Recent Work on Metaphilosophy


Timothy Williamson. The Philosophy of Philosophy. Malden/Oxford (Blackwell Publishing), 2007, 332 pp.


Herman Cappelen. Philosophy without Intuitions. Oxford (OUP), 2012, 237 pp.

Colin McGinn. Truth by Analysis. Games, Names, and Philosophy. Oxford (OUP), 2012, 189 pp.


The Philosophy of Philosophy is a book on how to do philosophy.

How to define and how to do philosophy is a topic as old as philosophy itself. As philosophy, especially nowadays, comes under attack of various claims of reductionism and charges of being superfluous philosophers devote considerable effort in defending the usefulness and distinctiveness of what they are doing. Philosophy is a many and not a one. There are several understandings of philosophy, and this has been a strength of philosophy as philosophy precedes any introduction of methodology and restriction on procedure and topic. Accordingly a philosophy of philosophy had to be monumental to come to terms with all branches of philosophy and the all the ways to do philosophy (analytic philosophy, various form of phenomenology, postmodernism, various neo-NN form of following an important position/philosopher of the tradition [like Neo-Kantianism, Hegelians …]).

In this respect Williamson’s book is one-sided and strongly focused. Although he mentions some other forms of philosophy in the introduction he only concerns himself with philosophy in the analytic tradition. And he is very outspoken about its superiority: ‘Analytic philosophy at its best uses logical rigor and semantic sophistication to achieve a sharpness of philosophical vision unobtainable by other means.’ (46) – ‘The rise of modern logic from Frege onwards has provided philosophers with conceptual instruments of unprecedented power and precision, enabling them to formulate hypotheses with more clarity and determine their consequences with more reliability than ever before.’(45, cf.279)

Up to some arrogance: ‘Impatience with the long haul of technical reflection is a form of shallowness, often thinly disguised by histrionic advocacy of depth. Serious philosophy is always likely to bore those with short attention-spans:’(45, cf.289)

In fact Williamson argues from a very specific position within current analytic philosophy, and makes use of a couple of claims controversial within analytic philosophy (like theories of direct reference and de re modalities, having a rigid actuality operator, how to understand tacit knowledge and so forth). Williamson often refers to his own work and results (e.g. in the theory of vagueness), and accordingly sometimes (e.g. discussing vagueness or probabilistic reasoning) the discussion seem to be too engaged in setting out one theory on a topic than reflecting on ways to develop philosophical theories. So Philosophy of Philosophy could also be titled One Way to Do Analytic Philosophy. Williamson acknowledges that his book ‘makes no claim to comprehensiveness. … it explores some interrelated issues that strike me as interesting and not well understood’ (8).

Philosophy of Philosophy also makes in some chapters use of the formal devices of analytic philosophy. Although the formalisms are elementary they presuppose some background in modal logic and the theory of counterfactuals (as two formal appendices are provided as well).

The book, therefore, aims neither at the layman or the general philosopher, but is of oft interest to analytic philosophers who reflect on their way of doing analytic philosophy. It is especially worthwhile for those analytic philosophers who rather disagree with Williamson’s sub-branch of analytic philosophy, as the book invites them to compare their methodological self-understanding to Williamson’s and defend it in the light of his criticisms.

Williamson defends armchair reasoning. He claims that one may arrive at substantial insights by reflecting on the conditions and consequence of our judgments (often in present work called ‘our intuitions’, a phrase Williamson does not endorse at all, cf.215). The precondition to arrive at such insights in methodological thoroughness. We have to use our faculties (of modal knowledge, inter alia) and express the conceptual structures we find as clearly as possible.

One chapter therefore deals with knowledge of metaphysical modality, and one with thought experiments as a way to do philosophy. Williamson sees our ability to come to terms with modal knowledge as grounded in our ability to use counterfactuals and counterfactual reasoning: ‘our overall capacity for somewhat reliable thought about counterfactual possibilities’ (137). Therefore the book stresses the importance of formal systems of counterfactual reasoning, and sees the alethic modalities (‘necessity’ and ‘possibility’) as being reducible to our use of counterfactuals: ‘metaphysically modal thinking is logically equivalent to a special case of counterfactual thinking’ (158, cf.162). Thought experiments exploit this faculty as well: ‘Paradigm thought experiments in philosophy are simply valid arguments about counterfactual possibilities’ (207).

