is rather trying to indicate, at one remove, the transcendental conditions of human knowing and critical public communication.Footnote ³⁸ This is then our closest approximation to truth and objectivity. Hence private faith or private meanings are distinct from public meaning, which for Kant is another way of making a space for faith. Divine language, vis-à-vis Hamann or Swedenborg, understood as a kind of mysterium tremendum, related to an inexpressible God, is viewed by Kant as – charitably – just private enthusiasm.Footnote ³⁹ Hamann, in response to Kant, would no doubt have, once again, claimed here that the Kantian argument on categories was just another strategy for avoiding the language question and consequently piling up more abstract words and mysterious entia rationis.

10. Critical concluding remarks

There are certain intrinsic constraints governing the debate between Kant and Hamann that prevent a conclusive overview or assessment. As indicated earlier, Hamann was clearly not fully acquainted with the full range of Kant’s writings; he died before many of the important developments in Kant’s thought had even taken place, including, for example, the B edition of his first Critique. This simple fact alone inevitably places limits on how far we take Hamann’s critique. Further, we don’t know exactly how Kant might have responded philosophically to the issue of language in either Hamann or indeed Herder. As some Kant scholars have noted, the issue of language remains an inexplicable gap in Kant’s systematic thought. All that we do have in Kant’s works are some random and unsystematic comments on the language question. In this incomplete sense, Hamann was indubitably correct in his assessment of Kant’s Critique of Pure Reason, namely, that it did neglect the issue of language. In this restricted context, it also follows that Hamann’s critique of Kantian reason – as wholly neglecting the question of language – remains a credible and defensible position. We can, of course, speculate how Kant might have responded to Hamann’s arguments. I suggested earlier, for example, that Kant might have used the same transcendental argument (as used against Hume) in this case against Hamann. However, as pointed out in the discussion, Hamann would undoubtedly have responded immediately by turning the linguistic argument back on Kant, namely, that transcendental reason remains just another facet of language. That is to say, both reason and metaphysics, in whatever shape they appear, remain just words and nothing more than words. Again, it is important to emphasise in Hamann’s work that he rejected – in parallel to Hume – the whole disciplinary architecture of metaphysics. Ordinary language, in all its convoluted complexity, was reality for Hamann; hence, no metaphysical abstractions or assumptions were required or implied. In this general setting, the debate between Kant and Hamann, on the question of reason and language, looks like a philosophical stalemate of sorts.

However, there are further criticisms that can be made against Hamann, not usually covered in Hamann’s scholarship, which address his idiosyncratic conception of language. The first is something I have avoided talking about until the close of the previous section, largely because it can allow too quick a dismissal of Hamann. This is the argument that ordinary language rests ultimately upon divine language. Ordinary language, in all its sensuous often contradictory complexity, carries for Hamann a divine revelation that needs to be deciphered prophetically or poetically.Footnote ⁴⁰ In point, poetic or prophetic language for Hamann exhibits this divine spirit most effectively. However, how exactly do we interpret this aspect of Hamann’s work? Language in Hamann might, more sympathetically, be seen here as theonomous – that is, as a vehicle of religious faith. Faith is not antagonistic to reason, as he stated in a letter to Jacobi: ‘Faith needs reason, as reason needs faith’ (Hamann quoted in Gregor Smith Reference Gregor Smith1960: 59). What reason can do in this setting is to inaugurate Socratic ignorance, which constitutes a learning experience, allowing faith to evolve. To have faith is therefore to dwell in a learned ignorance.Footnote ⁴¹ For Hamann, this is not a shallow ignorance but is conversely a process concerning one’s whole existence.Footnote ⁴² This perspective, in point, has distinct echoes of Kierkegaard’s work. In fact, one might (as Walter Lowrie suggested) describe Hamann here as a proto-Christian existentialist. Existence therefore precedes conceptual essence.Footnote ⁴³

The above existential point might provide a little more credence or warrant to Hamann’s religious reading of language, although to modern ears it can still strain credulity. One further significant oddity about Hamann’s work here is that he never offers a clear definition of language. Given his critical views on reason and abstraction, one could possibly see why he might not want to move in this definitional direction, but it still remains an unsatisfactory lacuna. How do we think about language if we don’t know what it is precisely? Simply referring to poetic and prophetic language as a preferred medium does not exactly assist our understanding of the question of language. Further, the distinction made by Hamann between ordinary and abstract language also remains vague. Hamann plainly tries to utilise the distinction in his defence of ordinary or popular language, but, prima facie, there seems to be nothing ordinary about the claim that the ordinary is saturated with the divine.

Leaving aside the divine, we might still find the ‘ordinary’ as intrinsically puzzling. There is an unexpected echo here in the ordinary language movement of mid-twentieth-century British philosophy, namely, the view, for example, that the common stock of words contains all the important distinctions that are worth drawing. Ordinary language is convenient on one level since it does provide ‘a very simple way of disposing of immense quantities of metaphysical and other argument, without the smallest trouble or exertion’ (Heath Reference Heath and Lewis1963: 185). Yet it is still not clear why we should place such an inflexible reliance upon the guidance of ordinary use in language.Footnote ⁴⁴ There are, for example, no pre-existent consensual meanings for either ‘use’ or indeed ‘ordinary’. Thus, in essence, why are ‘ordinary’ and ‘use’ given such an authoritative status in Hamann? Who speaks authoritatively for the ordinary? Surely all such words are subject to quite reasonable conceptual articulation and contestation?Footnote ⁴⁵ It is difficult to see how Hamann could have answered these queries satisfactorily without engaging in some abstract reasoned argument.

Finally, returning to reason and put rather crudely, the real question at issue here (vis-à-vis Hamann) is this: is reason simply another word within language, or alternatively is language just another concept deployed within reasoned discussion? Further, in claiming that language and words are prior to reason, one must (I suppose) be assuming that this latter proposition is a reasonable assertion. Hamann therefore claims a special rational insight into the status of language as prior to rationality, which has the distinct aroma of circularity. More controversially, for some at least, is that there is no awareness in Hamann of the powerful, if elusive, epistemic role of the faculty of imagination (even within Kant’s first Critique), or alternatively the role of feeling within cognition in Kant’s ethical work.Footnote ⁴⁶ In this general sense, I do not share the view, put forward by some recent Hamann scholars, that Hamann is the ‘best critic of Kant’ (see Betz Reference Betz2004). He clearly is not; however, at the same time, we should not simply dismiss him out of hand. He does raise a number of pertinent philosophical issues on the place and role of language in our overall understanding.Footnote ⁴⁷