Vol. 3 No. 20 · 5 November 1981

Isaiah Berlin responds to the foregoing criticisms of his work

Isaiah Berlin

3889 words

Professor Aarsleff kindly sent me a copy of his article before publication; his courtesy enables me to provide an immediate rejoinder. I read it with mounting astonishment. Since I know Professor Aarsleff to be an erudite and scrupulous scholar, I find it difficult to understand what moved him to protest so vehemently that Vico’s view of language, which lies at the heart of his system, is wholly lacking in originality. I can only surmise that, irritated by the sudden appearance in the main path of the Enlightenment of this unwelcome intruder, with his unscientific etymological theories and the value put on them by his admirers, Professor Aarsleff simply wants him out of the way. The claims (which Professor Aarsleff regards as totally hollow) advanced by most students of Vico’s ideas, and indeed by himself, for regarding him as a boldly original thinker, spring from the belief that no one before him had said that it was only the study of the evolution of language, myth, ritual and other social institutions that made it possible to reconstruct in some degree of concrete detail the mentalities and outlooks of primitive societies and to trace the patterns of their development, stage by stage. This could, in Vico’s view, be achieved by examining men’s attitudes to God, nature, one another, and in particular their self-images as these are embodied in social institutions, especially in forms of language and of religious and artistic self-expression, connected in his mind with class conflict and social tension: these institutions provided the most vivid and accessible evidence of cultural growth. Vico stressed that his method of imaginative insight into what the world must have looked like to men, especially in early times, who wrote, spoke, worshipped, fought, dispensed justice, created works of art in specific ways, differed in principle from the methods commonly employed by natural scientists and those influenced by them in his own time.

Most students of Vico believe that his method of investigation entailed a novel approach in many fields – the theory of knowledge, aesthetics, jurisprudence, education, the study of antiquities, and, of course, anthropology and linguistics: but above all, that it was this shift in perspective that led to the idea of a culture as the expression of a developing and all-pervading Volksgeist (Aarsleff quaintly traces the origin of this concept to Condillac), for which the methods of the Cartesians and Lockeans of Vico’s time did not seem to him adequate. Professor Aarsleff maintains that this approach, so far from being original, consists of ideas widespread in the 17th century, and commonplaces at that; that so far from being an innovator’s, Vico’s central ideas are a mere echo of the utterances of far greater thinkers – Leibniz, Locke, Mersenne and so on. It follows that only crass ignorance of the thought of the 17th century – after all, one of the most intensively studied periods in the history of Western philosophy – could have misled us all so grossly, from Croce and Dilthey and Collingwood (writers not generally thought of as ignorant of this field of knowledge) to the ever-growing number of students of the Neapolitan thinker, especially in Italy, whose work fills the Vico bibliographies as well as the ‘fat volumes of papers’ on him, of which Professor Aarsleff speaks with such evident annoyance. Professor Harold Fisch, the doyen of Vichian scholarship in English-speaking countries, whom Professor Aarsleff holds particularly responsible for inflating Vico’s transatlantic reputation, is well able to stand up for himself: but it is against myself, by my own confession his respectful follower, that the main attack is made. On this, I have the following comments.

Even if Professor Aarsleff’s principal arguments were valid, this would still not prove Vico’s lack of originality. It could be said of Marx that every one of his central ideas can be found in earlier writings – indeed, Elie Halévy went so far as to call him somewhere (I do not know how seriously) ‘a gifted pupil of Hodgskin’. All Marx’s ideas may indeed have been anticipated – all, that is, save the whole, the transforming synthesis. A contemporary scholar, no less learned than Professor Aarsleff, is reported to have remarked, ‘We do not need Hume,’ because all Hume’s notions could be compounded out of the ideas of earlier sceptics. Originality is a contestable concept; most theorists have forerunners. A great deal of work has been done on Vico’s antecedents, especially by Italian scholars – Croce, Badaloni, Corsano, Cantelli, and many others; indeed, I have myself attempted to contribute to this topic, and suggested that some of the most important influences on Vico (apart from those he cites himself) are to be found among the French Protestant jurists of the 16th century, of whom Professor Aarsleff says nothing. But, be that as it may, the notion that a thinker, or a vision of nature or of human society, can be dissolved into a collection of isolated antecedent elements, and is therefore ‘not needed’, is obviously absurd. Few among the great thinkers after the Greeks would survive this test. ‘Change for a Napoleon’ – a 19th-century gold coin – ‘is not a Napoleon.’

However, Vico does not require this defence, sufficient as it is. The doctrines of Leibniz, or Locke, or Mersenne (not, curiously enough, Hobbes, whose influence Vico does acknowledge), cited by Professor Aarsleff as evidence that Vico was a mere copyist, do not, it seems to me, sustain any part of his conclusion. Let me come to his detailed charges more or less in his own order.