Defending arm chair reasoning Williamson does not consider philosophy as a mere reflection of language or meaning. He proposes – as one chapter title has it – to take philosophical questions at face value, i.e. to consider them as concerning their topic (say a theory of properties) directly and not being about our way of taking about that topic. Philosophers use methods of reasoning ‘required over a vast range of non-philosophical inquiry’ (3). They reflect often on language, but turning to language to reason about vagueness, say, does not make vagueness a mere linguistic issue. ‘Many non-philosophical questions that are not about thought or language cannot be resolved without inquiry into thought or language.’ (41), which does not make philosophy, according to Williamson, a discipline reflecting on language.

In this vain several chapters criticize analytic philosophy in as much as it claims to rest on a linguistic or conceptual turn of philosophy. In that tradition philosophy might be seen as conceptual analysis, where conceptual analysis is understood as following the structure of linguistic meaning. Williamson disagrees and claims ‘that the differences in subject matter between philosophy and the other sciences are also less deep than is often supposed’ (3). Scientific results, because of this, may be important to philosophy. ‘Our evidence in philosophy consists of a miscellaneous mass of knowledge, expressed in terms of all kind, some from ordinary language, some from the theoretical vocabulary of various disciplines. … Whatever we know is legitimate evidence’ (277). In this way philosophy is not set apart from any other science. Naturalism tries to exploit that. Nonetheless, according to Williamson, neither reductionism/naturalism nor the philosophy of mind have ‘come to play the organizing role in philosophy that philosophy of language once did’ (18). And even if some sentences which express philosophical results are analytic in the sense that substitution of synonyms turns them into logical truths (i.e. are ‘Frege analytic’): ‘it does not follow that we can gain cognitive access to them simply on the basis of our logical and linguistic competence’ (70). In fact completely competent speakers may deny some such truths (often for reasons of their philosophical background theories); there are no law-like ‘understanding-assent-links’ (85): ‘No given argument or statement is immune from rejection by a linguistically competent speaker’ (97). Philosophy so cannot appeal to the linguistically obvious or evident. And what some practice considers to be obvious or linguistically evident need not be so on close analysis. ‘What strikes us today as the best candidate for analytic or conceptual truth some innovative thinker may call into question tomorrow for intelligible reasons’ (126).

Williamson regards armchair reasoning (‘in which experience plays no strictly evidential role’, 169) as detached from empirical reasoning, because the beliefs acquired in the armchair have no direct link to experience, experience only (formerly once) played a role in acquiring some judgments, now the object of reflection. This theory of experience less than ‘evidential’ but more than ‘purely enabling’ (168, where ‘enabling’ means experience playing a role somehow like in acquiring a concept) might need further elaboration.

Williamson attacks several versions of understanding conceptual analysis (e.g. one which sees conceptual claims grounded in meaning, and thus being a priori and secure and another one which defines those claims as analytic the very understanding of which secures assent by competent speakers). He uses strong language again: ‘the conceptual turn and a fortiori the linguistic turn look like wrong turnings’ (21, cf. 53). These chapters are the most important for current analytic philosophers, even if they do not agree with Williamson’s diagnosis. One may ask oneself (as analytic philosopher) at which point one parts ways with Williamson and which of his presuppositions one therefore has to question. The common thread of these chapters is the question whether conceptual analysis delivers substantial insights or ‘just’ analytic statements (considered as mere trivia). ‘Many philosophically relevant truths are clearly not conceptual truths in any useful sense’ (49).

Williamson rightly stresses that serious analytic philosophy cannot forsake the claim that analysis provides substantial insight: ‘we should not assume that analytic truths are insubstantial in any further sense’ (52, cf.71), i.e. being obvious or trivial. Once this is granted one may even soften Williamson’s attack on conceptual analysis.

