Analytical Philosophy and Its Forgetfulness of the Continent. Gottfried Gabriel in conversation with Todor Polimenov


Analytical Philosophy and Its Forgetfulness of the Continent
Gottfried Gabriel in conversation with Todor Polimenov*


Gottfried Gabriel is interviewed by Todor Polimenov about the relationship between analytic and continental philosophy.

Table of contents

Polimenov: In one of his essays, Michael Dummett defines analytic philosophy as post-Fregean philosophy (1977: 441). What he means by this is that for Frege, philosophy of language is the foundation of all philosophy. According to Frege (as is apparent especially in his way of approaching philosophy of mathematics), we can only properly analyze thoughts through the analysis of language. Elsewhere, Dummett states that Frege has been the first to rightly emphasize that we cannot have a thought without expressing it in language (1978: 116). Finally, in his book on the origins of analytic philosophy, he speaks of Frege’s legacy to analytic philosophy: the linguistic turn. This legacy, according to Dummett, consists not so much in the “details of Frege’s philosophy” (which have, among others, influenced Russell, Wittgenstein, and Carnap in many respects), but rather in a basic idea which Dummett summarizes as “the extrusion of thoughts from the mind” (1993: 22). Accordingly, already in his first book on Frege, Dummett compared Frege’s place in the history of philosophy to the philosophical revolution that started with Descartes’ epistemological turn towards the subject (1973: 665 ff).

By contrast, you, Professor Gabriel, concur with those interpreters (such as Hans Sluga, Gregory Currie, or Wolfgang Carl, among others) who see Frege’s main philosophical interest to be within in epistemology rather than in philosophy of language. Thus, in a number of essays you have provided detailed evidence for Frege’s Neo-Kantian background. In particular, you have demonstrated fundamental similarities between on the one hand Frege’s main philosophical theses and, on the other hand, those of value-theoretic Neo-Kantianism – which has its origins in Lotze and was further developed by Windelband, Rickert, and Bauch. In a similar vein, you have drawn attention to revealing parallels to traditional logicians such as Herbart, Bolzano, Trendelenburg, Sigwart, and Brentano, for example concerning the reduction of Kant’s forms of judgment. And while, in your book on modern epistemology, you explicitly agree that the transformation of epistemology into philosophy of language started with Frege, you only exemplarily refer to his logical analysis of the different uses of “to be” in this context (Gabriel 1993: 129–136).

So my first question is: What do you think was Frege’s role in the development of what we today call analytic philosophy? Or, put differently: To what extend is Dummett’s definition justified?

Gabriel: I agree that Frege is the father of analytic philosophy, although of course we must not overlook the role other authors such as Russell and Husserl have played. As concerns philosophy of language, Dummett is right insofar as Frege has indeed made substantial contributions to this field. In fact, I have myself published a collection of posthumous writings under the title Schriften zur Logik und Sprachphilosophie (Writings on Logic and Philosophy of Language) (Frege 1971). So there’s no doubt that Frege was a distinguished philosopher of language. However – and here I disagree with Dummett – his primary interest lay elsewhere. Frege himself emphasized that his main concern was logicism, i.e. the demonstration that arithmetic is nothing but a further development of logic, and thus an analytic discipline. Now, the concept of analyticity clearly is an epistemological concept according to the classical view that started with Kant. This means that logicism is an epistemological project; and in this sense, I think that Frege – in line with his own statement – primarily pursued an epistemological aim.

I have absolutely no problem admitting that he was also a great philosopher of language, although we should – as you have implicitly done – keep in mind here the distinction between philosophy of language and linguistic philosophy (where linguistic analysis is the only method of doing philosophy). It’s important that Frege, in contrast to Wittgenstein, was not a linguistic transcendentalist. To him, linguistic analysis was not necessarily the condition of the possibility of understanding thoughts. Thus, in this transcendental sense, it is rather Wittgenstein and the authors of the Oxford School, such as G. Ryle or J.L. Austin, who are linguistic philosophers. Russell (another father of analytic philosophy), by the way, was also explicitly no linguistic transcendentalist and was quite critical of the Oxford philosophers.

I therefore think that it would be inadequate to one-sidedly define analytic philosophy in terms of linguistic analysis. It’s definitely right that Frege respected linguistic analysis as a method and also practiced it himself. However, language was more of a vehicle for him, a medium for the expression of thoughts rather than the condition of the possibility of grasping a thought. The way Frege starts with and applies linguistic analysis is basically similar to Lotze’s approach, who said that we have to start with language and follow language. This does not mean, however, that philosophy is nothing but linguistic analysis. So that is where I disagree with Dummett – not only in the interpretation of Frege, but also systematically.