1. He complains that I ‘take Goethe severely to task for finding [Vico] a sibylline writer.’ This is not so. I did suggest that Goethe wrote about him but never troubled to read him. His letter speaks of ‘visions of the future’ in the New Science, when there is little prophecy, only a cyclical theory of history, in any of the writings. But Goethe is in good company: Professor Aarsleff is walking through open doors when he denies that Vico had any influence in the 18th century. Indeed, in my book I went a good deal further, and wondered who read him, then or now. Professor R.T. Clark did attempt to establish a link between Herder and Vico via Cesarotti’s writings, and I believe that a connection can be established via Calepio and Bodmer, but there is no proof that Herder had so much as heard of Vico when he first formulated his own historicist doctrines in the early 1770s. Vico remained virtually unknown outside Italy, despite a handful of isolated references elsewhere, until Jules Michelet made him famous in 1824-5. His influence in England was negligible or non-existent – on this point I am inclined to agree with René Wellek (cited in evidence against me by Professor Aarsleff) and not with Fisch. The words of Ballanche (which I used as an epigraph to my essay), that Vico’s fate was to ‘rise from the grave when he had no more to teach’ – for the Germans had done it all – are sadly true. Professor Aarsleff’s assertions that I believe, or have said, the opposite of this are groundless.

2. Now to Professor Aarsleff’s central contention: that Vico’s ideas about the evolution of linguistic forms, which do, indeed, occupy a central place in his argument, are nothing more than so many 17th-century ‘commonplaces’. That Leibniz took a life-long interest in language, especially as it relates to logic, is well-known. That, like Vico, he was interested in the affinity of languages, did not think them all to originate in Sweden, and wrote more than one thousand pages on the subject, I take on trust from Professor Aarsleff, who has probably read more words written by Leibniz than any living man. But I have yet to learn that Leibniz or Locke, or anyone in the 17th century, expounded an anthropological doctrine according to which the evolution of successive Weltanschauungen of primitive or ‘heroic’ (e.g. Homeric) societies is accurately reflected in changes in the forms of language, the study of which is therefore an indispensable instrument in determining the path of this evolution. Vico asks, and I know of no one who asked such pointed questions before him, how sentences (which may well become poetic diction in Classical Roman verse) like omnia plena Jovis – ‘everything is full of Jove’ – who is at once the sky and the father of the Olympian gods – can possibly have originated. Poseidon is both a bearded deity and all the seas of the world, Cybele is an enormous woman and also the whole earth – these are words and images which, Vico remarks, have little meaning for us, but evidently once expressed a vision of the world to those who first naturally spoke in this manner: what, then, can this vision have been? What kind of society was it to which such expressions made sense? Was this discussed by those from whom Professor Aarsleff’s ‘commonplaces’ were said to flow – Leibniz or Locke, Spinoza or the Fellows of the Royal Society? Yet it is this in Vico that so deeply fascinated Erich Auerbach, one of the great literary scholars of our time. Of this, Professor Aarsleff says not a word. Yet it is in asking questions such as these, and in suggesting how solutions might be obtained (however fancifully at times), that Vico’s originality and genius lie. This is very different from observing that languages differ, that metaphors derive from earlier reports of sense experience, or that customs vary among men – as Leibniz and Locke and their 18th-century successors, especially Montesquieu, declared. These reflections may well, for all I know, have grown to be commonplaces by the 18th century, but what Vico said was clearly far more revolutionary.

3. Professor Aarsleff rightly asserts that Vico maintained that only he who had made a thing can truly know it: therefore God, who made the external world, alone can know it, as we, who have not made it, cannot. He then, no doubt correctly, tells us that 17th-century thinkers – Bishop Wilkins, for example, and Mersenne – had said this before Vico. If Professor Aarsleff were to look again at my essay, he would find that (following men more learned than I am) I trace this doctrine, not to the 17th century, but rather further back – to Augustine, for example. This doctrine may well have become a ‘commonplace’ by the 13th century: indeed, Croce supposes Vico to have derived it from Aquinas, and, then again, perhaps from Sanchez three centuries later; Vico evidently did not need the famous Warden of Wadham, Wilkins, to tell him this. Vico’s originality consists, not in reaffirming the traditional distinction between knowledge of the man-made and that of the ‘natural’, but in declaring that mathematics, as a man-made set of rules, was not a body of objective knowledge, whencesoever derived – Platonic ideas, or the natural light, or intuitive insight into the unalterable structure of reality, or the attributes of the physical world – but was a set of arbitrary rules, applicable to, but not derived from, the observation of the real world. Leibniz was only the most gifted and famous of those who in his century took mathematics to be a system of relationships of ideas ‘that embodied certain eternal truths according to which the phenomena of nature are governed’. The conventionalist theory of mathematics (even though Buffon and Berkeley and perhaps others may have adumbrated something of this kind against the metaphysical realists of their time) became a really live issue only in this century. I do not know enough about the history of mathematical thought to be certain that some sceptic had not anticipated Vico in this view: but clearly no 17th-century rationalist could possibly have held it. Professor Aarsleff’s statements about what some 19th-century thinkers believed, or thought they believed, about the relationship of mathematics to physical reality, seem to me to be totally irrelevant. Vico was not, of course, the first to distinguish a priori from empirical, or certain from probable, judgments: these were indeed ancient ‘commonplaces’, and neither I nor, so far as I know, anyone else has ever described Vico’s acceptance of them as ‘momentous’. Once more, Professor Aarsleff seems to me to be barking up the wrong tree. The point is that for Vico mathematics does not, as it does for Plato or 17th-century rationalists, give true knowledge of some realm, whether of physical nature or the world of essences: it is more akin to a game, where the correct application of arbitrary rules ensures validity, but gives no information about anything. Even Kronecker, more than a hundred years later, did not say that the whole of mathematics is Menschenwerk. Vico’s notion that myth, legend and metaphor are doors to the past cannot owe much to Locke’s very general remark about the relationship of abstract ideas to sense, but probably goes back to the 16th century – to Bodin and the Neapolitans, of whom Badaloni has written in so illuminating a fashion. Unlike James Mill on Kant, I simply do not see what it is that Professor Aarsleff would be at.

