The Problem of Validity Proofs Timm Lampert, Michael Baumgartner ∗ Abstract In philosophical contexts, logical formalisms are often resorted to as a means to render the validity and invalidity of informal arguments formally transparent. Since Oliver (1967) and Massey (1975), however, it has been recognized in the literature that identifying valid ar- guments is easier than identifying invalid ones. Still, any viable theory of adequate logical formalization should at least reliably identify valid arguments. This paper argues that ac- counts of logical formalization as developed by Blau (1977) and Brun (2004) do not meet that benchmark. The paper ends by suggesting different strategies to remedy the problem. 1 Introduction One of the main functions that is commonly attributed to logical formalisms in philosophical contexts consists in rendering the validity or invalidity of informal arguments transparent. The (in)validity of an argument is not unambiguously de- terminable based on its natural language form. Logical formalisms are widely seen to provide the theoretical fundament for systematic and reliable validity assess- ments. However, before formalisms can be put to work in the reconstruction of natural language arguments, the latter’s component statements must be transferred into the syntax of a pertaining formalism, i.e. they must be formalized. Theories of formalization aim to provide a systematic understanding of logi- cal formalizations. The project of developing such theories can be (and has been) approached in three different ways: (A) by conceptually explicating the notion of formalization, (B) by specifying criteria of adequate formalization, and (C) by defining an effective formalization procedure. For reasons given in what follows we take approaches (A) and (C) to face severe problems and, hence, join authors as Blau (1977) and Brun (2004) in giving preference to approach (B). As formal- izing a statement, in one way or another, amounts to representing or identifying the logical form (or the logical structure or the logical features) of the pertaining statement (cf. Borg and Lepore 2002, 99, Blau 1977, 3, Brun 2004, ch. 4), expli- cations of the notion of formalization must, in one way or another, draw on the * We profited a lot from intensive discussions with Georg Brun about his book and earlier drafts of this paper. Moreover, we are grateful to Sebastian Leugger, Johannes Marti, Tim Raez, and to the anonymous referees for this journal for helpful comments on an earlier draft. Finally, Michael Baumgartner would like to thank the Swiss National Science Foundation for generous support of this work (grant PP001-114812/1). 1