Polimenov: Now that analytic philosophy has come of age, it is time it understood its own history. During the last few decades, there has already been a trend towards acknowledging the historical origins of analytic philosophy and thus – as you put it – to reconsider its “forgetfulness of the continent”. In this regard, you note that Frege has “already been accorded his proper place in Neo-Kantianism” during this development (Gabriel 2000a: 489; transl. 2003a: 32; further 2003b: 125). Now, your long-time assistant and Jena colleague Wolfgang Kienzler criticizes this view of “Frege as a Neo-Kantian” in his habilitation thesis. He argues (2009: 26 ff.) that a position emphasizing the commonalities between Frege and the philosophical tradition loses sight of the specific and original character of Frege’s work and does not do justice to his radicality and innovation. How do you think about this alternative today: is Frege an epistemologist or a philosopher of language? Is there a substantive difference – is it about a “rectification of history” – or do we just have the relation of two complementary viewpoints (i.e. are the two aspects compatible)?

Gabriel: If you put the question like this, then in a way I have already answered it: I do not see a real conflict here. However, we must distinguish between the way Frege actually proceeds and his underlying main interest. Of course he proceeds via linguistic analysis, but it’s not in fact his goal to do philosophy of language; he is primarily interested in epistemological questions.

But you also asked about Frege’s relation to the philosophy of his time. Well, my essay Frege als Neukantianer (Frege as a Neo-Kantian) (Gabriel 1986; cf. 2002) that you mentioned was in a way an act of science policy, so to speak, because I was not so happy about Dummett talking as though everything had started with Frege. I totally agree with Dummett when it comes to appreciating Frege’s role in philosophy. However, there seems to be a certain arrogance in analytic philosophy that comes through in his assessment. It’s a bit like saying: Forget about everything else that may have been. This is a provocation that I think was part of a temporary attitude within analytic philosophy, which becomes clear when we consider who first suggested this kind of view – who were Dummett’s teachers, for example? One philosopher who has played a key role in this “forgetfulness of the continent” typical of analytic philosophy was Gilbert Ryle, of course. We should not forget that Ryle studied under Husserl and that one of his first publications was a review of Heidegger’s Sein und Zeit. After his return from the continent, he wanted a new beginning in Great Britain (with the journal Mind); in a way, he wanted to do “philosophy without footnotes”, without any reference to history. That was his program. Of course, Ryle was still familiar with what had been done before. But it’s quite common that the next generation will forget the past, and that’s what happened with Dummett. It was this policy – to cut off analytic philosophy from the past – that my essay argued against. I wanted to demonstrate that it’s not as simple as that; that Frege has to be placed in the context of Neo-Kantianism and the philosophy of the 19th century, especially within the tradition of Lotze’s theory of validity. And Herbart, too, needs to be mentioned here, since he pointed Frege the way towards an analysis of the relation between existential and number statements.

By contextualizing Frege, I do not mean to belittle his achievements, of course. We have to distinguish here (and that’s where my opinion differs from Kienzler’s): As far as logic is concerned (and also the specific field of philosophy of mathematics), Frege’s accomplishments are no doubt singular. But when it comes to epistemology and the philosophical framework his considerations are embedded in, Frege does not stand alone. It’s quite clear, for example, that he directly deals with Kant in his Grundlagen der Arithmetik, especially with the question whether arithmetic, in defining numbers as logical objects, can provide a counterexample to Kant’s view that there are no intelligible objects. The whole project of logicism is embedded within questions familiar from traditional continental philosophy and especially from German philosophy. This is the context within which Frege’s thinking moves. We cannot just cut off this tradition.

If we take a closer look at the development of propositional and predicate logic, it becomes clear that Frege took up aspects from traditional logic; but of course, his own achievements are unique. Again, this is not about questioning Frege’s originality; I just want to clear up the misunderstanding that we can forget about everything before Frege. The essay with the provocative title Frege als Neukantianer was directed against this misunderstanding.

Polimenov: You’re currently writing a book on Frege …

Gabriel: That’s right, I’m planning to summarize my previous work that deals with contextualizing Frege – in appreciation of his achievements.


Polimenov: In Great Britain and the United States, Frege’s influence on analytic philosophy was to a high degree second-hand up until around 1950, as we can see in some of Dummett’s accounts. In particular, it was mediated through very few, but influential philosophers such as Russell, Wittgenstein, Carnap, and Church. Although Peter Geach began translating Frege’s philosophical works into English at the same time and held a lecture on Frege in Cambridge, and although Austin translated Frege’s Grundlagen for a course at Oxford, Dummett (1993: 170) remembers in a conversation with Joachim Schulte that to him, Frege (just as Mauthner) was not much more than one of those names in the Tractatus during his studies in Oxford.