An even bolder move by Vico was to transfer the distinction of man-made versus natural from mathematics to a field where knowledge about the world was obtainable – to the empirical sphere of human history, which men, since they have made it, can, according to him, know better than they do the external world, which they have not created. The meaning and validity of this doctrine has been disputed ever since. Suffice it to say that the concept of Verstehen, of understanding versus knowledge, which is much discussed at present, depends upon a particular interpretation of it. Vico’s view is consciously opposed to Descartes’s notorious contempt for history as a field of intellectual endeavour: this was certainly not dreamt of in Wilkins’s or Mersenne’s philosophies. Of this, there is nothing in Professor Aarsleff’s strictures against either Vico or myself: yet it is the heart of the entire matter. It is one of the philosophical discoveries (I cannot think why he believes that I allow discoveries only in the natural sciences) on which Vico’s reputation rests.

4. I did, indeed, express the belief that the possibility of a logically perfect language, which was obviously a chimera for Vico, occurs from time to time in the history of scientific rationalism. Professor Aarsleff informs the reader that ‘nothing could be further from the truth.’ These are hard words. Does Professor Aarsleff really believe that Leibniz (who thought well of Wilkins and other contemporary authors of schemes for a universal language) was not all his life haunted by the ideas originally adumbrated in De Arte Combinatoria, which he wrote in his extreme youth, and then in the Characteristica Universalis, of which he said, ‘Telescopes and microscopes have not been so useful to the eye as this instrument would be in adding to the capacity of human thought’ – so that controversies between rational men on scientific issues could easily be settled by those involved if they sat down with a pencil and slate and said to each other: ‘Let us calculate’? The Characteristica Universalis was intended to be a system that would mirror the basic structure of the world of which a priori knowledge could be obtained – in 1714 Leibniz still seems to have believed in it as a kind of general algebra that would give directions to reason; it could not correct errors of fact, but would guarantee the validity of reasoning, being a kind of alphabet of rational thought accessible to rational men everywhere. This idea may be visionary, but it is difficult to believe that it had no effect at all on Leibniz’s most famous commentator, Bertrand Russell, and his logical atomism, and so perhaps on Wittgenstein’s Tractatus. Does Professor Aarsleff truly believe that this idea is not echoed in Condillac’s words, ‘to study a science is to do nothing else than to learn a well-made [by which he meant logically-constructed] language,’ or that his disciple Turgot was not echoing this with his ‘the relation of language to philosophy is similar to the application of mathematics to physics’? Condorcet did not think too well of Condillac, yet he, in his turn, wrote an Essai d’une Langue Universelle – a language which he intended to be one of the cleansing weapons against the accumulated ignorance, superstition, irrationality and obscurantism of centuries, and which, for that very reason, was duly attacked by the Catholic reactionaries. Professor Aarsleff accuses me of saying that the intention was to substitute ‘artificial’ language for the natural languages of the world. I did not, of course, say this: the logically perfect language was to serve as an instrument of communication between rational men seeking to discover truth about the nature of things by means of scientific analysis. The dispute between those who believed that the basic method used in the pursuit of knowledge was quantitative (Calculemus! was Condorcet’s enlightened and brave, but wildly over-simple Leibnizian motto) and those who maintained that this left out the qualitative, imponderable, at times inexpressible aspects of experience, which all men live by – this battle had been joined in the 17th century, was fought vigorously towards the end of the 18th and indeed continues in our own time.

5. Professor Aarsleff tells us that one of the greatest champions of the qualitative approach, Herder, in his famous essay on the origins of language, was so deeply influenced by Condillac that he was taken to task for this by his mentor Hamann. It is my turn to speak of something as being ‘very far from the truth’. Herder did, indeed, incur the displeasure of Hamann, not for following Condillac (he did not), but for not declaring language to be a miraculous gift from God. In a famous controversy about language in the 18th century, Rousseau, who believed language to be a human invention, and Condillac, who thought it was a natural development from the cries of animals, were ranged against Suessmilch, who maintained that language was a gift conferred on Adam by a special act of the creator. Herder tried to occupy a middle ground. He denied both that language was an invention and that it had animal origins, but maintained that the very essence of being a man was virtually identical with a capacity for the use of language. The development of langauge was part and parcel of human development as such: there was therefore no real problem of its specific origin. This was a position with which Goethe concurred. But it was clearly not good enough for Hamann, and under pressure from him, Herder had to recant (but later returned to his heresy). There is no question of the influence of Condillac here. In his essay, Herder describes Condillac’s theory of language as ‘empty’ since, according to him, ‘words were formed because words existed before men existed.’ There is no point in following ‘the thread of this interpreter any further, since it is completely broken’: its only consequence was to lead Rousseau to his own equally fallacious hypothesis – for, according to Herder, there is no continuity between animals and human beings, but a clean break. Still further from the facts is Professor Aarsleff’s assertion that my belief, found only ‘in the older literature’, that Hamann was Herder’s teacher, is ‘unfounded’. Of this, I really do not know what to make. The best evidence for Herder’s discipleship is in his letters to Hamann. If Professor Aarsleff had read these letters with the attention that he has devoted to Leibniz or Wilkins, he would have found that Herder, a proud and prickly man who stood aside from most of his contemporaries, venerated among them Hamann and him alone. Herder, unlike Hamann, was not an enemy but a critic of the French Enlightenment, and freely admitted his great debt to it. He took an interest in, and used the findings of, the natural sciences of his day – this aspect of his work is enlarged upon, and, if anything, overstressed, in my essay. But to deny that he owed Hamann the central ideas, the development of which in his own unique fashion created his name and influence, is eccentric. His conviction that all human activity, and especially art, religion and custom, are rooted in the life of particular societies, not in individuals or mankind at large, and that they function primarily as means of communication; his populism, with its occasional nationalistic tinge; and, finally, his pluralism of changing values, which went with his detestation of French universalism, dogmatic Neo-Classicism, materialistic utilitarianism – all this is derived principally (for better or for worse) from the Magus of the North. His letters leave no doubt that this is what he himself regarded as his greatest debt. He spoke of Hamann as an oriental sage whose disciples were happy to receive morsels that fell from his table. Why Professor Aarsleff chooses to deny this, or at least to doubt it, puzzles me almost more than anything else in his peculiar article. As for Condillac, he may have been the inspirer of the socio-linguistic surveys made by French savants in 1790 and 1800, but it seems to me that the growing knowledge in France of Celtic, Norse, Persian and Indian literatures and customs, and, by 1800, the retour d’Egypte, are sufficient to account for the conception of languages as a means of understanding cultures, without benefit of Vico or Herder.

