2.1 – Economists’ received grounds of finality

8 Over 1899 and 1900, Veblen published three papers, or three largely self-contained and separately published parts of one very long paper, all with the title ‘The preconceptions of economic science [3]’. The enquiry there reported by Veblen is one in metaphysics, this being Veblen’s word for ontology as I am employing the latter term, meaning the study, or a conception, of the more fundamental features of social being including the natures of social phenomena. Veblen’s interest was in the ontological or metaphysical conceptions or ‘preconceptions’ of earlier generations of economists, specifically in their « underlying metaphysics of scientific research and purpose [4] ». Those preconceptions on which Veblen focussed most related to the ‘received grounds of finality’, namely the preconceptions that determine or ‘ground’ the form to which researchers felt their results must conform to be acceptable as final and ready for public consumption and scrutiny. Veblen uncovers how these have continually evolved over time and in a systematic fashion ; that is, he details how « changes which have supervened in the preconceptions of the earlier economists constitute a somewhat orderly succession [5] » in these received grounds of finality :

9

The feature of chief interest in this development has been a gradual change in the received grounds of finality to which the successive generations of economists have brought their theoretical output, on which they have been content to rest their conclusions, and beyond which they have not been moved to push their analysis of events or their scrutiny of phenomena. There has been a fairly unbroken sequence of development in what may be called the canons of economic reality ; or, to put it in other words, there has been a precession of the point of view from which facts have been handled and valued for the purpose of economic science [6].

10 With a coverage that includes the underlying metaphysical preconceptions of the Physiocrats, Adam Smith, the utilitarian economists, especially Jeremy Bentham, and culminating in the more recent British contributors such as John Stuart Mill and especially John Elliott Cairnes, Veblen traces in particular how presuppositions of event correlations constrained analyses throughout. In this, Veblen notes especially how, accompanying this emphasis on the correlation of events, is an early metaphysical commitment to animism, the belief that supernatural powers organise and animate the material world, and he details how this « more archaic metaphysics of the science, […] saw in the orderly correlation and sequence of events a constraining guidance of an extra-causal, teleological kind [7] ». He also observes how this early commitment to animism gradually dissolved to be replaced by visions of developments guided by natural law and a natural order, on the way :

11

The history of the science shows a long and devious course of disintegrating animism, – from the days of the scholastic writers, who discussed usury from the point of view of its relation to the divine suzerainty, to the Physiocrats, who rested their case on an « ordre naturel » and a « loi naturelle » that decides what is substantially true and, in a general way, guides the course of events by the constraint of logical congruence. There has been something of a change from Adam Smith, whose recourse in perplexity was to the guidance of « an unseen hand » to Mill and Cairnes, who formulated the laws of « natural » wages and « normal » value [8].

12 If overtime, then, a presumption of events correlations prevailed, the associated or accompanying account of the relevant agencies continually changed, becoming in particular notably less animistic.

2.2 – Taxonomic versus evolutionary economic science

13 Motivating this study of ontological or ‘metaphysical’ preconceptions, Veblen had a particular overriding objective. He was seeking to determine whether, and if so, how fast, the commitment to a conception of science that he labelled ‘taxonomic’ was being replaced by an ‘evolutionary’ science. To this end, he was interested in assessing the relative fates of the different specific and basic « grounds of finality for science » associated with these different conceptions of science.

14 For Veblen a taxonomic science is one that is concerned with the normal or regular. In terms of method, it is, in essence, the search for correlations or event regularities. In the earlier periods covered by Veblen’s research, the content of normality was vital to the taxonomic project, taking the form of, or including, states regarded as « natural », « normal » or « good » to which economic processes tended. With time, however, this focus had lessened and taxonomic analyses had lost their ‘colour’, with the result that, in contemporary contributions, the notion of the regular or normal was mostly taken to apply only at the level of the patterning of events. And so the primary focus was method, with the taxonomic approach in question thus amounting, as I say, to the search for correlations.

