Contrary to a common opinion, some theses of scientific anarchism are historically connected not only to Popper's and "second" Wittgenstein's thoughts, but also to some ideas affirmed by the advocates of "physicalism" (like Neurath) during the neopositivistie debate on protocol sentences. The common basis of "physicalism" and "anarchism" is a repulse of the "atomistic" theory of meaning. That is making more adequate the epistemological description of knowledge. But both Neurath and Feyerabend err in thinking that this repulse entails a conception of "truth" as "coherence" instead of "correspondence". Against such a conviction the "realistic" requirement expressed by SchHck through the concept of "Konstatiemng" is still very valid.

This paper outlines the main features of the conception of empirical knowledge presented by Moritz Schlick in his paper 'Über das Fundament der Erkenntnis', and contains a detaüed analysis of Schlick's concept of "Konstatierung". It is argued that in spite of its basically foundationalist appearance, Schlick's theory resembles in important respects contemporary coherence theories of knowledge.

The relation between theory and reality is an important problem for phüosophy of science. Positivists or logical empkicists of the Vienna Circle have tried to solve it by postulating several types of so-called basic statements induced by immediate experience or observation. According to Neurath protocol sentences are distinguished from other synthetic sentences only m virtue of their syntactical form. Since consistency is a relation concerning sentences only, not a sentence and any immediate experience,he contends that there remains an unbridgeable gap between observationstatements and observation itself and consequently he adopts a coherentist strategy of corroboration, which Schlick, as a thorough empiricist, wants to avoid. For that purpose he introduces the notion of "Konstatierung". Such ascertainments play a decisive role as endpoints in the corroboration of theories insofar as they serve to verify predictions which express expected ascertainments. In spite of the fact that they are synthetic wc recognize their truth when grasping their meaning and, therefore, error is precluded in their case, while protocol sentences remain mere hypotheses. Quine attempts to overcome the weaknesses of all prior positions by proposing a purely behavioristic definition of 'observation sentence' which avoids the pitfalls of former sense-data theories as well as the ones of cultural relativism or irrationalism. Considering science to be a bridge linking together our sensory stimuli he comes to the view that empirical support of a theory is a consequence of the logical relation between the theory formulation and the observation categoricals derivable from it. He considers the ideological and epistemological neutrality of his conception as one of its main merits. Its feasibility depends, however, on the question whether, by his behavioristic method, he effectively can explain how linguistic expressions acquire their meanings.

What I want to focus on in this paper is the question of the connection between the positivism of the Vienna Circle — the "scientific conception of the world" — and politics. The Vienna Circle will be considered first ''als soziale Bewegung'' and second from the point of view of "Sozialforschung". The paper is a case study in the problem of the relation of a theory to practice, and more particularly, of the relation of a technical epistemological and methodological theory of science to social practice. The critical assessment of the nature of the Vienna Circle as a social movement starts from the consideration of the views of Otto Neurath; and concludes with some theses about the failure of this social movement precisely as it relates to the logical positivist and logical empiricist theory of science, and to its scientific conception of the world.

Wittgenstein was not only an inspirational figure for Schlick but also contributed to scientific philosophy as Neurath demanded. His verificationism is one instance of this, but it is also shown in his treatment of probability (where his ideas were developed further by Waismann). Wittgenstein revived Bolzano's logical interpretation of probability, anticipating Carnap and many moderns. He construed laws of nature as hypotheses that we had to assume. It is the general form of these hypotheses (what he later called a worldview) and not (pace von Wright) relative frequency that provides the basis for judgements of probability.

Utilitarianism, as well as many other political and moral doctrines, presupposes that the problem of interpersonal utility comparisons can be solved. Otto Neurath gave a comparatively early (1912) and explicit statement of this problem, and he suggested that it cannot be solved. This may still be the dominant view. It is argued that recent attempts to solve the problem (by e.g. Schick, Rescher, Harsanyi, Brandt, Jeffrey, Arrow, and Hare) are unsatisfactory, but that the oldest suggestion - i.e. the method of minimal units or just-noticeable differences - is acceptable from the point of view of utilitarianism.

The dispute between Schlick and Neurath over het foundations of empirical knowledge illustrates the difficulties m trymg to draw epistemological conclusions from a verificationist theory of meaning. It also shows how assummg the general correctness of science does not automatically avoid, or provide an easy answer to, skepticism. But while neither Schlick nor Neurath arrived at a satisfactory account of empüical knowledge, there are promising hmts of a better theory m their writmgs. Following up these hints, and drawing on further ideas m Hempel, Carnap and particularly Quine, I suggest the direction I think a naturalistic epistemology should take.

Carnap opposes physicalist language to phenomenal language. His elementary physicalist sentences convey descriptions which physicists still regard as phenomenal and subjective. A second order physicalism (principle of special relatively) is required in order to express physical laws. Carnap makes the phenomenal language a proper part of the physicalist language. This relation is compared to the relation that general relativity establishes between geometry and physiscs.

The original version of the Principle of Verifiabüity (PV), formulated as "The meaning of a proposition is the method of its verification" (Schlick, quoting Wittgenstein), can be criticised as ungrammatical. Schlick's claim that it was a "truism" reflecting commonsense and scientific practice is refuted by PV's paradoxical consequences. Its users faüed to distinguish between operational and situational readings, the latter of which invokes a mythology of comparison with "facts". Wittgenstein rightly described PV as a "rule of thumb" of limited usefulness.