What was the situation like in Germany at that time? Remembering his studies under Frege in Jena before World War I, Carnap (1963: 4) mentions in his autobiography that Frege’s work was “practically unknown” in Germany back then. During the inter-war period, Vienna, Graz, Lviv, Warsaw, and Prague – where many fundamental analytic approaches have been developed – were part of the broader cultural sphere of Central Europe where German was the language of publication. Despite this, Łukasiewicz (1935: 112) observes regretfully in the mid-30s that even in Germany, it does not seem to be generally known that Frege was the founder of modern propositional logic (not to speak of a recognition of his philosophical import).

So, what was it like in Germany? Was Frege really forgotten? Has philosophical interest in him only been raised from the outside? And what role did the new edition of his writings play? You have yourself been involved in the editing of Frege’s unpublished writings…

Gabriel: When I think back to the time of my own studies, I remember attending an undergraduate seminar on Frege taught by Friedrich Kambartel during one of my first semesters. I knew immediately that this was the kind of philosophy I wanted to do. At the University of Münster, where I studied, Frege has never been forgotten. Already in the 1930s, Heinrich Scholz began collecting Frege’s unpublished works. This collection was lost due to unfortunate circumstances (a bombing raid during World War II), so that today we only have the copies that Scholz made back then and not all of the writings. But thus, in Münster, Frege has always been present, and Scholz also occasionally held lectures on him outside Münster.

One problem was that, after the partition of the faculties, the logician Scholz, who originally was a philosopher, stayed at the faculty of mathematics in order to found the Institute of Mathematical Logic and Fundamental Research (Institut für mathematische Logik und Grundlagenforschung). In Germany, the faculties of philosophy originally also accommodated the natural sciences and mathematics along with the humanities. At various points in time, the faculties were partitioned, so that two separate faculties were formed, one of natural sciences and mathematics and one of philosophy. Scholz, who had started out as a philosopher of religion, therefore had to choose one of these faculties. Unfortunately, he decided upon the faculty of mathematics, and hence formal logic basically lost its place within philosophy. Something similar probably also happened at other universities. I was lucky to begin my studies of philosophy in Münster, where I could hear logic under the philosophically inclined mathematician Hans Hermes, one of Scholz’s students. That’s also how I came to participate in the edition of Frege, by the way.

Polimenov: But what was it like elsewhere in Germany? Did Günther Patzig’s new edition of Frege’s writings in the 1960s play a key role here?

Gabriel: Yes, of course, but that was a little later. Let’s put it like this: Patzig’s editions definitely were decisive for the reception of Frege. As I said, the interest in Frege has always been present in Münster. But it’s clear that without texts, one cannot hold a seminar. Owing to Scholz’s initiative, the seminary library at the Institute of Mathematical Logic and Fundamental Research in Münster held collections of photocopies of Frege’s essays. So here, it was possible to study Frege. There were also (and still are, as I hope) original copies of Frege’s books. However, since there were no student’s editions of his writings, seminars on Frege were impossible. It has to be said that Patzig’s editions made it possible to study Frege’s writings more widely for the first time. And this happened rather quickly, and at all universities.

Polimenov: But is it not the case that Patzig’s work was to some extent inspired from outside movements, through the growing interest in Frege in the English-speaking world?

Gabriel: I think that the motivation came from within a German tradition. Josef König was Patzig’s teacher, and Josef König was interested in logic. He in turn studied under Georg Misch, who published the edition of Lotze’s Logik with the Meiner Verlag. Already under König, some works on Frege were written, including Marcus Bierich’s dissertation where the connection between Frege and Lotze was described for the first time. In other words, it was not only the influence of analytic philosophy; there was also an interest that had grown within the German tradition itself. The Frege-tradition had not ceased, it had only been pushed into the background due to National Socialism. After the war, interest was revived. This development was complemented by the reception of Carnap (this holds true especially for Patzig). These two things go hand in hand. The displaced analytic philosophers – such as Carnap – were being read in Germany. Carnap’s book Meaning and Necessity (which deals with Frege in detail), in particular, played an important role here. We know that Carnap and Patzig were in direct contact.

I do not mean to deny that the Anglo-Saxon influence played an important role, but claiming that Frege was totally forgotten in Germany would not do justice to the German tradition. Münster is a good example to demonstrate that the Frege-tradition was never really discontinued.

As concerns the reception by his contemporaries, it is often claimed that Frege was largely ignored. That’s also not entirely true. In fact, the Neo-Kantians always viewed him as an ally in “the battle” – if I may say so – against naturalism and psychologism. And that’s not only true of Jena, where Bruno Bauch and his students explicitly referred to Frege. (The connection between Bauch and Frege has been investigated in Sven Schlotter’s Jena dissertation. Schlotter (2004) has also shown that Frege himself absorbed ideas from this circle.) Implicit references can also be found elsewhere, especially with Rickert, who, like Bauch, stood in the tradition of Lotze.