15 Of course, where taxonomic scientists accept a reality of causal entities bound up with and underpinning (or anyway acting consistent with) these correlations, and most do, then the causal entities must be formulated in particular sorts of ways. Specifically, it must be supposed that these causal entities are of a sort as, first, to exert the same independent, invariable, effect whatever the context and, second, to exist, or at least to act, in isolation. I use the term atom to refer metaphorically to a causal factor of this sort ; the noted set of conditions are those I am systematising as the atomistic ontological conception.

16 Why these two conditions ? Correlations of the generic form ‘whenever event X then event Y’ (however simple or complex, whether linear or non-linear, deterministic or stochastic, a priori or a posteriori etc.) simply require for their existence that causal factors are atomistic in this sense to guarantee that every time conditions X are found this causal agent produces the same effect Y. It is required that these causal factors act in isolation to ensure no other factors interfere with a realisation of the outcome Y.

17 Veblen provides a colourful example in describing the atom of a (then recent) form of taxonomic science as an isolated, passive, inert, homogenous, and self-contained ‘globule of desire’ :

18

In all the received formulations of economic theory […] the human material with which the enquiry is concerned is conceived in hedonistic terms ; that is to say, in terms of a passive and inert and immutably given nature […] The hedonistic conception of man is that of a lightning calculator of pleasures and pains, who oscillates like a homogeneous globule of desire of happiness under the impulse of stimuli that shift him about the area, but leave him intact. He has neither antecedent nor consequent. He is an isolated, definitive human datum, in stable equilibrium except for the buffets of the impinging forces that displace him in one direction or another. Self-imposed in elemental space, he spins symmetrically about his own spiritual axis until the parallelogram of forces bears down upon him, where upon he follows the line of the resultant. When the force of the impact is spent, he comes to rest, a self-contained globule of desire as before. Spiritually, the hedonistic man is not a prime mover. He is not the seat of a process of living, except in the sense that he is subject to a series of permutations enforced upon him by circumstances external and alien to him [9].

19 The relational-processual ontological conception that I am identifying as the second metaphysical preconception of interest to Veblen is effectively the antithesis of this atomistic contrivance. For according to it, causal factors are far from fixed or atomistic but (and like everything else that is social) in continual transformative process. And rather than acting in isolation, such factors (as again with most other social phenomena) are everywhere considered to be relationally constituted. Moreover, where regularity or stability emerges it is typically found not at the level of events at all but at a deeper one of underlying processes of causation and so on. These features are consistent with the future being open and unfolding in unpredictable causal sequence rather than in conformity with pregiven event regularities. It is these four features of process, relationality, depth, and openness that I take to be core to the relational-processual conception.

20 Whereas a commitment to atomism is uniform across contributors to taxonomic science whatever the particular form of the science that is adopted, those that accept the relational-processual conception tend to emphasise different aspects of it according to context. Thus, if Marx can reasonably be said to emphasise social relationality [10], and Keynes focusses on openness in his concern with uncertainty [11], for Veblen the primary concerns are process, and specifically evolutionary process, openness and especially lack of teleology, as well as depth in emphasising cumulative causation in place of correlation.

21 Thus Veblen, drawing on Darwinian evolutionary theory, stresses how the evolutionary scientist « is unwilling to depart from the test of causal relation or quantitative sequence [12]», asking of everything only « why ? », and seeking an answer in terms of cause and effect. It is simply the case that one thing leads to another in a non-predetermined fashion. As mutations occur, or environments of selection change, there can be a change in the traits selected. But there is nothing superior or especially harmonious or natural about the entities that survive or scenarios that come to dominate. They are simply that, namely the ones that survive. All order that comes about is unplanned ; in effect reality manifests an appearance of order if it puts in an appearance at all [13]. The world is open and in process, not harmonious and predetermined in line with notions of normality or regularity. In contrast the taxonomic economist presupposes that « this ground of cause and effect is not definitive [14] ». Rather, the ultimate term in the taxonomist’s systematisation of knowledge is something like a « natural law », an association of phenomena, an empirical generalisation or correlation typically regarded as « natural » or « normal », or a « consistent propensity » with any exceptions regarded as mere disturbing factors. In the evolutionary sciences generally « the evolutionary method and the evolutionary ideals have been placed in antithesis to the taxonomic methods and ideals of pre-evolutionary days [15]».