One could somewhat more clearly than Williamson differentiate between ‘analytic’ as a description of the procedure or a classification of the result of philosophy. Even if analytic sentences rest on definitional or conceptual structures definitions may be useful or not. Useful definitions capture facts corresponding to these definitions. A truth resting on such definitions, maybe expressed as analytic truth, may thus nevertheless be a substantial truth. This may be even more so in case of concepts long in use where we discover by reflection the structures they try to capture. Following this line of reasoning one may agree with Williamson: ‘nothing has been done to rule out the hypothesis that [the analytic truth] expresses a profound metaphysical necessity about the nature of the world’ (61, cf.63).

These chapters (about half of the book) are worthwhile reading – and maybe arguing against – for anyone reckoning him- or herself to be part of the analytic tradition. The Philosophy of Philosophy is far from comprehensive as philosophy goes, but it is superb in coming to grips with one’s methodological self-understanding as analytic philosopher.

Cappelen, who is sympathetic to much in Williamson's approach, and McGinn disagree widely not only about the way philosophy is and has been actually practised, but also in their proposals what philosophy should be like. They claim each that they themselves follow and endorse the best of philosophical traditions. Cappelen sees this in a variety of methods which cannot by some principled qualities (like being a priori or resulting in analytic sentences) be set apart from the sciences. McGinn, on the contrary, sees the distinctive character of philosophy in the method of conceptual analysis. They agree in some of their adversaries. Both reject Experimental Philosophy and Ordinary Language Philosophy, which might be surprising given McGinn’s endorsement of conceptual analysis and Cappelen’s endorsement of empirical methods. One reason of these rejections is their – at last – shared thesis that philosophy deals with the world and its (types of) entities and not with psychology (i.e. only the human mind and its apprehension of the world) or language (i.e. only with the way the world is represented in language).
Cappelen’s main concern is to refute the ‘Centrality Thesis’, which claims that contemporary analytic philosophers rely on intuitions as evidence for philosophical theories. The crucial term in this thesis is, of course, “intuition”. Cappelen’s whole argument depends on the manner “intuition” is defined. He separates his refutation of the Centrality Thesis into two parts: (i) the claim that on those occasion when philosophers do engage in “intution”-talk, the reference to ‘intuitions’ can be substituted by a reference to folk theories, prejudices or common knowledge, (ii) the claim that in those supposedly paradigmatic examples of philosophy centring on ‘intuitions’ (mostly when philosophers invoke thought experiments) intuition in the crucial sense of the Centrality Thesis play no role. Both claims are empirical, since they depend on a comprehensive or at least representative survey of ‘contemporary analytic philosophy’ (given a suitable definition of, at least, “belonging to the analytic tradition” and “contemporary”). Such a survey might require a research project of its own. And thus whether Cappelen’s claims are empirically adequate is hard to tell. With respect to the sought for uses of “intuition” we have to remember what the logician says: absence of evidence is not evidence of absence!
Let us suppose that Cappelen’s claim (i) has been established. What does this show? It shows that describing the philosopher’s data or sources does not require casting these as “intuitions”. It does not show that philosophy does not, as Cappelen likes to argue, rely on a special onset of its investigations. An Ordinary Language Philosopher may reply to Cappelen: Ordinary Language Philosophy does not rest its claims on 'intuitions'. “Intuition” has too many different meanings: these range from opinions based on 'common sense' or folk theories (so that claiming something to be 'intuitively' so may hedge the claim made, as Cappelen points out) to the intellectual apprehension of conceptual insight (e.g. in some philosophies of mathematics). One might call reliance on one's unsystematic (i.e. pretheoretical) understanding of language rules 'intuitive', but as the term has been used for quite different forms of belief, it may be better to speak of an expression of language competence or linguistic judgement or meta-linguistic beliefs. In Ordinary Language Philosophy we investigate possible cases under the directive of what we would say if. The evidence which turns up with this rests on linguistic judgements or pre-theoretical knowledge of language. Philosophical 'intuitions' either appeal to shared convictions in some (folk) theory, which carries little argumentative weight, or are guided by a (partial) apprehension of rules of word usage. Intuitions in this latter sense are philosophically useful and necessary as an element of analysis. Positively one may surmise that the new fashionable recourse to 'intuitions' stems from a dissatisfaction with 'naturalism' and mere stipulation of language forms. Pleas to 'intuitions' want to come close to a source of philosophical insight, where this source better be language.