If Professor Aarsleff can believe that Michelet, who, towards the end of his life, wrote, ‘I had no master but Vico; his principles of the living force of humanity creating itself made both my book and my teaching,’ would have caught fire from the obiter dicta or the linguistic theories of Leibniz, Locke, Condillac or their followers, he is capable of believing anything. One last-minute concession I am ready to make him. He complains that I call Vico the founder of the German historical school. I should have called him its unjustly ignored anticipator. Most of its members had most probably never heard of him. When Vico’s theories about Homer came to the notice of the Homeric scholar F.A. Wolf, and Vico’s writings on early Rome to that of the great Roman historian Niebuhr, these eminent men were not pleased. Pereant qui ante nos nostra dixerunt. At least Vico, whatever his other anxieties, had no cause, pace Professor Aarsleff, to feel this particular form of resentment.

6. Finally, let me say this: I could understand it if Professor Aarsleff had argued that I have drawn the line between the Enlightenment and its critics – the creators of the Geisteswissenschaften – too sharply: I should have disagreed, but might have been impelled to reexamine my formulation of the dichotomy. But this attempt to play down Vico’s importance goes very much further. It says a good deal for the vitality of Vico’s ideas that, after two and a half centuries, a particular interpretation of them can still stir so much passionate feeling, if only of acute exasperation, in a serious scholar’s breast. Nothing in Professor Aarsleff’s article causes me to modify my views: and I suspect that nothing I have said will have any effect on him.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN


Please include name, address, and a telephone number.


Vol. 4 No. 2 · 4 February 1982

SIR: In the rather curious exchange between Professor Aarsleff and Sir Isaiah Berlin (LRB, 5 November 1981), in the course of a good deal of pedanticism on both sides, it seems to me that essential issues got lost sight of. The question to be asked is: what purpose lay behind the study of language in the 17th, 18th and early 19th centuries? Neither Locke nor his contemporaries studied language as an object of pure intellectualising. Locke had a most important practical purpose – to reform the evils of the ‘understanding’ and its misuse, which, he believed, had terribly afflicted the world during its whole history. He thought that words should reflect clear and distinct ‘ideas’, the perceptions of actual, existing things. The relations between these simple ideas constitute our notions of values, of laws, and everything else. When Professor Aarsleff says that, in Locke’s opinion, ‘if languages were made by illiterate men, they bear no testimony to the truth of things, but only to the minds of those who made the words,’ Locke would say that, unfortunately, this is correct: but – and here is where Locke momentously leaves the company of such modern writers as Saussure – he emphasises that this state of affairs need and should not be.

The mind, Locke stressed, has an innate power of seeing the true relationships between the simple ideas, which are universal, and universally true. ‘These simple ideas, when offered to the mind, the understanding can no more refuse to have, nor alter, when they are imprinted, nor blot them out, and make new ones itself, than a mirror can refuse, alter or obliterate the images or ideas which the objects set before it do therein produce.’ Professor Aarsleff, it seems to me, is clearly inaccurate in attributing to Locke the notion that ‘since ideas [simple ideas] are private, there is an irremovable subjective element in language which precludes that one speaker can ever truly know what another meant to say.’

Professor Aarsleff is also wrong in saying that ‘languages, being made by men, are pre-logical and bear no relation to the true constitution of things.’ Insofar as this statement is applied to the age of Locke, it is clearly wrong. Languages most certainly do have a relationship, even though imperfect, to the order of reality which exists, immutable and clearly perceptible, outside us. In his Manifesto, ‘The Conduct of the Understanding’, Locke set forward, with great enthusiasm and urgency, a programme for vast and sweeping reform, which would produce a correct way of thinking. If he had had any such notion as the one given above, he would have given up in despair before starting to write. What we need to do, he says, is to ‘reflect’ – to observe, impartially and calmly, the relations between the simple ideas in our minds. Then we are guaranteed success. But, because of passion, ignorance, absorption in our own thinking, associations of ideas, etc, we all too seldom do regard the clear and obvious connections between the ideas printed on our mental retinas.