Polimenov: Until recently (2007), you have been devoted to the edition of the last volumes of the Historisches Wörterbuch der Philosophie (Historical Dictionary of Philosophy), a project which started in 1971 and to which you have also contributed many articles. All in all, you have authored a considerable amount of lexicon articles over the years; notably, you have continuously contributed to the Enzyklopädie Philosophie und Wissenschaftstheorie (Encyclopedia of Philosophy and Philosophy of Science) edited by Jürgen Mittelstraß. Now, that encyclopedia is almost considered the pride of Erlangen constructive philosophy.

Which role did the Erlangen school, with Lorenzen’s keywords “logic” and “philosophy of science”, play in the establishment and proliferation of analytic philosophizing in post-war Germany? (After most proponents of early analytic philosophy had left the German-speaking countries for the Anglo-Saxon world during the Nazi regime – in contrast to Heidegger, who stayed on the “continent”.)

Gabriel: I myself come from the Erlangen-Constance school and there’s no doubt that it played a major role in promoting analytic philosophy. As can be seen, for example, in the fact that one of the first post-war dissertations on Frege was written by a member of that school, namely by Christian Thiel. Erlangen and Constance were not the only places relevant in this context, however. Wolfgang Stegmüller in Munich – who did not have any connections to Erlangen – has been even more influential in propagating analytic philosophy, and especially the Carnap-tradition.

Let me highlight a crucial difference: The constructivism of the Erlangen-Constance school emphasized the normative basis of research, and practical philosophy has always played a crucial role here. That’s where it differs from Carnap, who was an emotivist in questions of ethics. Constructivism, by contrast, relies on Kant, in particular on Kant’s second critique, his Kritik der praktischen Vernunft (Critique of Practical Reason). So analytic philosophy was not limited to philosophy of science and logic. In addition, there was a constructivist ethics, while for Carnap, giving rational justifications in ethics was unthinkable. So that is where a certain opposition to Carnap arose: glancing at Kant, constructivism explicitly set itself apart from logical empiricism.

Where the Erlangen-Constance school agreed with logical empiricism in the tradition of the Vienna Circle was in their general understanding of philosophy: that philosophy is about the explication of concepts, in the ideal of precision, as well as an appreciation of the crucial role that logic plays in this context. Stegmüller was the leading proponent of the other line, the one that originated directly from the Vienna Circle. The philosophical method was the same, but not the domain that analytic philosophy was to be applied to. These differences are also reflected in publications. Patzig was basically an intermediary between these two positions. He was in good contact with constructivism, but also with logical empiricism, especially with Carnap. So, the places most relevant for the development of analytic philosophy in Germany were Munich (Stegmüller), Erlangen (Lorenzen), Constance (Kambartel, Mittelstraß), and Göttingen (Patzig) – apart from Münster, where the logicians were not philosophers, but mathematicians.

Polimenov: What about Tugendhat in Heidelberg a little later: Was he a lone fighter?

Gabriel: Tugendhat was a bit of a special case, insofar as he came from Heidegger. But he was also in Münster, in the circle around Joachim Ritter, from which Kambartel came, too. In a sense, Tugendhat could indeed be called a “lone fighter”, because he was not concerned with analytic philosophy to begin with. He came to Germany to study with Heidegger, and then found that there were also other things worth considering. But Tugendhat definitely has to be mentioned, as well as the whole circle that gathered around him in Heidelberg and later studied under him in Berlin. You’re right, this circle absolutely has to be mentioned.

Polimenov: Don’t you think, then, that the Erlangen-Constance program of a methodological and dialogical philosophizing can be regarded as the German counterpart to the post-war development of analytic philosophy within the English-speaking countries?

Gabriel: It would be one-sided to speak of a “counterpart” here, because that would mean excluding the other philosophers I have mentioned. And also, the differences have become less pronounced in the meantime.

By the way, the constructivist idea that science and also language have a normative basis has been revived in recent analytic philosophy. When I read Robert Brandom, I recognize almost word by word thoughts that I heard from Kambartel as a student. (Kambartel and Brandom also are in lively exchange now.) So it can be said that in some respects, constructivism was actually ahead of analytic philosophy. To me, Brandom’s current philosophy seems – whether he is aware of it or not – like a (considerably) extended remake of what I, as a student, heard and discussed in the circles of Erlangen-Constance constructivism, even though I later took another direction.