22 As I say, a major motivation for Veblen’s conducting his historical study was to identify, and if possible to quicken, the process by which economics became evolutionary, and to this end he was interested in how preconceptions of normality had fared in recent and ongoing developments [16]. And because an evolutionary orientation had been adopted elsewhere Veblen thought that economics must follow the trend, and indeed was already ‘caught in it’ :

23

Under the stress of modern technological exigencies, men’s every-day habits of thought are falling into the lines that in the sciences constitute the evolutionary method ; and knowledge which proceeds on a higher, more archaic plane is becoming alien and meaningless to them. The social and political sciences must follow the drift, for they are already caught in it [17].

24 Veblen, then, saw an evolutionary economics as ultimately inevitable and was keen to hurry the process along.

2.3 – Neoclassicism and ontological inconsistency

25 Perhaps a reason that Veblen supposed that economics in particular was already ‘caught in’ the trend of adopting the evolutionary approach was a recognition that some economists coming out of the classical tradition were already revealing support for an evolutionary vision. In this regard Veblen identifies Alfred Marshall, alongside John Neville Keynes (the philosopher father of the economist John Maynard Keynes), as exemplary, as in the vanguard in effect.

26 Veblen, indeed, welcomes this development but observes that even Keynes and Marshall are unable to break sufficiently with the taxonomic ideal of science at the level of method, and this prevents the achievement of a meaningful account of the genesis and developmental continuity of such phenomena. Thus, although acknowledging the causal insights of Marshall and Neville Keynes, and welcoming in particular Marshall’s concern with institutions, Veblen stresses that in the end a significant break with taxonomy is not achieved in methodological practice. Rather he concludes that :

27

There is a curious reminiscence of the perfect taxonomic day in Mr. Keynes’s characterisation of political economy as a ‘positive science,’ the sole province of which is to establish economic uniformities [18].

28 And, Veblen continues,

29

in this resort to the associationist expedient of defining a natural law as a ‘uniformity,’ Mr. Keynes is also borne out by Professor Marshall [19].

30 So, Marshall like Neville Keynes is inconsistent. The vision they both advance is evolutionary, but in each case the « taxonomic bearing is, after all, the dominant feature [20] ». And the sort of inconsistency in question is between the atomistic ontological presuppositions of the (taxonomic) method adopted and the causal-processual (evolutionary) version of the relational-processual social ontology that forms their (evolutionary) vision. It is precisely this inconsistency that is the defining feature, of neoclassical economics, or of neoclassical economists according to Veblen. Contributors like Marshall are classical in maintaining (the then dominant form of) a taxonomic orientation in their choice of method, but they achieve the status of neoclassical by virtue of their simultaneously expressing and supporting a vision, a worldly ontology, that is evolutionary [21].

31 In sum, then, three sorts of ontological orientations or conceptions were in effect discerned in the history of economics undertaken by Veblen, those that I am calling the atomistic, the relational-processual, and an incoherent mix of the two adopted by those economists, including Marshall, that Veblen labelled neoclassical.

3.1 – Cambridge social ontology

36 The focus of the Cambridge group, as noted, has been the process-relational conception of social ontology, and in its concern with social positioning theory [25] specifically the focus has been on developing the relational aspect. A feature that marks this theory out from most contending alternatives of this sort is that although it constitutes a contribution to socio-philosophical ontology (concerned with the nature of social being per se or in general) it flies close to socio-scientific ontology (a concern with the constitution of specific social existents, such as the nature of money, the corporation, Cambridge high tables, and nurses, etc.) in claiming that all (types) of specific social existents are in fact constituted in the same sort of way.

37 Specifically, a contention of the theory is that novel social phenomena are constituted through the (relational) organisation of (mostly) pre-exiting entities (human persons and other phenomena) to form (relational) components of totalities including, especially where persons are involved, human communities. And even more specifically, the theory contends that these components are constituted by way of the pre-existing entities being allocated to social positions, not least community positions.

38 My focus here is on social totalities in the form of human communities. In the community context each social position consists in a package or set of rights and obligations. Further, each position is associated with a function, which is implicit (and more or less apparent) in the content of the rights and obligations that constitute the position. And the positions themselves divide into person positions and non-person or ‘object’ positions, according to whether parties associated with them are thereby afforded direct access to associated rights and obligations. Those with direct access are designated (specific forms of) person positions. Normally, it is human persons that are allocated to person positions with other entities allocated to non-person or object positions [26].

39 Entities allocated to either sort of position give rise to components of the relevant community, with allowed and required actions (person components) or uses (non-person or object components) determined by the rights and obligations constituting the associated positions. Rights (obligations) constitutive of any particular person position are matched to obligations (rights) also constitutive of (typically different) person positions. Each matched right/obligation pair is a (internal or constitutive) social (power-over) relation. The set of rights and obligations that constitute any non-person position is comprised of a subset of the rights and obligations which constitute person positions in the relevant community – those specifically that bear on the (allowed and required) uses of the occupants of the object position qua positioned occupants or components.

40 So, teachers and students are person positions comprised of packages of rights and obligations that when occupied give rise to components of a school or university (also labelled teachers and students) where many of the rights of one are matched to the obligations of the other, and vice versa. The same sorts of things can be said of employers and employees, landlords/ladies and tenants, and so on. And these person rights and obligations determine how school or workplace equipment, etc., may be used in those communities, and so on.

41 Certain consequences can be seen to flow from even such a brief sketch of the theory. Very obviously any positioned X or community component is not reducible to the X positioned in forming it. Specifically, unlike X, the positioned X is a relational feature of a social totality with a function. This insight serves to resolve puzzles that endure throughout social theorising. Thus, a woman is not reducible to the biological entity allocated to the position woman in a community’s gender system, money is not reducible to the entity (perhaps a precious metal, or a commodity or bank debt) that occupies the position money in a community’s accounting system, and a corporation qua a legal person is not reducible to the community that occupies the position corporation (see endnote 26), itself nested in the position legal person in a community’s legal system [27]. All components are relational in a manner that the corresponding position occupants are not [28].

3.2 – Rainforest Realism

42 A second recent explicit and systematic approach to social ontological theorising is pursued in, or at least has been introduced into, economics by Don Ross [29], being referred to as rainforest realism. It is a conception that Ross develops with the philosopher James Ladyman [30]. Whereas the Cambridge group start their analysis from observed social phenomena and in particular from (often puzzling) generalised features of experience and thereupon seek to elaborate a social ontological conception as an explanation or an account of their conditions of existence, the starting point for Ladyman and Ross is a set of findings of the non-social sciences, especially of (fundamental) physics. For their ultimate goal is to achieve a naturalised metaphysics that unites (in a non-reductionist fashion) the various sciences, including economics, by reference to fundamental physics, where quantum mathematics is the accepted form [31].

43 Ross, in making his case in the context of economics, recognises the two ontological conceptions that I have argued have dominated the discipline, but accepts neither precisely as stated, offering instead a formulation, seemingly presented as a version of the relational-processual conception, but without the relationality. Specifically, Ross accepts the openness and process or better dynamism features of the relational-processual conception along with a stress on complexity. But, in drawing on an interpretation of quantum mechanics, Ross rejects any notion of ontological depth and so of relationality, as least of the constitutive sort. Rather we live in a flat world of regularities or ‘patterns’.

44 More precisely, the ontology accepted by Ladyman and Ross is one of ‘real patterns.’ These are patterns that cannot be reduced in the sense of being identified in, or inferred from, more general data patterns, but yet are far more bountiful than « all the science performed in a finite galaxy will ever be able to fully catalogue [32] ».

45 It might be supposed that Ladyman and Ross are really embracing an ontology of isolated atoms, albeit of a complex form that prevails only over restricted domains. For prima facie such isolated atoms seem to be the only sort of material entity consistent with the noted real patterns. However, Ladyman and Ross explicitly reject the atomistic conception of ontology. The reason for this is that the interpretation of quantum mechanics upon which they draw rejects the idea of entities altogether, whether atomistic or otherwise. Or at best any ‘entities’ are held to take the form simply of real patterns.

46 Drawing further on their interpretation of quantum theory, Ladyman and Ross speculate that the best attempts to represent these real patterns in natural language will always fail, involving aspects that necessarily serve to distort. Rather our only hope, if there is any at all, of representing these patterns, Ross supposes, entails a reliance on methods of mathematics [33].

47

My view is that if there is any representational technology that would be fully adequate to the actual structure of social reality, it will be mathematical rather than linguistic [34].

48

One cannot come close to accurately describing quantum reality without mathematics. Indeed, as argued in Ladyman and Ross (2007), using natural language at all to represent the content of quantum physics invariably has the effect of ‘domesticating’ it, in other words concealing the radicalism of its implications for the accuracy of folk ontology [35].

49 Ross shares with proponents of Cambridge social ontology the assessment that mathematical economic modelling, as so far practiced, has achieved little insight. Unlike the Cambridge school, Ross does not reason that this is because mathematical modelling is the wrong method. Rather the failure is due more to poor usage and limited tools [36]. Ross has some optimism for the future, however, because he expects it to produce advances in technology that will facilitate methods more relevant to identifying underlying fundamental patterns. These advances, though, will involve computer programs, artificial intelligence, and the like, that leave little scope for human input, generating results that will facilitate prediction and control, though allowing limited human understanding.

50

What I expect will happen is that statistical estimation software will become steadily more sophisticated, especially when it is set inside the powerful learning systems that artificial intelligence researchers and engineers are at last delivering [37].

51

On this scenario we would expect to see escalating improvements in the accuracy of prediction of social phenomena, but the theoretical knowledge underlying these improvements would only be implicit in the evolving computer models and would increasingly elude explicit representation by social scientists [38].

52 Why earlier did I associate this ontological orientation, at least if taken up by economists, as a version of Veblen’s neoclassicism ? I do so for reasons provided by Ross himself in playfully accepting the label. Specifically, Ross recognises the inconsistency expressed in Veblen’s notion of neoclassical economics but embraces it. Ross refers to himself as a humble neoclassical in a piece entitled Neoclassicism forever. His argument, in essence, is that such is the degree of complexity of reality, and so limited are we in our ability to grasp it, that, despite his noted optimism, he suspects that we will never be able to do better than (be neoclassical in the sense of) making use of methods whose presuppositions do not wholly match the nature of social reality.

53

Lawson thinks that the neoclassical tension between modelling technology and ontology that he identifies is something we should seek to overcome and, indeed, could escape from immediately by adopting the philosophy of critical realism. I will argue by contrast that such tension is the social scientist’s fated condition [39].

54 Specifically, although Ross anticipates major advances in relevant forms of mathematics, he suspects that we will never achieve « a level of mathematical power that is fully adequate to the deep strangeness, to folk ontology, of the structure of the world [40] ». The explanation is that « increasingly powerful computational technology will deprive us, as a species, of the motivation and the capacity to keep seriously trying for such a limit [41] ». In consequence, all economists who recognise this situation, who maintain Ross’s vision of reality, will be left in the camp of the neoclassical :

55

All economists, therefore – or, at least, those who do not deny that the world is open and complex – will forever remain neoclassical up to a point. That is, the mathematics they use for expressing their understanding – including of the predictions churned out by econometric machines – will always fall somewhat short of the complexity they acknowledge [42].

3.3 – Alternative formulations of quantum theory

56 Clearly a core feature in the reasoning of Ladyman and Ross is an acceptance of a basic ontology of ‘real patterns’ amenable to analysis only through the employment of mathematics This is an ontology or ‘metaphysics’ that Ladyman and Ross associate with the findings of quantum physics. It is worth noting, then, that social positioning theory, which draws rather different implications for going forward, does, or so have argued [43], also sponsor social theories that qualify as forms of quantum theories. The differences in play here stem from contrasting assessments made of the features of quantum experimental and related findings that it is essential for a theory to cover or accommodate to qualify as quantum.

57 Elsewhere I advance two criteria for any theory to qualify as a quantum theory [44]. The first is that the subject matter consists of phenomena that are quantised, that take the form of elements that are not continuous but discrete. This is precisely the sort of outcome that processes of positioning achieve in the social domain [45]. The second is that any posited entities manifest patterns associated with one kind of phenomenon in certain interactions, and patterns associated with different kinds of phenomena in other interactions (as if the phenomena qua entities themselves do not really exist or at best are rendered determinate only in interactions). Thus, under (quantum) experimental conditions electrons manifest patterns consistent with their being particles under some forms of interaction and with their being waves under others.

58 I also argue, however, that this phenomenon, from the perspective of social positioning theory, is easily recognised as a feature of social reality too and is also straightforwardly handled from within that perspective [46]. In brief, with all human persons positioned in a multitude of ways, each gives rise to a ‘social individual’, anchored in the given human person, but with each simultaneously comprising very many different components, perhaps of numerous different communities. Not all of these can be activated simultaneously, however. Rather, the specific components that do manifest at any point will depend in large part on the sorts of interactions with others in which the social individual engages. Thus, if social individual X is many components at once including, say, a member some local community, a university lecturer, a partner, parent, a member of a political party, or sports club, etc., then, on walking down the street, X may well manifest as an ordinary community member at first, but on encountering, say, a student that this lecturer is supervising, thereupon engages in behaviour consistent with being a lecturer. If instead X bumps into a partner or offspring etc., then the sort of component and so component behaviour manifested will be different.

59 In short, X manifests forms of behaviour that change according to the other with whom X interacts. In this way, through maintaining a depth, and specifically a relational, ontology, the second accepted criterion of quantum theorising can be met without requiring that entities posited do not really exist or cannot pre-exist any interaction with which they are involved. On this account, in other words, that which is relativised to interactions is not the existence of individual entities but the form in which continually existing social individuals manifest [47].

60 The recognition of ontological depth is essential to the foregoing reasoning, and worth stressing. If instead a flat ontology were to be adopted for the social realm, i.e., one that lacks depth-features like relationality, then the sort of phenomena that I have associated with the second criterion would be unintelligible. The social positioning sponsored account relies fundamentally on the depth ontology for its explanation. I stress this because standard interpretations of quantum mechanical theorising, almost as an a priori principle, adopt an empiricist orientation and thereby do maintain an associated flat ontology. From this perspective the results of quantum physical experiments (that form the bases for the above noted second criterion) are certainly explanatorily unintelligible. Elsewhere [48], I speculate that a form of quantum field theory characterised by ontological depth may prove successful eventually in not merely accommodating quantum mechanical results but rendering them intelligible. Others of course, including various (quantum) physicists, notably Albert Einstein and David Bohm, have all along rejected the empiricist orientation of the standard interpretation, interpreting the lack of intelligibility of the quantum experimental results from its perspective as implying that this theory is at the very least incomplete.

61 But whatever the reason for this lack of intelligibility in this context of quantum mechanics specifically, in this domain it is nevertheless found that a form of mathematics (resting on unintelligible notions) proves to be predictively successful [49]; it is found to serve as an instrument of prediction. It is seemingly in large part because this is so that Ladyman and Ross are encouraged in the view that fundamental reality cannot be grasped by natural language, but only by forms of mathematics.