Every stage of inquiry requires choices the evidence cannot make for the researcher

The argument

Those who deny that objectivity is attainable argue that the ideal breaks down well before anyone reaches the stage of reading a result. Someone has to choose which question is worth asking out of the many available, which effects count as the ones to measure, what will be treated as the background to be held constant and what as the variable of interest. Nothing in the evidence makes those choices, because they are made before the evidence exists. On this account the inquiry has already been shaped by the time measurement begins. The same argument is made about observation itself. What a researcher records is not raw sensation but a description in the vocabulary of a theory, and that vocabulary decides what can be noticed. A reading becomes a signal or an artefact depending on expectations about how the apparatus behaves, and an unexpected result can be reported as a discovery or set aside as a fault, with no rule determining which. Critics argue that experienced researchers make these calls constantly and could not work if they did not, which is the point: the judgment is unavoidable, so the neutrality is not available. Finally they note that the standards themselves embody choices. How much evidence is enough, how to weigh the risk of accepting a false claim against the risk of missing a true one, which simplifications are acceptable: each is a decision about what to value, and different decisions yield different conclusions from the same data. Critics conclude that such judgments are not failures of rigour to be eliminated by better practice, but conditions of doing the work at all, and that a practice built on them cannot be called objective in the sense its defenders want.

Premises

Counter-arguments

Defenders of objectivity answer that this describes how inquiry starts rather than how it ends. Choices about what to study and how to measure it are indeed made by people, but they are then written down, and a choice that was distorting shows up when someone who made a different one fails to reproduce the result. They add that the theory-laden character of observation is a reason for instruments and procedures fixed in advance, not an argument against them, since both exist to settle beforehand what will count as a signal. The appeal to standards is said to prove too much: if judgment anywhere defeats objectivity everywhere, then no practice could be objective, and the word would describe nothing rather than describing science imperfectly.

Rejecting the premises

[Rejecting P1] Choices made before the evidence exists are recorded and can be examined, and a choice that distorted the outcome shows up when someone who chose differently cannot reproduce it. [Rejecting P2] That observation is described in theoretical vocabulary is the reason instruments and procedures are fixed in advance, which settles beforehand what will count as a signal rather than leaving it to the observer. [Rejecting P3] Standards are chosen, but they are chosen publicly and applied uniformly, and a result that depends on which standard was used is identified as such rather than passing unnoticed. [Rejecting C] If any element of judgment defeated objectivity, nothing could be objective, so the argument proves too much: it shows objectivity is demanding, not that it is impossible.