Polimenov: Let’s return to Frege. His view on the relation between logic and philosophy, which has become a commonplace in analytic philosophy today (especially with ideal language philosophers), is modestly expressed in a hypothetical statement in his Begriffsschrift (1879: VI f.):

If it is a task of philosophy to break the power of words over the human mind, by uncovering illusions that through the use of language often almost unavoidably arise concerning the relations of concepts, by freeing thought from the taint of ordinary linguistic means of expression, then my Begriffsschrift, further developed for these purposes, can become a useful tool for philosophers. (Transl. Beaney 1997: 50 f.)

As is well-known, to break the power of words over the human mind by means of logical analysis has become a central concern of analytic philosophy. Philosophical analysis of language is based mainly on the logical theory which Frege developed as truth-functional and quantificational logic. It is therefore generally accepted that the philosophical impact of logic as a central organon crucially depends on its modern form as introduced by Frege.

Now, during the past few years there have been attempts in Germany to question Frege’s status as the renovator of modern logic. I’m talking about Michael Wolff’s much-discussed book on Kant and his book on logic, where he builds on traditional distinctions to argue for roughly the following thesis: The true logic for the employment of the understanding in Kant’s sense is syllogistic logic; Frege’s mathematical logic, by contrast, is a specific logic suitable only for mathematics. Thus, Frege has not managed to embed syllogistic logic within a more general logical framework; rather, he has obscured its status as a general logic.

What do you think about this? Do we have to take such a thesis seriously? After all, you have yourself written a lot on the development of modern logic by Frege…

Gabriel: Before I answer your question, let me briefly get back to an earlier point: The passage that you have quoted, that famous passage from the Begriffsschrift, actually demonstrates very nicely that Frege was not really a philosopher of language. When he says: “If it is a task of philosophy to break the power of words over the human mind…,” he recognizes that the mind can play a role independently of the power of words. So here it becomes clear that Frege was not actually a linguistic transcendentalist; that his real aim was to free himself from the power of words.

Polimenov: But what are we to make, then, of the also quite prominent passage from Wittgenstein’s Philosophical Investigations – which sounds similar to Frege’s remark – where Wittgenstein describes philosophy as “a struggle against the bewitchment of our understanding by the resources of our language” (PI 2009: § 109)?

Gabriel: It has to be kept in mind that Wittgenstein (whether we agree with him or not) thought that everyday language is basically OK as it is. It is rather the language of philosophers that bewitches our understanding: Philosophers are not bewitched by everyday language; they bewitch themselves. Frege, on the other hand, does not trust everyday language. For Wittgenstein (and this holds true both of the Tractatus and the Philosophical Investigations), “all philosophy is critique of language”. This well-known sentence from the Tractatus (TLP 4.0031) can be related to the passage from the Begriffsschrift that you have quoted, insofar as for Frege philosophy is also critique of language (though not only that). But there’s another sentence (also from the Tractatus) – “A thought is a proposition with a sense. (Der Gedanke ist der sinnvolle Satz.)” (TLP 4) – that Frege would never have written. In fact, it’s Wittgenstein, not Frege, who holds the views that Dummett ascribes to Frege, namely that thoughts have no status independently of language.

So Wittgenstein views all philosophy as critique of language. He is a linguistic transcendentalist. In the Tractatus, he is concerned with the logical deep structure of language, which differs from the grammar of everyday language. In his Philosophical Investigations, he insists that everyday language is OK, and so we must not question it. In that sense, the bewitchment is a bewitchment that philosophers have themselves created through their metaphysics. It’s not a bewitchment that lies within language itself. And therefore, we cannot just equate the quotation from Wittgenstein with that from Frege.

But you asked about Wolff. I don’t agree with him at all. Actually, I have quite some trouble even understanding his point. And there’s something I’m a little upset about: In his first book (in the appendix on Frege), Wolff writes that up to now, nobody has recognized that Frege, in his Begriffsschrift, § 4, deals with the theory of judgment in traditional logic. I am surprised about this statement, since in my introduction of Lotze’s Logik, I have elaborated on Frege’s transformation of the traditional theory of judgment in detail (Gabriel 1989). And Wolff knows this text, for in another passage, he criticizes some minor point. He nowhere addresses the actual analysis, though.

I think that Frege’s logic as developed in his Begriffsschrift is, on the one hand, closely linked to traditional logic. On the other hand, it is also an attempt to improve traditional logic and to overcome its limitations. So it’s quite absurd to dismiss this type of logic as a purely mathematical logic – as logic only for mathematicians. Rather, Frege’s logic is an extension of traditional logic. Wolff completely overlooks the fact that traditional logicians were actually already aware of the weak points that Frege criticizes in the traditional (especially Kantian) doctrine of forms of judgment. That’s why we need to take into account the development of traditional logic between Kant and Frege and to study authors such as Herbart, Bolzano, Lotze, Windelband, and Sigwart. The attempt to confine Frege’s logic to mathematics is just as absurd as the contrary view that traditional logic is no longer worth dealing with. Frege has improved traditional logic by providing more different ways of expression. Just consider the possibilities of combining existential and universal quantifiers. To build up a front between traditional and modern logic would only encourage authors such as Dummett in their forgetfulness of the continent.