As to Vico, it seems to me he expresses another view of the human mentality. In short, he expresses something like the romantic idea of the Volk – the notion that each people has developed, from a deep, mysterious source, peculiar to itself, a unique sense of the universe, expressed in folk art, the early epic etc. He is partially anticipated – in his attitude towards the pre-literate ‘age of poets’ – by Blackwell and others: but I am inclined to agree with Sir Isaiah that Vico is more an ‘anticipator’ than an influencer. He expresses, early, attitudes that were to become commonplace in the age of high Romanticism.

Ernest Tuveson
University of California, Berkeley

Vol. 4 No. 10 · 3 June 1982

SIR: Sir Isaiah’s response to my article (LRB, 5 November 1981) casts fresh light on the background and nature of his opinions. The issue is not whether Vico is original, but that Berlin’s knowledge and arguments do not show in what this originality consists. The issue is Berlin. Criticism of his work is taken to minimise Vico’s originality, a position that for Berlin can only be rooted in ‘lack of understanding’, ‘national pride’, ‘ideological antipathy’, or ‘jealous concern for the reputation of some other thinker’. Since Berlin believes that I am an erudite and scrupulous scholar, he surmises that I am irritated by Vico’s intrusion into the Enlightenment and by ‘his unscientific etymological theories and the value put upon them by his admirers’. I have now for twenty years repeatedly argued the reverse in print; my aim has always been to understand past language study as an aspect of intellectual history without the least prior regard for current opinion of its soundness. It is sad to find Berlin so eager for an ad hominem argument that even an uninformed one will do.

But since this puts me in the company of Goethe, I do not feel bad about it. Of Goethe’s account of Vico, Berlin wrote that ‘as anyone can tell who has the smallest acquaintance with the doctrines of the New Science, Goethe’s remarks bear little relation to the text: he evidently did not bother to read him.’ On this basis I said that Berlin took Goethe ‘severely to task’, but now Berlin answers that ‘this is not so. I did suggest that Goethe wrote about him but never troubled to read him.’ Are we asked to believe that Berlin’s original words merely made a suggestion? I mention this only to ask, if Berlin has such difficulty reading himself himself, how in fact he reads Vico, and how we in turn are to read Berlin on Vico with any hope of understanding. My present response is addressed to those who still hope that Vico and Herder studies can become fruitful and informed.

In what Berlin calls my ‘peculiar article’, I said the belief that Hamann was Herder’s teacher would clearly seem to be unfounded. That I should choose ‘to deny this, or at least doubt it, puzzles [Berlin] almost more than anything else’, a problem that would easily have been resolved, he suggests, if I had read Herder’s letters to Hamann ‘with the same attention that [I have] devoted to Leibniz and Wilkins’. I have indeed read those letters, but I do not take the expression of veneration and friendship to mean discipleship. I have also read Herder’s letters to other correspondents, and they offer even less reason to accept what Berlin finds it ‘eccentric’ not to see. Early in 1769, Herder wrote to Hamann: ‘Soon we shall no longer understand each other.’ Then followed three years of silence, during which time Herder wrote the famous essay on the origin of language. Of this essay Hamann wrote a review that Herder called ‘a spiteful lampoon.’ Herder eagerly read Diderot, considered him the greatest French philosopher, and at one time joined him with Plato and Shaftesbury in the triumvirate of the deepest philosophers he knew. In his study Herder kept Lessing’s, Luther’s and Hamann’s portraits along with those of the ducal grandees he served, but in his letters to Hamann he avoided all mention of his high regard for Spinoza. Years before Hamann died, Herder paid tribute to his knowledge and wisdom, but added that ‘it is unfortunately not for our time.’

Berlin’s belief in Herder’s discipleship might have been interesting if it were new, but it is old and familiar; some three generations ago it could even have been considered obvious. Berlin calls Robert Clark’s work (1955) on Herder ‘a magnificent biography’. Has it escaped Berlin that Clark’s main thesis, forcefully stated in the Introduction and argued throughout the work, is a rejection of Berlin’s old thesis ‘as degrading to Herder … and as ignoring the real set of problems that engaged Herder’s active and individual mind’? Yet Berlin supposes that a mere reading of letters he thinks I haven’t read will be enough to set me free from my eccentricity. More than twenty years ago the late Wilhelm Dobbek (who knew Herder’s works, correspondence and unpublished manuscripts better than anyone) wrote that after 1770 the two men from Königsberg no longer understood each other and that Hamann in the end for Herder was no more than a fond link to his early years and native city. Berlin’s opinion has for much more than a generation belonged to the realm of academic folklore, yet he does not hesitate to make it the basis for declaring a suspicion about lacunae in my reading.