Alexius Meinong developed a notion of defective objects in order to account for various logical and psychological paradoxes. The notion is of historical interest, since it presages recent work on the logical paradoxes by Herzberger and Kripke. But it fails to do the job it was designed for. However, a technique implicit in Meinong's investigation is more successful and can be adapted to resolve a similar paradox discovered by Romane Clark in a revised version of Meinong's Theory of Objects due to Rapaport. One family of paradoxes remains, but it is argued that they are unavoidable and relatively harmless.

Voluntarism is the view that it is from our intimate awareness of the exercise of our wills in performing actions that we arrive at our concept of causality. This view has generally been thought to be indefensible since Hume attacked it in the Treatise and Enquiry. A variant of the position is stated and defended. The views of Castaiieda, and psychologists such as Maine de Biran, Michotte, and Piaget add clarity and enhance the plausibility of the view.

The law of excluded middle is usually considered as intrinsically connected with the realistic standpoint and incompatible with the idealistic position. This is just what Ajdukiewicz claims in his critique of transcendental idealism. The analysis of Ajdukiewicz's argumentation raises the problem of validity of the law of excluded middle for vague (or incomplete) languages. The problem is being solved by differentiating between the logical (or ontological) and the metalogical (or semantical) law of excluded middle: in contrast to the former, the latter is claimed to be invalid for the languages in question, without thereby embracing the idealist position.

The purpose of this paper is first to discuss and criticize some general theses about interpretation. The concept of interpretation is shown to be more complex than these general theses indicate. Distinctions are suggested between different types o f interpretations, and an attempt is made to state criteria of interpretation and arguments which can be used to support or criticize proposed interpretations. The relations between the various types of interpretations and the criteria (arguments) are then explored. Can they be combined? How? Can they comie into conflict with each other? How are such conflicts in that case to be solved? It is argued that at least partly different criteria and arguments are used, and ought to be used, when different types of interpretations are proposed, checked, and criticized. Sometimes a particular criterion is given, and ought to be given, different weight when different types of interpretations are considered.

Various criticisms have been brought against a Platonistic construal of the musical work: that is, against the view that the musical work is a universal or kind or type, of which the performances are instances or tokens. Some of these criticisms are: (1) that musical works possess perceptual properties and universals do not; (2) that musical works are created and universals cannot be; (3) that universals cannot be destroyed and musical works can; (4) that parts of tokens of the same type can be interchanged and still yield tokens of that type, whereas we cannot interchange parts of performances of the same work and still get performances of the work. Of these claims, (1) and (2) seem to be true, but are not incompatible with a Platonistic construal of the musical work, whereas (3) and (4) just seem to be false and, therefore, of no concern to the musical Platonist.

Problems arising from two issues are examined and resolved: those having to do with reference and denotation involving fictional entities (associated with avoiding truth-value gaps and with adhering to extensional accounts) and those having to do with the realist/idealist controversy - and with confusions due to mingling the two issues. Discussion ranges over the views of Russell, Quine, Strawson, Searle, Beardsley, Ryle, Wolterstorff, van Inwagen, de Man, Bakhtin, Goodman, and others. The solutions offered depend on sorting actual persons, actual stories, and imaginary or fictitious persons; and on treating reference in a purely grammatical way, without ontological import in itself but without precluding ontological interpretation.

Peter Kivy has developed a general philosophical account of musical expressiveness based on baroque writings. But he omitted the association which baroque accounts make between the arts of music and rhetoric. It will be argued that one cannot capture the specifics of baroque musical expressiveness without taking account of baroque rhetorical theory. The detailed analysis of an example will demonstrate how rhetorical analysis of baroque music can fill in the details of Kivy's schematic account of musical expressiveness.

Music often pictures emotion through representing its expression and is thereby able to bear insight into significant aspects of emotional life. Scruton's arguments for denying that music is significantly representational is shown to fail, musical pictures having their own sort of determinacy. Musical representation is dramatic. Musical sounds play the role of expression. They themselves are portrayed as expressing the emotions which we thus represented. But musical drama is distinct from literary drama.

Numerous writers on photography and motion pictures have claimed that photographically originated pictures are essentially different from handmade pictures. Arguments made on behalf of the essential difference of photographs from other kinds of pictures generally depend upon one or another of two models of the photographic process: the visual model claims that photographs are closely allied to vision and show what we would have seen from the standpoint of the camera at the time of exposure; the mechanical or automatic model claims that irrespective of what a photograph looks like, it is a reliable index of what was the case at the moment of exposure. Each of these models is examined and shown to be faulty on either or both factual and/or conceptual grounds. Stanley Cavell's assertions about the "automatic" nature of photography are examined in some detail and shown to be either equivocal or false. It is suggested, in closing, that sharp, categorial differences between photographs and handmade pictures do not exist and that questions about the differences between photographs and, say, paintings, can be solved only within the context of viewing particular photographs and particular paintings. In sum, claims about the ontological distinctions between photographs and handmade pictures cannot be warranted.

Talk of fictions is usually problematic. One reason is our habitual difficulty in distinguishing clearly between discourse about fiction and fictional discourse. And part of our problem is understanding more clearly what such various discourse refers to. In this paper I would like to examine critically a recent influential account of "fictional discourse" with a view towards offering several proposals for reconstructing that account.