Polimenov: You once put it like this: “Although, for a long time, people had the impression that Frege was from Oxford, Wittgenstein from Cambridge, and Carnap from Los Angeles”, eventually they realized that these “pioneers of analytic philosophy actually came from the continent (Frege from Jena, Wittgenstein from Vienna, and Carnap from Jena and Vienna)” (Gabriel 2000b: 186). You have yourself substantially contributed to this rising awareness of the common origins of analytic and continental philosophy. Your interest here seems to go beyond the purely historical perspective of overcoming the “forgetfulness of the continent”. In a systematic interest, you want to contribute to surmounting the mutual reservations of the two parties through a reflection of the common roots that continental and analytic philosophy share (Gabriel 2000a: 495; transl. 2003a: 38 f.). In your review of Michael Friedman’s book Α Parting of the Ways: Carnap, Cassirer, and Heidegger, you express the hope for a re-unification of the two traditions, with a view to the possibility of a fruitful combination of continental wit and analytic acumen. Can you explain what you mean by that?

Gabriel: The concepts of wit and acumen make up one of my favorite distinctions, which was also the topic of my inaugural lecture at Jena (on aesthetic wit and logical acumen). This juxtaposition goes back to the enlightenment philosopher A.G. Baumgarten. Baumgarten, the founder of aesthetics as a philosophical discipline, speaks of logic and aesthetics as “sisters” to express that the two belong together. Acumen is the faculty of logic, wit that of aesthetics. Ultimately, this is about defining the relation between logical and analogical thinking, i.e. between logical differentiating on the one hand and analogies such as metaphors and comparisons on the other.

Michael Friedman defines the relation between continental and analytic philosophy in terms of the contrast between the poetic use of language – referring to Heidegger – and logical argumentative discourse – here he means Carnap. In this view, logical argumentative discourse is the form of presentation of analytic philosophy while continental philosophy applies the poetic use of language.

Even though, as I said, the common terms “analytic” and “continental” are not without problems, Friedman’s characterization makes an important point. We need to recognize that there are different forms of philosophy and therefore, poetic forms should also be taken seriously (even if they are not scientific). We cannot equate the academic form of philosophy with philosophy in general. Being scientific is not a necessary condition for cognition. Arts and literature also have a cognitive value. This is not based on the discursive realization of claims to validity, but rather on successful aesthetic presentation of the conditio humana. The presentation of situations, attitudes, or ways of life is also relevant in many branches of philosophy. So why not make use of the corresponding forms of presentation? That’s why I think that what we need is a reconciliation of logical and poetic discourse. Both forms of discourse are fully justified, depending on what the concrete aim of presentation is. They do not necessarily oppose one another.

My greater “tolerance” in this respect may partly result from the fact that I have been concerned not only with logic, but also with aesthetics. In particular, I have argued for the cognitive value of literature. Once that is established, it is easier to accept that there are intermediate forms of philosophy which are more closely linked to literature; and that Begriffsdichtung (as it is sometimes called) can provide new insights. It is noteworthy in this context that the author of classics of analytic philosophy, Wittgenstein, made use of such intermediate forms in his texts. Wittgenstein’s writings form a unique connection between philosophy and literature.

Polimenov: But is there a sharp boundary between those different forms of cognition in philosophy? In the clarification of logically simple concepts, for example – which cannot be defined by breaking them down into simpler concepts – we need to rely on metaphors, as Frege often emphasized. This relationship between “metaphorical hints” and “categorial elucidations” is also a topic you have repeatedly dealt with…

Gabriel: Here you mention an additional argument. Frege’s metaphor of “saturated” and “unsaturated”, which he adapted from chemistry to elucidate the categorial distinction between objects and functions, is a good example. Logically simple concepts cannot be defined; they can only be elucidated. Frege explicitly states that in this case, we can and must apply comparisons and metaphors. This means that even in logic – when we are concerned with basic concepts – we cannot do without poetic use of language in a broad sense. You do not need to be a poet, but metaphors are rhetoric or poetic figures. It’s true that we cannot draw a “sharp boundary” here, as you suggested in your question. Incidentally, fundamental evidence for the cognitive function of metaphors has been provided by the analytic philosopher Nelson Goodman.


Polimenov: You hold that a reconciliation of analytic and continental philosophy will have to begin with a metaphilosophical discussion about the relation between science, philosophy, and poetry; that is, a discussion about the respective forms of language used in these different disciplines. You are convinced that their different forms of representation correspond to different forms of cognition.