Vico’s and Herder’s conceptions of the nature and origin of language are the foundations of their philosophies. These doctrines therefore have a crucial place in Berlin’s argument, and his claims for originality rest on the knowledge he adduces to support them. In his response Berlin cites the doctrines of Herder, Condillac and Süssmilch, all of whom are among the most important writers on the subject. In a short passage Berlin manages to betray ignorance of the writings and arguments of every one of them. Süssmilch, Berlin says,‘maintained that language was a gift conferred upon Adam by a special act of the creator.’ No, the brilliant demographer did not say that. He argued that since language must be either divine or human, a demonstration of the impossibility of human origin would leave only divine origin. His argument was, as he stated, strictly philosophical and set out to show the insufficiency of all known arguments in favour of human origin. Süssmilch’s essay may well be the most profound contribution to the subject ever written. Berlin’s ignorance is old and goes back to Herder’s misrepresentation of Süssmilch – which Herder scholars have found it convenient to trust right down to Berlin more than two hundred years later, though published correspondence has long shown that what especially disturbed Herder when his essay went into print was his treatment of Süssmilch. Berlin is again relying on 19th-century opinion. In an exemplary article (Germanic Review, 1978), Professor Bruce Kieffer has shown how thoroughly Herder distorted Süssmilch’s argument and how scholars have trusted those distortions.

Berlin is equally ignorant about Condillac, who thought, he says, that language ‘was a natural development from the cries of animals’. No, the cries of animals have a crucial role in Herder’s account, but have no place in Condillac’s. To say that language was in any sense ‘a natural development’ begs the entire question, but since Berlin thinks Herder did not hold that view, he presumably believes that language for Condillac was somehow a materialist, automatic, passive, unthinking product, the result of instinct much like a spider’s web. Having cited Herder’s false account of Condillac as his authority, Berlin concludes that ‘according to Herder, there is no continuity between animals and human beings, but a clean break.’ This is also Condillac’s position: animals do not speak because they have only instinct, while man does because he has reason. For Condillac, the break is as clean as for Herder, and both held that ‘nature made man for language,’ to use the words of Herder, who, like Condillac, placed the difference in man’s organic being. ‘It is not surprising,’ said Condillac, ‘that only man, who is as superior in regard to organic being as by the nature of the spirit that animates him, has the gift of speech.’

Herder’s linguistic thought and his contribution to Romantic aesthetics and historicism are held to find expression in such statements as these: ‘Everything confirms that each language expresses the character of the people who speak it. Of all writers, it is in the poets that the genius of language finds its most vivid expression. For anyone who knows languages well, they become like a portrait of the character and genius of each nation.’ These words are in fact Condillac’s, published in 1746 in a text that Herder certainly knew when he first began to write on language, poetry etc in the 1760s. At that time Herder wrote a short essay which is rightly judged a key to his linguistic thought. In his well-known monograph on the idea of language from Dante to Vico, Apel observes that this essay comes so close to Vico that it is ‘hardly believable’ that there is not a ‘real dependence’ on Vico. But when Herder read Condillac ‘with delight’, it seemed to him that Condillac cast light on his own essay. As long ago as 1945, René Wellek stated emphatically that Vico and Condillac expound ‘a strikingly similar theory of the evolution of language and poetry’. This evident fact has been treated in detail in the secondary literature. In his response Berlin twice asserts that Herder owed Condillac nothing, and he speaks with equal confidence on another point. In ‘Vico and Berlin’ I wrote what I have argued elsewhere, that Condillac ‘more than anyone provided the linguistic and philosophical foundation for the concept of the Volksgeist and the expression of culture in language’. Berlin condescendingly responds that I ‘quaintly trace’ this concept to Condillac. In view of the questions I have raised here, can any reader trust Berlin’s sense of what makes an opinion worthwhile on these subjects?

As with Süssmilch, so with Condillac, Berlin’s failure stems from trust in Herder and in the folklore based on Herder, which to this day is the conventional wisdom on the subject. But Berlin’s confidence extends to the linguistic thought of the 17th and 18th centuries at large.

Berlin sums up Vico’s important principle in these words: ‘We think and can think only in symbols, whether words or images; the two are one.’ This claim, says Berlin, was ‘still novel enough when it was made a hundred years later’ in de Maistre’s ‘thought and language are two magnificent synonyms,’ a conception Berlin also credits to Hamann and Herder. Berlin and his old authorities do not know that this principle was firmly established before 1700 as a consequence of the doctrine (stated, for example, by Bacon and often repeated) that languages were not made either by Adam or by philosophers but ‘were framed and applied according to the conceit and capacities’ of ordinary people. It was advocated by Mersenne and his followers both in France and England, not least in the popular and often-printed conferences of Renaudot (in half a dozen volumes in French which were in large part also issued in English). In the early 1630s, the conferences report that ‘all things made by the institution of man, such as language, are as diverse as are the opinions of men’ (i.e. Weltanschauung), and in the first days of 1638 they discussed the very same question about the origin of language that Herder answered some one hundred and forty years later. This same principle was fundamental doctrine with John Wilkins and the Royal Society until, enforced by Locke, it became central to 18th-century linguistic thought. In 1703 an often quoted French treatise on languages said that ‘speech is the mirror of the soul, man paints for himself in language. It is with entire nations as with an individual. Their language is the living expression of their customs, of their genius, of the entire thinking of their soul and all the passions of their heart.’ Condillac was not saying anything new when he wrote that ‘the art of speaking, the art of writing, the art of reasoning and the art of thinking are fundamentally one and the same.’

