![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
ABSTRACT
On the centenary of the Treatise on Probability's publication, this paper explores six related topics: varieties of realism; Ramsey's critique; Keynes's response; radical uncertainty; the continuity/discontinuity question regarding Keynes's philosophy and economics; and the possibility of a non-Platonic interpretation of logical probability. Several new perspectives are presented.
Acknowledgements
I thank the anonymous referees for helpful comments.
Disclosure Statement
No potential conflict of interest was reported by the author(s).
Notes
1 None of these, however, are immune from theory-dependence.
2 This aligns with Keynes's 1938 view of science as ‘the application of logic and rational analysis to the material presented as sense-data’ (Keynes, Citation1972, p. 438).
3 Platonism is a much debated doctrine with various forms. In the present context, a key aspect is the following. Behind observed reality there lies a more real and perfect reality inhabited by Forms or Ideas that are metaphysical in being non-spatiotemporal, eternal and changeless. The Idea of X expresses what X really is, so that understanding that Idea delivers comprehension of X's many particular occurrences in spatiotemporal reality.
4 These contrasts relate to differences between both the probability interpretations of Keynes and Ramsey, and the economic theories presented by the GT and orthodoxy (whether Pigovian or Walrasian).
5 By contrast, Ramsey based his probability interpretation on ‘ideal’ agents (see below).
6 Interestingly, self-reference is not a property of other probability interpretations such as frequentism or subjectivism.
7 Note that the word rational excludes purely subjective interpretations. Non-arbitrary reasons must underpin the attributed degree of rational belief in both logicism and subjectivism. That is, there must be grounds other than a complete absence of grounds (see below).
8 See Khlentzos (Citation2021) for a survey. Concerning Keynes's acknowledgement of the influence of Platonism, see Keynes (Citation1972 pp. 438, 442, 444–445).
9 For Laplace all probabilities were numerical; if we didn't know their magnitudes, they became unknown probabilities.
10 Sahlin's introduction to Ramsey (Citation1990b) has a misprint removed in Sahlin (Citation1990, p. 234).
11 This contrast then carries over to economics: the orthodox treatment of uncertainty using the subjective probabilities approach that Ramsey helped establish, and Keynes's (TP-based) treatment of uncertainty in the GT.
12 Ramsey's associations with pragmatism have recently received increased attention; see Dokic and Engel (Citation2002) and Misak (Citation2016; Citation2020), for example.
13 Numericality appeals to mathematicians because it allows the probability calculus universal applicability.
14 Keynes did not write enough on Ramsey's thought for the converse to be explored.
15 See Ramsey (Citation1922a, Citation1922b and Citation1926). This complete misconception is commonly accepted, as in Braithwaite (Citation1975, pp. 238–239), Gillies (Citation2006, p. 205) and Misak (Citation2016, p. 175; Citation2020, p. 114).
16 Keynes Citation1973a, pp. 37–43).
17 Possibly Ramsey overlooked this due to his rejection of Keynes's principle of indifference.
18 Other misunderstandings of Keynes by Ramsey also exist.
19 See also below.
20 Note that while Keynes's arguments usually run forwards from premises to conclusions, they can also run backwards (or abductively) from conclusions to most likely premises. As it was Pierce who advanced abductive reasoning, it is surprising that Ramsey never noted this connection in any of his various criticisms of Keynes.
21 Keynes Papers. King's College, Cambridge.
22 This final sentence is linked to remarks made in the TP and 1926 concerning vague knowledge (see below). Both frequencies and psychology play major roles in Ramsey's epistemological comments concerning ‘human logic’.
23 Consilient with this is his remark that ‘Inductive processes have formed … at all times a vital, habitual part of the mind's machinery’. (Keynes Citation1973a, p. 241).
24 Earlier figures (and later ones such as Pascal and Butler) framed religious belief as a wager in which rational probabilistic calculation recommended belief in deities (Keynes Citation1973a, p. 355). On Keynes, Darwin and Paley, see O’Donnell (Citation2020).
25 Post-1926, Ramsey became more critical of his earlier views (Ramsey Citation1931; Citation1990a).
26 Keynes (Citation1972, p. 336, also p. 337).
27 Misak does not refer to the discussions in economics since the 1980s.
28 This second type is impossible in Ramsey's account since agents never lack probabilities.
29 See also his Galton lecture discussed below.
30 This is related to the GT's earlier observation concerning long-term expectations: ‘Worldly wisdom teaches that it is better for reputation to fail conventionally’ (Keynes, Citation1936, p. 158). When handled appropriately, the contributions of psychology and behavioural economics are relevant here.
31 This partly prefigures Friedman's 1950s ‘as if’ methodology.
32 If pure theory constructed along these lines is to reconnect with reality, the addition of something further is required in a subsequent stage of theorising, as J.S. Mill recognised.
33 This reflects Keynes's QJE remarks on longevity.
34 See O’Donnell (Citation1989, Citation1991).
35 It is also consistent with Keynes's 1938 remarks to Townshend that, in making decisions when consequences cannot be evaluated, we ‘necessarily' fall back on other factors such as ‘habit, instinct, preference, desire, will etc’ (Keynes Citation1979, p. 294).
36 Virtually identical mistaken criticisms and remarks occur in Braithwaite (Citation1975).
37 In this context, Keynes's 1938 memoir, ‘My Early Beliefs’, is not a key source of understanding. Ramsey is unmentioned, and it is a complex, sometimes inaccurate, and sometimes misleading document, as argued in O’Donnell (Citation1982, Citation1989, Citation1991, Citation2020).
38 This paper expands on the TP–GT continuity thesis I initially developed in 1978–79, presented in O’Donnell (Citation1982) and further elaborated in O’Donnell (Citation1989; Citation1991; Citation2013). For the 1982 and 1989 works, an empirical test was conducted: every word in the available volumes of the Collected Writings was read to see if Keynes used probability or probable in ways inconsistent with the logical approach. No instances were found. Other researchers are invited to repeat the test.