Since rhetoric is traditionally the discipline concerned with linguistic forms of representation, you speak of a “return of rhetoric” or of a broadening of the “logic of language forms” towards a more comprehensive “rhetoric of language forms”. In this process it will become clear, as you say, that “the distinction between analytic and continental philosophy is historically explicable, but systematically mistaken” (Gabriel 2000a: 495; transl. 2003a: 39). What we have here are just different ways of viewing philosophy, depending on whether “poetic metaphor or logical analysis takes the lead in philosophical discourse.” This means that philosophy based on poetry and philosophy based on logic constitute complementary forms of cognition (2003b: 128). Could you explain that in some more detail?

Gabriel: As already mentioned, I want to emphasize that there is no absolute opposition between poetic language and logical-argumentative language. Rather, these different forms of language supplement one another, they are complementary. This is what the example of “saturated” and “unsaturated” was meant to demonstrate; it shows that even in logic, we have to rely on poetic language to some extent. Of course there’s no doubt that Heidegger’s later writings speak a completely different language than Frege. But there are also intermediate forms. The continuum of forms of philosophy can be compared to the color spectrum. Just as we can distinguish red from green, we can discriminate between the language of Heidegger and that of Frege. You can tell the difference. But just as there are smooth transitions in the color spectrum, so that you can go from red to green, there are also intermediate forms in philosophy. The thesis that the different forms of philosophy are complementary concerns not only the extremes, but also these in-between cases. Ultimately, the issue at stake is a reconciliation of analytic and continental philosophy by means of analyzing the transitions among the different forms of philosophy. And one important part of this is the epistemological rehabilitation of rhetoric.

For years (or for decades or centuries – that varies), rhetoric was part of the doctrine of ornatus. What was not recognized is that rhetoric is also about forms of cognition (even if not about strict proofs, which are the domain of logic). Just think of Aristotle. So in this sense, I am also interested in a cognitivist rehabilitation of rhetoric. Complete separation of logic and poetics is thus avoided, because rhetoric lies in between (which shows that there are indeed transitions). On the other hand, and just as important, we also have to counter the deconstructionist monopolization of rhetoric. I agree with deconstruction that some parts of philosophy are also poetry. But in contrast to deconstruction, I hold on to the claim to cognition in philosophy, including poetic philosophizing. That means my position is, firstly, distinct from that of hardliners within analytic philosophy who say: “Logic and only logic!” But secondly, it also differs from that of deconstructionists, who say: “There are metaphors in philosophy, too. Since metaphors do not convey cognition, neither does philosophy. There’s nothing but rhetoric.” I debate both of these positions when I argue for a concept of cognition not limited to propositional truth. Along with my research on Frege, elaborating the idea of complementary forms of cognition – including non-propositional forms (in between logic and poetry (Gabriel 1991)) – has been central to my work for a long time.

