This paper relates to theories of knowledge underpinning educational research. The central concern is to discuss theories of knowledge in relation to interpretations of learning in educational research, that is, to ascribe meaning to observed behaviour by the formulation of interpretative theories. A departure is taken in Greeno, Collins, & Resnick, (1996), who have roughly distinguished three general research perspectives on learning, which they matched with corresponding views on knowing and learning: behaviourism / empiricism (associated with Thorndyke, Skinner and e.g. Locke), constructivism/ rationalism (Piaget and e.g. Descartes) and situated perspectives / sociohistorical / pragmatist (e.g. Vygotsky and Heidegger). It is argued in the paper that all three perspectives provide important insights into research on learning, but due to their respectively definitions of ‘learning’ each alone runs the risk of reducing learning and interpretations of learning to single aspects. Constructivism can be regarded as a reaction to behaviourism, which did not account for the subjective aspect of learning. Constructivism and its focus on “the inside” of learning, on the other hand, have been criticised for not acknowledging situational aspects (Schoultz, Säljö, & Wyndhamn, 2001; cf. Resnick, Levine & Teasley, 1991). Knowledge with in socioculturalism is essentially intersubjective. It appears to look upon situations in a way that may end in a position where we get caught up in language and discourses (cf. Rönnström, 2011), not accounting for the inside aspects of learning. 

In interpretation of an observed behaviour, including speech behaviour, it may also be possible to make an alternative interpretation that fits equally well with the observed behaviour. Such situation is analogous to what has been referred to in philosophy of science that theories are underdetermined by the data. Underdetermination has been defined by Psillos (2005) as “…a relation between the propositions that express the (relevant) evidence and the propositions that constitute the theory.” (p. 575), and relate to different kinds of holism (cf. Downes, 1998, p. 353; Quine, 1951).  Also known as the Quine-Duhem thesis. A strong version of the thesis states that there is always another theory that can explain the data equally well (but cf. Psillos, 2005). A weaker interpretation of the thesis states that there can be underdetermination (Newton-Smith & Luke, 1978). As such, it poses problem for the interpreter, who neither knows the learner’s beliefs nor can take the meaning of utterances for granted, as that would be to assume what is to be explained. Quine (1951) showed what he regarded as the implication of holism and underdetermination in his famous Gavagai example.  Quine may by that adopt a sceptical position, that although one interpretative theory is true, we are not in position to decide which it is (cf. Psillos, 2005).                                                                                                                                                                                                                   

In order to try to avoid scepticism, interpretation is therefore also discussed in light of Donald Davidson’s (2001a, b, c) theories of knowledge and interpretation. Davidson has been used in an educational context before (e.g., Roth, 2009; Rönnström, 2011; Wahlström, 2010). Here the focus is on how Davidson’s theories of knowledge and interpretation constitute aspects of an ontological and epistemological stance, that may meet the problem of underdetermination and holism, and in what sense such a stance provides an alternative to earlier learning perspectives on knowledge.

It is suggested that his theories may provide aspects of a stance that can form the basis for interpretations of learning in educational research. The paper discusses what such a stance may imply with regard to the nature and location of knowledge and the status of the learning situation, as it methodological considerations are inextricably connected, or even follow from epistemological and ontological assumptions (cf. Bereiter, 1994). 

The research questions are discussed by an interpretation of Davidsons philosophy (2001a, b, c). The problem to be solved by reference to Davidson is that an action is always an action under a specific description hence the same observed behaviour can be described in different ways. Different descriptions may not be contradictory but turning to Davidson means stressing that for the interpreter it is of core importance to understand the learner’s language use, as we otherwise cannot know whether there is a difference in language or in beliefs. It is thus argued in the paper that interpretation is always radical (Davidson, 2001a; Rönnström, 2011). First the arguments for looking at interpretation as radical is presented. Davidsons version of the principle of charity is then outlined, arguing that it minimises the possible negative effects of holism and underdetermination. It is argued that the principle of charity constitutes a priori assumptions we have to make when interpreting learning processes. These assumptions are suggested also to work as heuristics in interpretation. Interpretation then takes the form of triangulation, in which the principle of charity is applied. Reference to shared world is hence crucial and the argumentation relies on Davidsons reference to our ability to see the same salient features in the world. It implies stressing a causal connection between the contents of learners’ thoughts and the world so that a physical situation is causally related to beliefs (cf. Davidson, 2001b, p. 213). It implies a kind of externalism (Davidson, 2001c) which is interpreted in the paper as an alternative to empiricism, as it maintains an objectivist ideal. Such position does not rely on a correspondence theory recognised in logical positivism. Language relates to the world in a natural way by means of triangulation, without the presumptions of any intermediaries (Davidson, 2001a; cf. Roth, 2009). It rather relies on subjectivity, intersubjectivity and objectivity that stand in a dependent relation to each other (Davidson, 2001b, p. 220). The argument in favour for Davidsons theory is further based on his view on meaning and evidence being public, by way of people communicating about objects and events in the world. In such interaction, we will sometimes experience that we are wrong, and we are also able to realise that this is due to an objective fact.