Cappelen’s second claim depends not just on his examples being chosen representatively, but foremost on his definition of “intuition”. Cappelen defines “intuition” by three features: (i) intuitions supposedly come with a phenomenality that makes what one intuits seem true, (ii) intuitive judgments need no justification, they provide justification (have ‘rock status’), (iii) they are based on conceptual competence. Cappeln then goes on to show that the eight paradigm cases of contemporary analytic philosophy supposedly relying on intuitions (ranging from Perry’ shopping cart case to Chalmer’s zombies) do not rely on intuitions in the sense defined by Cappelen. Again let us suppose that he has established this. What does it show? It might very well show that the definition made unrealistic demands on ‘intuitions’. Most crucial in this respect is feature (ii), heavily relied upon by Cappelen in his analysis of the paradigm cases. In my experience the extensive reference made to ‘intuitions’ in the philosophy of language and epistemology (and especially so in philosophical logic, although Cappelen at one point seemingly wants to exclude that area) does not include the non-revisability of intuitions at all. Often special emphasis is put on the progress from our initial to our ‘reflected’ intuitions. So it seems Cappelen is chasing a man of straw here.
Nonetheless Cappelen’s book, apart from the laudable – even so partial – liberalism with respect to philosophy’s methods, is a valuable advance in metaphilosophy: it challenges all those who use ‘intuition’-talk and rely on ‘intuitions’ in their philosophical methodology to come up with clarifications what they mean by that and how they meet Cappelen’s challenge.
McGinn might be understood as taking up that challenge by boldly stating that philosophy investigates the essence of things in the world by conceptual analysis (i.e. relying on our conceptual competence). McGinn’s defense of conceptual analysis on first sight might be seen as reminding philosophers on a viable method, set aside because of too much undeserved bad press (on the hands of Quine and many others). McGinn, however, claims that philosophy is nothing but conceptual analysis. Such methodological imperialism has done philosophy no good in its history, and the opposite claim might be more recommendable: philosophy refers to any viable method used anywhere in the sciences and adds some special methods (like formalization and – contrary to Cappelen, Quine and many others – conceptual analysis). Conceptual analysis cannot be all of philosophy. For a start: that philosophy is conceptual analysis had to be established by conceptual analysis itself. The concept of philosophy had to entail that philosophy is nothing but conceptual analysis. McGinn does not establish this.
McGinn puts forth only a few conceptual analyses in fact. Centre stage he rejects the undefinability of the concept of ‘play’ and sides and endorses Suits’ analysis of it [the repetition of which here would distract us too long]. He defines ‘knowledge’ as ‘non-fluky true belief’. Although this gets the (in)famous Gettier-cases of the backs of the theory of knowledge, one may doubt whether this goes far enough. Why not define “knowledge” as “true belief”? There is nothing inconsistent about both the following sentences: “Peter knew it, but only by accident”, “Peter knew it, but could give no reasons”. In any case one may agree with McGinn that conceptual analysis is viable and useful. How far does he want to carry it? Is there a conceptual analysis of the concepts of ‘stone’ or ‘rose’? Although McGinn mentions on the sidelines Strawson’s idea of ‘connective analysis’, which elucidates the links between concepts, his main idea of analysis seems to be decomposition into necessary and sufficient combinations of constituent concepts (some of which finally have to be atomic). The tradition of Ordinary Language Philosophy includes many more methods (ranging from substitution tests over the controversial paradigm case arguments to contrastive analysis and rephrasing by assertability conditions, inter alia). McGinn, however, foams about the ‘linguistic turn’. He wants it ‘burned’ and to be ‘stamped out’.