Translated by Franziska Tropschug


  1. Beaney, M. ed., 1997. The Frege Reader. Oxford: Blackwell Publishers.
  2. Carnap. R., 1963. “Intellectual Autobiography”. In: P.A. Schilpp, ed. 1963. The Philosophy of Rudolf Carnap. La Salle, Illinois: Open Court, pp. 3–84.
  3. Dummett, M., 1973. Frege: Philosophy of Language. 2nd ed. 1981. Cambridge, MA: Harvard University Press.
  4. Dummett, M., 1977. “Can Analytical Philosophy be Systematic, and Ought it to Be?” In: M. Dummett, 1978. Truth and Other Enigmas. London: Gerald Duckworth & Co., pp. 437–458.
  5. Dummett, M., 1978. “Frege’s Distinction between Sense and Reference”. In: M. Dummett, 1978. Truth and Other Enigmas. London: Gerald Duckworth & Co., pp. 116–144.
  6. Dummett, M., 1993. Origins of Analytical Philosophy. London: Gerald Duckworth & Co.
  7. Frege, G., 1879. Begriffsschrift, eine der arithmetischen nachgebildete Formelsprache des reinen Denkens. Halle a.S.: Louis Nebert Verlag.
  8. Frege, G., 1971. Schriften zur Logik und Sprachphilosophie. Aus dem Nachlaß. Ed. by G. Gabriel. Hamburg: Felix Meiner Verlag.
  9. Gabriel, G., 1986. “Frege als Neukantianer”. Kant-Studien 77, pp. 84–101.
  10. Gabriel, G., 1989. “Lotze und die Entstehung der modernen Logik bei Frege” [editor’s introduction]. In: R. H. Lotze, 1989. Logik. Erstes Buch. Vom Denken. Ed. by G. Gabriel. Hamburg: Felix Meiner Verlag, pp. XI–XLIII.
  11. Gabriel, G., 1991. Zwischen Logik und Literatur. Erkenntnisformen von Dichtung, Philosophie und Wissenschaft. Stuttgart: Verlag J.B. Metzler.
  12. Gabriel, G., 1993. Grundprobleme der Erkenntnistheorie. Von Descartes zu Wittgenstein. Paderborn: Ferdinand Schöningh Verlag.
  13. Gabriel, G., 2000a. “Carnap und Heidegger. Zum Verhältnis von analytischer und kontinentaler Philosophie”. Deutsche Zeitschrift für Philosophie 48, pp. 487–497.
  14. Gabriel, G., 2000b. “Kontinentales Erbe und analytische Methode. Nelson Goodman und die Tradition”. Erkenntnis 52, pp. 185–198.
  15. Gabriel, G., 2002. “Frege, Lotze, and the Continental Roots of Early Analytic Philosophy”. In: E.H. Reck, ed. 2002. From Frege to Wittgenstein. Perspectives on Early Analytic Philosophy. Oxford: Oxford University Press, pp. 39–51.
  16. Gabriel, G., 2003a. “Carnap’s ‘Elimination of Metaphysics through Logical Analysis of Language’: A Retrospective Consideration of the Relationship between Continental and Analytic Philosophy”. In: P. Parrini, W.C. Salmon and M.H. Salmon, eds. 2003. Logical Empiricism. Historical & Contemporary Perspectives. Pittsburgh: University of Pittsburgh Press, pp. 30–42.
  17. Gabriel, G., 2003b. Book Review: Michael Friedman. A Parting of the Ways: Carnap, Cassirer, and Heidegger. Erkenntnis 59, pp. 125–128.
  18. Kienzler, W., 2009. Begriff und Gegenstand. Eine historische und systematische Studie zur Entwicklung von Gottlob Freges Denken. Frankfurt am Main: Vittorio Klostermann.
  19. Łukasiewicz, J., 1935. “Zur Geschichte der Aussagenlogik”. Erkenntnis 5, pp. 111–131.
  20. Schlotter, S., 2004. Die Totalität der Kultur. Philosophisches Denken und politisches Handeln bei Bruno Bauch. Würzburg: Verlag Königshausen & Neumann.

Biographical note

Gottfried Gabriel is Professor Emeritus of Philosophy from the University of Jena. He defended his PhD in 1972 at the University of Constance, completed his Habilitation in 1976. He has been Lecturer of Philosophy at the University of Constance, interim visiting professor at the University of Campinas (UNICAMP), Professor of Philosophy at the University of Bochum, and from 1995 until 2009 Professor of Philosophy at the University of Jena. He works in the areas of epistemology, logic, aesthetics and philosophy of language. His publications include: Definitionen und Interessen (1972), Fiktion und Wahrheit (1975), Zwischen Logik und Literatur (1991),Grundprobleme der Erkenntnistheorie. Von Descartes zu Wittgenstein (1993, 2008), Logik und Rhetorik der Erkenntnis (1997), Ästhetik und Rhetorik des Geldes (2002), Einführung in die Logik (2005, 2007). Editor of Historisches Wörterbuch der Philosophie, vol. 11-13 (2001-2007).

Todor Polimenov is Assistant Professor in logic at the Department of Logic, Ethics and Aesthetics at Sofia University “St. Kliment Ohridski”, Bulgaria, where he defended his doctoral dissertation on Frege and the rise of modern logic in 2006. His research interests include the history and philosophy of logic, and the philosophy of language in the analytical tradition. He is the author of numerous articles, mostly in Bulgarian, but also in German, on Frege, Wittgenstein and Paul Lorenzen. He is the editor and translator of collections of writings by Frege and Carnap.

This conversation took place on May 14th, 2009 in Paris at the international conference “Carnap’s Ideal of Explication: Logic, Metalogic, and ‘Wissenschaftslogik’” (Université Paris 1 Panthéon-Sorbonne). It was first published as “Die analytische Philosophie und ihre Kontingentvergessenheit. Gottfried Gabriel im Gespräch mit Todor Polimenov” (in German, with a Bulgarian translation) in Critique & Humanism vol. 31, 1/2010, Special Issue: Analytic Philosophy. Guest editors: A. Beshkova, E. Latinov and T. Polimenov, pp. 8–35.
Gottfried Gabriel and Todor Polimenov . Date: HTML version first published 5.11.2012; last change 18.12.2013. XML TEI markup by WAB (Rune J. Falch, Alois Pichler, Špela Vidmar) 2012-13.
This page is made available under the Creative Commons General Public License "Attribution, Non-Commercial, Share-Alike", version 3.0 (CCPL BY-NC-SA).


  • There are currently no refbacks.