... ¹

Girton College, University of Cambridge, Huntingdon Road, Cambridge, UK. CB3 0JG

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...$^{\textrm{,}}$²

Physics and Chemistry of Solids Group, Department of Physics, University of Cambridge, Madingley Road, Cambridge, UK. CB3 0HE

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... ³

Mineral, Ice and Rock Physics Laboratory, Department of Earth Sciences, University College London, Gower Street, London, UK. WC1E 6BT

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... ⁴

vi5u0-website@yahoo.co.uk

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... logic⁵

The rules of logic are [67] themselves among the postulates.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... omission⁶

The author apologizes for any confusion caused by the fact that the definition of ``hypothesis'' here is different from that used by MacKay [42,43,44], or by the fact that the definition of ``value'' here is different from some commonplace definitions.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...Mies:1993:MFR⁷

In the latter case, the author has taken a small, and, it is hoped, justified, liberty, in identifying a work which does not mention the word ``standpoint'' as part of the standpoint epistemology; it is the author's assumption throughout that ``standpoint epistemology'' can be used to describe a collection of insights, mostly from the new social movements, much broader than that which explicitly goes under the name.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... made⁸

Kolakowski [39] does not unambiguously identify the entities whose number is to be minimized as postulates. However, both Jeffreys [36] and Rae [58] interpret the positivist Occam's razor as a principle of fewest postulates.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... science⁹

One historian of art [9] has gone as far as to suggest that research conducted, without an ideology in mind, can be understood as an artistic movement, which arose in the early twentieth century, rather than as a continuation of the earlier traditions of science.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... physics¹⁰

The author apologizes for any confusion caused by the fact that the definition of ``reductionism'' here is different from that used by Harding [28].

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... probability¹¹

The idea that not all phenomena are well described by mathematical language is unique neither to Bayesian inference, nor to standpoint epistemology. Without reference to either tradition, and in the context of astronomical observations, Eddington [17] has made this point eloquently.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... choice¹²

Care is needed over what is an ``initial'' prior probability distribution. The initial prior probability distribution is the $P_0(T)$ before approximate account (section 5.1) has been taken of previous evidence.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... affected¹³

The idea that theoretical beliefs inevitably affect the interpretation of empirical evidence is unique neither to Bayesian inference, nor to standpoint epistemology. Without reference to either tradition, and in the context of astronomical observations, Eddington [17] has made this point eloquently.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...associate¹⁴

The mapping from ideologies to uncertainty functions is many-to-one. Therefore, the ideology associated with a given uncertainty function cannot be uniquely specified.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... theory¹⁵

The commitment to philosophical realism involved in this supposition is not [36] a requirement for the adoption of Bayesian statistics as an epistemology, nor is it necessary to any of the arguments in other sections of this paper.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... mechanics¹⁶

The assignment of cases where the new postulates are derived rigorously from statistical properties of quantum mechanics to the ``emergent phenomena,'' rather than to the ``first principles,'' tradition is somewhat arbitrary; this choice has been made here to generate a clear distinction in Bayesian statistical properties.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... configurations¹⁷

Even the sterling work of researchers such as Lind et al. [16,41], in running band-structure calculations for multiple proposed structures for a particular sample, barely scratches the surface. However, current developments [33] in technology for enabling members of the public to donate CPU time may change this.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... impossible¹⁸

Kuhn's [40] description of certain pieces of evidence, which he terms ``mere facts'' (Kuhn's italics,) could be interpreted as a statement that those pieces of evidence are in this position.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.