Thus etymology became the avenue to the history of thought and philosophical understanding. Vico was not the first to understand the importance of ‘genetic etymology and philology’. Yet Berlin is convinced that Vico was, ‘so far as I know, the first to grasp the seminal and revolutionary truth that linguistic forms are one of the keys to the minds of those who use words, and indeed to the entire mental, social and cultural life of societies.’ Leibniz is only one represenstative 17th-century figure who has grasped this truth much better than Vico – Berlin for some reason prefers to believe that Leibniz’s linguistic interests were all devoted to the Characteristica Universalis. In 1748 Maupertuis (who knew Condillac and his work well) wrote that ‘in the construction of languages we can discover the vestiges of the first steps taken by the human mind,’ a principle he based on the conviction that ‘the signs by which men have designated their first ideas have so much influence on all our knowledge that I believe that researches into the origin of languages, and on the manner in which they have been formed, deserve as much attention and can be as useful in the study of philosophy as other methods that build systems on words with meanings that have not been thoroughly examined.’ During the 18th century, these principles were stated again and again by the very figures Berlin thinks were the opponents of what he calls ‘one of Vico’s most revolutionary discoveries’. For d’Alembert, the philosophy of language is ‘the march of the human spirit in the generation of its ideas’; for Turgot, citing the authority of Locke and Condillac, etymology is empirical metaphysics which studies words like grains of sand the human mind has left in its path as our guide to origins, so that ‘those who study the march of the human mind in the history of thought must never fail to march with the torch of etymology in hand.’ By the time Monboddo wrote, it had long been familiar doctrine ‘that, from the study of language, if it be properly conducted, the history of the human mind is best learned.’ They all agreed with Vico, of whom none of them had heard, that ‘our science is a history of human ideas, on which it seems the metaphysics of the human mind must proceed.’

Language and thought are indeed synonymous, etymology is the torch that illuminates the history of thought: these are basic conceptions of Enlightenment thought tout court. They were fully articulated before Vico, as Berlin says, ‘expounded an anthropological doctrine according to which the evolution of successive Weltanschauungen of primitive … societies is accurately reflected in changes in the forms of language, the study of which is therefore an indispensable instrument in determining the path of this evolution’. They were not as Berlin would have it ‘isolated antecedent elements’. If Berlin has ‘yet to learn this’, it is because he does not know the literature and trusts authorities that say nothing about it. Berlin’s knowledge is not equal to the extent of truth.

In his response Berlin again insists on Vico’s originality ‘in declaring that mathematics, as a man-made set of rules, was not a body of objective knowledge…but a set of arbitrary rules, applicable to, but not derived from, the observation of the world.’ With a nice piece of essentialist history and question-begging, Berlin declares that ‘clearly no 17th-century rationalist could possibily have held’ this view – such history is a great saver of the pains of reading and learning. But it is enough to read Descartes’s familiar writings against the objections to his Meditations to know that this view was indeed held in the 17th century. Descartes saw the threat it posed and argued strenuously against it as ‘the objection of objections’. For these excellent minds, he wrote, ‘all things that we can understand and conceive are only imaginations and fictions of our mind and bear no relation to reality.’ Berlin admits that Nicholas of Cusa held this view, but he apparently rules out that anyone could have held it in the 17th century: ‘Two hundred years before our time [Vico] conceived of mathematics as the invention of fictions.’ This view of the nature of mathematics is prominent in Gassendi and Mersenne.

Since the primary texts are well-known and have been amply treated in the distinguished secondary literature, there is no need to pursue the matter further. Berlin saves the day for Vico by passing over a text as important as Lenoble’s great work on Mersenne (now nearly forty years old), to say nothing of the readily available writings of Popkin and Rochot. Here Berlin could have learnt not only about mathematics but also about Mersenne’s empiricism, his views of the nature of language, customs and institutions, and his argument against natural law. He could also have received enlightenment about the sociological nature of Mersenne’s thought, and, finally, been reminded that the natural science that took shape in the 17th century followed Mersenne and not Descartes, helped by the popularity of Renaudot’s conferences and by John Wilkins’s commitment to Mersenne’s principles. Descartes’s physics was a disaster, but Mersenne’s mechanistic philosophy was not. Berlin’s singular fixation on Descartes supports his opinion that ‘mathematics was almost universally considered to be a form of factual knowledge,’ so that it could become ‘a momentous step’ for Vico to assert that ‘mathematical propositions are true only because we ourselves have made them.’ Descartes’s opponents were not, as Berlin would have it, merely making the familiar distinction between a priori and a posteriori knowledge. Mathematics was indeed Menschenwerk.

Berlin is spellbound by the idea of a logically perfect language. This is the only historical context he offers for what he sees as Vico’s revolutionary linguistic thought, and Vico’s rejection of the possibility of such a language becomes the test of his originality. But Berlin’s is not the view of well-informed scholars. On the contrary: Tullio de Mauro has cogently argued both that Vico’s belief in the universal validity of a ‘mental dictionary’ contains the same conception as Leibniz’s project for a Characteristica Universalis and that this conception also sustains the historicity of languages with all the lessons genetic etymology has to offer. Albert Heinekamp has independently advanced a similar argument, showing that Leibniz saw no conflict between his historical and philosophical linguistic interests. That such a conflict exists is a myth which has allowed the 19th century to believe that the true historical nature of language was its own discovery. Berlin believes I said that there was no active interest in the philosophical language, but what I said was that this interest was not dominant in Vico’s time or later (as if Vico alone grasped the historical dimension); nor did the French philosophers wish to put such a language in place of the existing natural languages. Berlin now says he ‘did not, of course’, say the latter, but he did in fact say it in a passage where he embroidered on Joseph de Maistre, and never until now had he made clear that such a language was to be used only in scientific analysis – as Condorcet, for instance, repeatedly stressed. But if a logically perfect language was never intended to replace the existing natural ones, to what does the perfect language pose a threat? Are chemical nomenclature, Linnean terminology, symbolic logic, mathematical notation etc the enemies of poetry and of efforts to understand history and human nature? In the 19th century many thought so, including de Maistre (to whom contempt for Locke was the beginning of wisdom in philosophy), but they were partisan and misinformed.

