Scientific Observation: Data, Phenomena, and Inconsistency Tolerance

Description

This talk addresses ‘observation’ as it pertains in (inconsistent tolerant) philosophy of science. An inconsistent set of sentences, in classical logic, logically explodes: the system can prove everything. In empirical science, inconsistency should be avoided, as a theory that proves and predicts everything is useless. However, inconsistent-tolerant systems avoid such explosive consequences.

Conventionally, observation is central to both philosophy of, and inconsistent tolerant, science. I show that the later provides an alternative account of the methodology (and logic) of science, but has issues: discussions about inconsistencies between theory and observation in science (and inconsistency toleration) appears inadequate as it overlooks the crucial distinction between data (what is directly observable) and phenomena (not directly observable, repeatable, explanatory and predictive). I aim to address this neglected aspect and offer a more nuanced view of scientific methodology.

Inconsistency between Theory and Observation: Data, Phenomena, and Toleration

Long Abstract: max 1000 words

In this talk I will be investigating the notion of ‘observation’ as it pertains to philosophy of science, especially in relation to discussions around inconsistencies between theory and observation.

Traditionalism, following Davey’s (2014) distinction, holds that an inconsistency cannot be held in science. This is because an inconsistent set of sentences, in classical logic, logically explodes: the system can prove everything. In empirical science, inconsistency should be avoided, as a theory that proves and predicts everything is useless. Hempel (2000) and Popper (2005) belong in this tradition. However, counter-traditionalists argue that it is not the case that we must always discard logically inconsistent theories. To support their view, counter-traditionalists usually provide three kinds of inconsistencies in science. This distinction is between inconsistencies between theory and observation, between theory and theory and internal to a theory. The first type of contradiction, viz. between theory and observation is different than the others in a substantive way and this will be my focus here. It is argued that in the history of science there have been inconsistent theories and that scientists can, and do, avoid logical explosion in some way. This job is being done by inconsistent-tolerant systems which avoid such explosive consequences.

I will be juxtaposing literature surrounding observation in philosophy of science and inconsistent tolerant science. My main claim is that literature surrounding inconsistent tolerant science needs development, either because it does not delve enough into the concept of observation (Martinez-Ordaz 2017; Alegre 2024), or because it misrepresents how observation is actually viewed in science. For example, discussions like ‘theory T predicting observation O, but upon experimentation we actually observe not-O’ (Priest 2002). I argue that terms such as ‘observation O’ are not adequate to describe the nature of observations in empirical science.

The reason for doing so is that I think that the inconsistent tolerant philosophy of science literature is interesting as it provides a robust, alternative and persuasive account of the internal methodology (and logic) of science, but at the same time it looks like it falls short in some aspects. For example, there is discussion about inconsistencies between theory and observation in empirical science, and inconsistency toleration approaches, which I think it is not enriched with certain advancements in philosophy of science regarding the concept of ‘observation’. Notably, Woodward and Bogen’s (1988) crucial distinction between data and phenomena. Data (what is directly observable) and phenomena (not directly observable, repeatable, explanatory and predictive) have been wrongly conflated, but are in fact different things. Data give rise to phenomena, and phenomena give rise to scientific theories.

An example of the nature of data and phenomena, and their relation with theory, can be extracted from Einstein’s general relativity: data that were collected, during an experiment testing the theory, where not precisely predicted by the theory. That is, “the locations of individual spots on photographic plates, were hardly themselves predicted by general relativity!” (Teller 2010: 816). What general relativity predicted was, in fact, the phenomenon of the bending light rays, and not the “individual spots on photographic plates”, the later constituting the data collected (ibid.)

Applying this distinction will not only help to solve the issues raised above, but also situations in which authors sometimes place direct observations under the term ‘observation’ and then talk about how (direct) observations are treated. And as it is well known, treating ‘observation’ to mean ‘direct observation’ leads to many issues, because we are conflating data (individual observations) with phenomena (which have stable characteristics and may yield different data), and these are two different things.

Bridging discussions between inconsistent tolerant science (which can involve discussions concerning non-classical logical frameworks in science) and general philosophy of science will not only benefit both, but also open a new path in how the role of observation functions. Therefore, this essay aims to address this neglected aspect of the literature, and by extension add something helpful and meaningful to the insightful literature regarding philosophy of science and inconsistency toleration.

Bibliography:

Alegre, L. F. B. (2024) Can We Test Inconsistent Empirical Theories. Forthcoming 

Davey, K. (2014) Can good science be logically inconsistent?. Synthese. 191. 10.1007/s11229-014-0470-x

Hempel, C. G. (2000) Selected philosophical essays. R. Jeffrey (ed.), New York: Cambridge University Press.

Martínez-Ordaz, M.d.R. (2017) Holism, Inconsistency Toleration and Inconsistencies between Theory and Observation. Humana Mente 10 (32):117-147.

Popper, K. (2005) The Logic of Scientific Discovery (2nd ed.) Ebook. Routledge. https://doi.org/10.4324/9780203994627  

Priest, G. (2002) Inconsistency and the Empirical Sciences. In: Meheus, J. (eds) Inconsistency in Science. Origins, vol 2. Springer, Dordrecht. https://doi.org/10.1007/978-94-017-0085-6_7 

Teller, P. (2010). “Saving the Phenomena” Today. Philosophy of Science, 77(5), 815–826. https:// doi.org/10.1086/656537 

Period	27 Aug 2025 → 30 Aug 2025
Event title	European Philosophy of Science Association
Event type	Conference
Location	Groningen, NetherlandsShow on map
Degree of Recognition	International