McGinn’s rejection of analysis of language is one of the shortcomings of his book. This may rest on a confusion about the role of language in linguistic analysis: language need not be taken by linguistic philosophers as the primary object of philosophy, but analysis of language (usage) is taken as one crucial or even the privileged method of getting at concepts. “Definition” applies to words at least as well as to concepts. Proposing a definition and testing it with cases (similar to proposing a hypothesis and testing it) explores whether the definition covers all cases by testing our judgements to the applicability of a term (the meaning of which contains the concept referring to the property ultimately under investigation). If language was not methodological essential one has (i) to account in some other way for the shared possession of concepts, which on the other hand every theory of concepts has to do, and (ii) to find some other way to identify a concept in question intersubjectively. Methodologically language helps to identify a concept in question as the core of the meaning of an expression employed. Further on, many concepts (especially those for social institutions) depend on language and rule governed communities. Concepts like ‘marriage’ cannot be separated from special speech acts that constitute respective social facts. Many if not most of the concepts interesting philosophers will be of this kind. Epistemological concepts like ‘knowledge’ are at least indirectly tied to language (e.g. by the link from feasible assertions to know something to justifying them towards an audience, of course using language and appealing to shared conventions). Moreover, one may argue that concepts involving powers of reflection and self-representation in thought need language ('inner speech') as representational device. This applies to all forms of shared knowledge ('common knowledge') essential for conventions, and arguably to a full-fledged concept of ‘belief’, as this involves reflecting on one's beliefs and their interrelations, and their relation to the world. So, although conceptual analysis – by definition – aims at concepts, the privileged method to do so is linguistic analysis. As conceptual analysis aims at concepts shared between individual natural languages, no individual natural language is essential for it, and all its cases of analysis have to be transferable in principle from one language to another.
The other main difficulty one should have with McGinn’s book is his claim that philosophy discovers the essence of things (“things” used here generically for all types of entities) by conceptual analysis. Two questions can be raised immediately: (i) what are ‘essences’ and (ii) why can we rely on concepts being properly tied to the essences of things?
The second question can only be answered by a theory of concept acquisition, possession or evolution. We need a theory why our conceptual equipment hooks up properly to properties in reality. Maybe some form of cognitive evolutionary theory can deliver that much. At points it seems that McGinn supports a representational theory of concepts (like concepts being types in the ‘language of thought’). Still evolution may have equipped us in many cases only with conceptual hook-up good enough for reproduction, not fit to capture the essence of things.
What are ‘essences’ anyway? Linguistic analysis answers this question by reducing essential truth to analytic truth, and thus finally to meaning constitutive definitions or conventions. Consider the following three sentences:
(1) “Cats are mammals” is analytic/necessary true.
(2) The concept ‘cat’ contains the concept ‘mammal’.
(3) Cats are essentially mammals.
We can freely move from (1) to (3), and back, by the authority of language: language developed and was adapted in its definitions of words to be successfully applied (interwoven with non-linguistic practices), the meaning of a word containing crucially a concept expressed, this concept referring to a property in reality. The essence captured thus is foremost a linguistic essence, due to definitions, but we are prone to revise or replace our definitions in the light of new important discoveries. It is ‘foremost a linguistic essence’ as the necessity involved is that of the linguistic framework. Many, McGinn included, want to have metaphysical essences (i.e. essences expressed in generalizations stronger than empirical/inductive generalizations, but not dependent on ‘mere linguistic conventions’). Again, there may be feasible theories for such a concept of essences (like two-dimensionalist semantics), but McGinn does not deliver one, his hints link essences to natural kinds, thus hinting at identifying essences with the nature of natural kinds. A non-trivial problem for such a theory will be its compatibility with a theory of concept possession as mentioned in the last paragraph: concepts as representations in our brain hook-up to reality by natural laws, but natural laws can vary while metaphysical identities stay constant, thus conceptual links need not coincide with metaphysical links. McGinn has to work out these parts of his theory.
McGinn’s book, therefore, on the one hand presents a strong case for the viability of conceptual analysis and thus defends a method philosophy should not forego. On the other hand, McGinn’s approach needs a lot more detailed exposition of its methodology and metaphysical background.
Although Cappelen and McGinn will not like it, one may do best by combining Cappelen’s open methodology with a dose of McGinn’s conceptual analysis, and thank them both for their attacks on Experimental Philosophy’s much too lose methodological reflection.

Manuel Bremer, Heinrich-Heine-Universität Düsseldorf, Germany