It is argued that Davidsons view on knowledge solves the problem of underdetermination. Learners' beliefs may actually be observable, or the potentiality is there in triangular situations. In other words, we do not need to give up the first-person perspective which rationalism implies: that we seem to know what we our- selves believe and that we have direct access to our beliefs in a way that other people do not have (Davidson, 2001b, pp. 205-220). We are born into a shared world and intersubjectivity accounts for people’s ability to communicate, which is also depended on objectivity, the possibility of being wrong. Hence, this can be regarded a naturalistic approach to knowledge, meaning and interpretation (cf. Bahagramini, 1998). Moving from the level of epistemology to an empirical level the paper discusses some possible research strategies, arguing that the principle of charity (Davidson, 2001b) and reference to saliency, that is, what in the learning situation that appears as the figure for the learner, may help us in the actual analysis of qualitative data. This by critically analysing the observed behaviour in the particular situation and in such triangulation process apply the principle of charity. Although interpretations are in one sense necessarily based on the interpreters’ own frames of reference, this does not mean that we need to reduce the learners’ perspectives on our own world view, or that learners are always right. However, saliency does not only appear as a phenomenon in relation to physical objects and events, but also in the symbolic world, thus requires that the analysis extend beyond the mere transcription of an interview or the description of an observation. A conclusion to be drawn is that the very question of what counts as data in the interpretation of complex learning processes is up for discussion.

Baghramian, M. (1998). “Davidson and Indeterminacy of Meaning” Twentieth World Congress of Philosophy, in Boston, Massachusetts from August 10-15, 1998. Retrieved 16062016, from https://www.bu.edu/wcp/Papers/Lang/LangBagh.htm Bereiter, C. (1994). Constructivism, socioculturalism, and Popper's world 3. Educational Researcher, 23(7), 21-23. Davidson, D. (2001a). Inquiries into truth and interpretation (2nd ed.). Oxford: Oxford Clarendon Press. Davidson, D. (2001b). Subjective, Intersubjective, Objective. Oxford: Oxford Clarendon Press. Davidson, D. (2001c). “Externalisms” in Interpreting Davidson, (Eds.). P. Kotatko et al., Stanford, CA: CSLI Press. Downes, W. (1998). Language and Society. Cambridge University Press, Cambridge. Greeno, J. G., Collins, A., & Resnick, L. B. (1996). Cognition and learning. In D. C. Berliner & R. C. Calfee (Eds.), Handbook of educational psychology (pp. 15-46). New York: Macmillan. Psillos, S. (2005). Underdetermination thesis, Duhem-Quine thesis. Encyclopedia of Philosophy. 2nd edition. Retrieved 06192016. Quine, W.V. (1951). “The Two Dogmas of Empiricism”. The Philosophical Review 60 (1951): 20-43. Resnick, L. B., Levine, J. M. & Teasley. S. D. (Eds.). (1991). Perspectives on social shared cognition. Washington, Rönnström, R. (2011). Cosmopolitan Communication and the Broken Dream of a Common Language. Educational Philosophy and Theory, 43:3, 260-282. Roth, K. (2009). Some Thoughts for a New Critical Language of Education: Truth, justification and deliberation. Philosophy & Social Criticism, 35: 685–703. Schoultz, J., Säljö, R., & Wyndhamn, J. (2001). Heavenly talk: Discourse, artefacts, and children’s understanding of elementary astronomy. Human Development, 44, 103-118. Wahlström, N. (2010). Learning to communicate or communicating to learn? A conceptual discussion on communication, meaning, and knowledge. Journal of Curriculum Studies. Vol. 42, 2010 – Issue 4.