Finally, three brief points. Berlin is fond of citing the authority of Erich Auerbach, who had written extensively on Vico and on Herder before Berlin. But he completely ignores the strong critique Auerbach had advanced against what became Berlin’s enterprise. Auerbach does not think that Vico anticipates German historicism and rejects as ‘foolish’ the efforts of some late 19th-century scholars to claim him as their forerunner. He sees no common ground between Vico’s thought and the romantic conception of the Volksgeist, he sees ‘an unscalable wall’ between the mental worlds of Vico and Herder, and he argues that it is no wonder that Vico was not known in late 18th-century Germany: the similarity is not at all as great as it appears to be ‘when modern terminology is introduced for what the authors originally said’. Given Berlin’s respect for Auerbach, would it not have been natural to acknowledge and discuss these fundamental differences? Auerbach’s criticism was directed against the 19th-century views Berlin cherishes.

In ‘Vico and Berlin’ I referred to the answers to Grégoire’s questionnaire in the early 1790s and to Degérando’s ‘Considerations on the methods to be followed in the observations of primitive nations’. The former shows that local officials all over France held views of the interrelations between language and culture which Berlin sees as the great Counter-Enlightenment creation of his favourites, and that they in fact cited their own French philosophers, including Voltaire and the 1703 passage I have quoted above. I referred to Degérando’s famous program because it reveals how powerfully French thought (and especially Condillac) determined the principles of linguistic anthropology in an exemplary manner that still commands admiration. It is characteristic that Berlin, apparently ignorant of these documents, meets them with a bit of guesswork that can be dismissed as quite beyond the realm of reality. He even surmises that recent knowledge of little-known languages would have sufficed to produce those views – surely a piece of excessive positivism, as if a new body of facts would automatically produce and explain the theory applied to them.

When pressed, Berlin is ever ready with guesses. He surmises that ‘if Professor Aarsleff can believe that Michelet [who admired Vico] would have caught fire from the obiter dicta or the linguistic theories of Leibniz, Locke, Condillac, or their followers, he is capable of believing anything.’ In other words, Berlin invents a hypothetical absurdity and attributes its acceptance to me on a matter I have said nothing about.

A closer look at what can be known shows something interesting. The great project that occupied Michelet in his youth was a work ‘on the genius and history of nations revealed in their language’, as he called it. First mentioned in 1819, it continued to occupy him under slightly different titles that all abundantly reveal the debt to 18th-century thought. Here Michelet had been set afire with enthusiasm for a project of Vichian dimensions. Michelet planned his reading in philosophy with special care. What did he read during these years of gestation? He read Locke, Condillac, Degérando’s important work on signs, Blair’s Rhetoric (a special favourite of the idèologues, who brought out a French translation in 1797), Destutt de Tracy, Dugald Stewart, Voltaire, Montesquieu and Gibbon. In Gibbon, Michelet found a note that fanned his interest: ‘There is room for a very interesting work, which should lay open the connection between the languages and manners of nations.’ Vico, Herder and de Maistre do not appear in this context until December of 1823, when they show up in a reading-list that accompanies a sketch of the great project. The sketch itself shows that Degérando’s work on signs was uppermost in Michelet’s mind.

On the basis of such evidence (which has been readily available for more than twenty years), I can indeed believe that Michelet ‘caught fire’ from what Berlin condescendingly calls ‘the obiter dicta or the linguistic theories of Leibniz, Locke, Condillac, or their followers’. Berlin’s absurdity was not only hypothetical: it was also deeply uninformed. Vico and his train entered Michelet’s world when the project had occupied his mind and directed his reading for more than four years. These years were among the most repressive and reactionary in French history. An ambitious young scholar would not have gotten far if he had proclaimed his allegiance to Condillac and French 18th-century philosophy. But with Vico one was safe.

Hans Aarsleff
Department of English, Princeton University

Isaiah Berlin writes: I see nothing in Professor Aarsleff’s latest reproof that requires me to retreat on any issue raised in it, whether of substance or of detail: but then his ideas and those of others whom he cites, both about the degree and the kinds of influence of earlier on later thinkers, seem to me wholly implausible (his account of the influence of Hamann on Herder, or of Vico on Michelet, seems to me particularly perverse). I will not continue to bandy texts with my opponent, if only out of regard for your own and your readers’ time and patience; indeed, it is polemics of this kind that brought much Medieval erudition first into contempt and then into justified oblivion. Professor Aarsleff says that my knowledge is inadequate to the task I set myself. This may be so, although nothing he has said so far seems to me to bear it out. What it does call to mind is Whitehead’s pertinent observation about scholars ‘who know so much and understand so little’. Professor Aarsleff’s two philippics seem to me to be excellent illustrations of this sad truth.

send letters to

The Editor
London Review of Books
28 Little Russell Street
London, WC1A 2HN


Please include name, address and a telephone number

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences