What kind of realist?

 

I've always felt that scientific realism is almost self-evidently true. Scientific theories and hypotheses put forward ideas that go beyond the evidence of direct experience. They postulate the existence of entities and forces that cannot be directly observed but whose effects can be teased out through the assumptions we have made about their characteristics. And when we have a theory that "succeeds" in explaining a domain of observation and experimentation, we have reason to believe that its hypothetical entities and forces actually exist. The existence of the hypothetical entities is the "best explanation" for the success of the theory or hypothesis.

This is not, of course, a deductively certain inference from the success of the theory to the reality of the unseen entities. There may be other explanations for the observational and experimental success of the theory. And the history of science in fact offers plenty of examples where this has turned out to be the case. Reality sometimes turns out to be more complicated, and structured differently, than our theories postulate.

This is the position that I would describe as "scientific realism". It represents a garden-variety ontology; it simply holds that the entities postulated by successful scientific theories are likely to exist in approximately the form they are postulated to possess.

There are coherent alternatives to scientific realism. Phenomenalism and instrumentalism are coherent interpretations of the success of scientific theories that do not postulate the real existence of unseen entities. Milton Friedman's instrumentalist treatment of economic theory is a case in point. However, instrumentalists have a hard time accounting for the success of scientific theories in the absence of a realist interpretation of the theoretical premises. Why should cloud chambers show the specific arcs and tracks that are predicted by theory if the underlying model of the mechanisms is not correct?

So how does all of this play out for the social sciences? In my view, the social sciences are substantially different from physics when it comes to hypothetical entities and theoretical hypotheses. The entities and forces to which we want to refer in the social world are not highly theoretical; rather, we can probe our concrete assumptions about these social entities and forces fairly directly. We don't need to turn to the Duhemian deductivism and theoretical holism that physics largely forces us into. Instead, we can devise strategies for probing them piecemeal.

So when we postulate that "class" is an important entity or structure in the modern world, our evidence for this claim is not largely based on inference to the best explanation and the overall success of class theory; it is instead the bundle of concrete researches that have been performed to identify, specify, and investigate the workings of class. Conceptual specification is more important that theoretical articulation and deduction: we need to know what a given researcher means to encompass in his or her use of the term "class structure". To take the photo above of Eton boys as an example -- what inferences can we draw about class from the photo? And what do we mean when we say that it illustrates an important social reality in the Britain of the 1930s, the reality of class? Is it a fact about attitudes; about the mechanisms of opportunity and selection; about the differential assignment of privilege; about modes of speech and thought?

My own philosophy of social science has several key features:

• I look at social science as inherently eclectic and pluralistic. There is no "best" method or "most fundamental" theory.
• I strongly suspect that social causation is fundamentally heterogeneous over multiple kinds of mechanisms and multiple temporalities. Outcomes are conjunctural, compositional, and contingent.
• I place a great deal of importance on empirical research and discovery. I am in that particular regard an enlightened "empiricist" about social and historical knowledge.
• I think there is an important place for theory and hypotheses in the social sciences. These need to be "theories of the middle range."
• I take an actor-centered approach to social theorizing. The substrate of the social world is individuals doing and thinking a range of things in various social settings.
• I am realist about a raft of social things: institutions, practices, value communities, social networks. All these social entities and structures exist as embodied in the thinking and acting of the socially constructed individuals who make them up, but they often have persistent and knowable properties that do not call for reduction to the micro level.
• I am realist about social causation, and I understand causation in terms of mechanisms.
• I am realist about the causal properties of at least some social entities -- structures, organizations, knowledge systems.
• I think ontology is important, but primarily at the level of the ontological assumptions implicated in various areas of scientific and historical research. Universal or philosophical ontology does not seem so important to me.

These commitments add up to a form of realism; but it isn't critical realism in the technical or substantive senses. It is a realism of a different stripe -- a pragmatic realism, a galilean realism, a scientific realism.

"Critical realism" is a term of art; it refers to a very specific bundle of philosophical and ontological ideas that have been developed by Roy Bhaskar and his followers. It makes substantive philosophical assumptions about how the social world works, and it depends resolutely on a philosophical method of discovery and justification.  And this means that the reasons we have for embracing realism of a more general kind do not necessarily extend to support for critical realism. One can be realist about the social world without accepting the assumptions and doctrines of critical realism. In fact, I suspect that the kind of realism I advocate here would be criticized as "empiricist" and "not truly realist" by the CR world.

There is much to admire in the literature of critical realism, both in the writings of Bhaskar and those who continue the research in this tradition. But it remains just one approach out of a spectrum of possible realist positions.

Kaidesoja's naturalistic social ontology

Tuukka Kaidesoja provides an important analysis and critique of Roy Bhaskar's philosophical method in his theory of critical realism in Naturalizing Critical Realist Social Ontology. This work provides a careful and detailed account of the content of Bhaskar's central ideas, as well as the relation those ideas have to other positions within and adjacent to critical realism. For Kaidesoja, the hope of discovering fundamental truths through transcendental reasoning is unpersuasive, and he advocates instead for a strategy of "naturalizing" the arguments for critical realism.

TK agrees with Bhaskar about the importance of ontological theory, and he thinks these topics are important for practitioners of the social sciences as well as philosophers. Here are some of the ways in which he characterizes the role of an ontological theory:

[Ontology is important] because specific research practices in social sciences as well as the theories and methods used in these practices always contain ontological assumptions and presuppositions no matter whether the practising social scientists and philosophers of social sciences acknowledge or discuss them. These assumptions and presuppositions concern, for example, the basic ontological categories of which the entities studied belong; the relationships between different kinds of entities studied; between them and those studied in the other social sciences and non-social sciences; and the causal structure of the social world (or the lack of such structure). In addition, ontological assumptions and presuppositions of this kind are not inconsequential in empirical research. Rather, they affect what are considered as proper social phenomena to be explained; what methods are thought to be suitable for studying different types of social phenomena; what are regarded as the sound explanations of these phenomena; and what are considered as possible factors in those explanations. Differences in opinion as to how to answer questions like these are reflected, for example, in the debates between the proponents of various forms of individualism (or microfoundationalism) and collectivism (or holism); and between the advocates of statistical causal modelling, the mechanism-based model of explanation and interpretative methods. (1-2)

So what is the ontology that Bhaskar articulates? According to Kaidesoja, it comes down to a fairly simple set of ideas:

The main ontological point in RTS then is that structures, or rather structured things (e.g. atoms, molecules, chemical substances and living organisms), possess causal powers by virtue of which they are able to generate empirically observable effects. (56)

Bhaskar describes the relationship between the structure of a thing and its power by using the concept of natural necessity. The essential structure of a thing both determines its causal powers -- or at least those powers that are explanatorily the most fundamental -- and constitutes its identity by fixing its membership in a natural kind. (57)

(These passages make clear the direct lineage from critical realism to causal powers theory.)

So how should we go about arriving at a defensible ontology for scientific knowledge? Bhaskar's answer is, through the philosophical strategy of transcendental argument. He wants to argue that certain ontological premises are the necessary precondition to the intelligibility of some aspect of the enterprise of science. Like Cruickshank, Kaidesoja attributes a philosophical apriorism to Bhaskar's theory of critical realism (5), and he holds that Bhaskar's method of argument is one grounded in apriori transcendental reasoning (82).

Kaidesoja argues against this aprioristic strategy and puts forward an alternative: "naturalized critical realist social ontology". Here is his preliminary description of this alternative:

In very rough terms, naturalists contend that theories in social ontology should be built by studying (1) the ontological assumptions and presuppositions of the epistemically successful practices of empirical social research (including well-confirmed theories produced in them); and (2) the well-established ontological assumptions advanced in other sciences, including natural sciences. This procedure is needed because naturalists hold that ontological theories cannot be justified by means of philosophical arguments that rely on a priori forms of conceptual analysis and reasoning. (2; italics mine)

So the heart of the approach that Kaidesoja advocates is the idea that the activity of formulating and evaluating scientific theories through empirical research is the only avenue we have for arriving at justified ideas about the world, including our most basic ontological beliefs. We might refer to this as a "boot-strapping" approach to ontology: we discover the more fundamental aspects of the world by constructing and evaluating scientific theories in various areas of phenomena, and then extracting the "ontological assumptions" these theories make.

This position makes a difference in the status of the resulting claims about ontology, according to Kaidesoja. Bhaskar wants to hold that the ontological claims established by transcendental arguments are different in kind from the claims about the physical or social world made by ordinary scientific theories (5). For Kaidesoja, by contrast, all ontological claims are on the same footing; they are part of the empirical scientific enterprise.

This means that all naturalist ontological theories should be understood as knowledge a posteriori which is always hypothetical, because, as will be later argued, there is no specifically philosophical or transcendental (as distinct from empirical) warrant for any philosophical ontology. (5)

Here is how Kaidesoja summarizes Bhaskar's typical transcendental argument:

In order to discuss them in detail, Bhaskar’s arguments in RTS can be analysed 

into the following steps:

1. X is 

a generally recognized 

natural 

scientific 

practice.
2. It is a 

necessary 

condition of 

the 

possibility (or 

intelligibility)

 of X 

that the world is 

P1,.

.

., Pn.
3. X 

is 

possible 

because 

it 

is 

real.
4. If 

the 

world 

were Q1, 

. 

. ., 

Qn, as is presupposed in competing philosophies of science, 

then X would 

be impossible 

or unintelligible.
5. Therefore, it is conditionally (i.e. given that X exists) necessary that the world is P1, 

.

.

., Pn. (88)

And here is the naturalistic argument form that Kaidesoja prefers:

1. X 

is 

an 

epistemically 
successful 

scientific 

practice 

described 

on 

the 

basis 

of empirical analysis of the practice. 
2. It is hypothetically (and in 
the explanatorily sense) a necessary condition of the epistemic successfulness of practice X under our description that the ontological structure of the world (or some of its aspects) really is as described in propositions P1, . . 

., 

Pn. 
3. Propositions P1, . . . , Pn are compatible with the ontological commitments of current 
scientific 
theories which have stood the test of critical evaluation by the relevant scientific community. 
4. The 
explicit ontological propositions or implicit ontological presuppositions of competing philosophical positions,  say Q1, 
. . . ,  Qn, are incompatible with propositions P1, . . . , Pn and the epistemic successfulness of X under our description remains impossible or unintelligible from the point of view of Q1, . . ., 

Qn. 
5. The best explanation of the epistemic successfulness of practice X under our description currently is that (a certain aspect or region of) the world is as described in propositions P1, 
. . . ,  
Pn. (98)

It seems to me that Kaidesoja's naturalistic alternative permits a very smooth respecification of the status and content of critical realism. Instead of arriving at conclusions that have philosophical certainty (philosophical transcendental ontology), we arrive at potentially the same conclusions based on reasoning to the best explanation. This was Richard Boyd's best argument for realism in the 1970s (what he called "methodological realism"), and it provides a philosophically modest way of giving rational credibility to the ontological conclusions critical realism wants to reach without presupposing the validity of philosophical transcendental arguments.

Since defenders of critical realism like Elder-Vass, Hartwig, and Groff have emphatically insisted that Bhaskar does not aspire to philosophical certainty with his scheme of argumentation, it may be that Kaidesoja's account will be understood as a clarification rather than an objection to the approach. The difference between the two argument forms here comes down to this: The naturalistic argument consistently replaces "reasoning derived from transcendental necessity" by "reasoning within the general framework of what we know about the world", but leaves the deductive flow of the argument unchanged. And this might be a reasonable way of accounting for the defenders' view that Bhaskar's philosophy has been fundamentally fallibilistic all along.

Reply to Elder-Vass, Hartwig, and Groff on critical realism

Critical realism is a hot topic now in sociological theory and philosophy of social science. It turns out that there are some pretty strong disagreements about the foundations of the theory. Recent posts here have highlighted my own (admittedly non-expert) reading of Bhaskar’s assumptions about ontology (link), my discussion of the limited and friendly critique of Bhaskar’s assumptions offered by Justin Cruickshank (link, link), and a preliminary view of the “naturalized critical realism” advocated by Tuukka Kaidesoja (link). (There is more to come on Kaidesoja’s work.) These posts — particularly those highlighting Cruickshank — have elicited strong rebuttals from Ruth Groff, Dave Elder-Vass, and Mervyn Hartwig (link, link). Here I would like to respond to some of the views advanced in the rebuttals by these experts from within critical realism.

Elder-Vass and Hartwig reject the core claims that I have attributed to Cruickshank in his critique of Bhaskar's philosophical method: that Bhaskar pursues an aprioristic philosophical method in arriving at the fundamental ideas of critical realism, and that he regards these ideas as having been established with  some kind of certainty by this method. (I should make it clear, of course, that this is my interpretation of Cruickshank; I hope I have not mis-represented him.) Against this aprioristic and infallibilist reading, Elder-Vass and Hartwig argue that Bhaskar's reasoning is not aprioristic and that he regards his conclusions as being fallible and historically conditioned.

Elder-Vass believes there are ample places in Bhaskar's work where he asserts the fallibilism of his conclusions. But the particular passage that E-V quotes from Scientific Realism and Human Emancipation seems to prove less than E-V supposes. Moreover, it seems to detract from intellectual virtues that Bhaskar himself wanted to assert: that there are good philosophical (i.e. non-empirical) reasons for accepting certain ontological statements. Does Bhaskar attribute rational credibility to philosophical arguments in arriving at substantive claims about the world? Unmistakably he does; his whole method is philosophical! And he seems to have quite a bit of confidence in the conclusions that he reaches when it comes to the fundamentals of ontology. Or in other words: he assigns a high level of justificatory weight to the philosophical arguments he offers for specific conclusions about ontology.

In fact, general statements about the fallibility of human knowledge don't help very much with the problem Cruickshank is raising. How fallible and for what reasons? For example, if the claims of critical-realist ontology are only "as fallible as" the claims of mathematics and logic, that is indeed to attribute a high degree of certainty to those ontological claims. On the other hand, if they are "as fallible as" statements about the virtues of the gods, then they are highly fallible indeed. So the general statement "all assertions are fallible" is too general to help very much. We want to know what the conditions of knowledge are for different kinds of assertions, and how confident we can be, give available reasons and evidence, that the given assertion is true. "Wood is made mostly of carbon and water," "electrons have negative charge of 1.6 * 10^-19 coulombs," "physical objects are located in three-dimensional space," and "a triangle encloses 180 degrees" are all statements that are in some sense fallible; but the ways in which they might go wrong are quite different from one to the next. Some are more empirical, some more theoretical, and some are metaphysical or mathematical. And the kind of justification or proof that is given for each is different. As a non-committed reader of Bhaskar, it does appear to me that Bhaskar relies on abstract philosophical arguments to reach ontological conclusions, and that he attributes a fairly high degree of confidence to those lines of reasoning.

So how fallible does Bhaskar think his theory of ontology is, and for what reasons, according to E-V and Hartwig? Does Bhaskar believe, for example, that perhaps experimentation could after all be coherently understood against a background of Humean regularity assumptions? Plainly not; that is the whole point of his argument, to rule out that possibility. And he seeks to rule it out by offering philosophical arguments to establish the point. To take a fairly random example from RTS:

However if deducibility is the only criterion for explanation and the source of the surplus-element is its explanation there will be an infinite number of surplus-elements for any statement. Hence any statement can be said to be law-like on an infinite number of grounds. Deducibility alone cannot explicate the distinction between necessary and accidental or nomic and non-nomic universals. (kl 3018)

This is plainly a purely philosophical (logical) argument; it is reductio ad adsurdum. And Bhaskar plainly believes it presents an insurmountable barrier to the Humean; or in other words, it establishes the necessity of the anti-Humean position on this particular point. So the idea that Bhaskar applies a warning label at various points (“knowledge is fallible”) doesn’t resolve the issue of whether he attributes too much weight to the power of philosophical arguments to resolve ontological issues.

Hartwig provides useful clarification by summarizing the logic of a transcendental argument. The argument form itself is deductively valid and trivial, essentially modus ponens.  So we can be completely certain that if the premises are true then the conclusion is true. That is not where the philosophy comes in. Rather, the heavy lifting for the transcendental argument is in establishing the major premise. What kind of argument is needed in order to establish an "only-if" statement? Take the Kantian version: [only if the world is spatio-temporally-causally structured] then [empirical experience is possible]. We can offer strong philosophical reasons for believing that empirical experience is possible. But how do we get the "only-if" assertion? How do we know that there is no other form of structure that could give unity to empirical experience? How do we know that a degraded spatio-temporal-causal ordering would not nonetheless admit of empirical experience? (Things sometimes result from anomaly and show up discontinuously in unexpected places; how do we know that such a slightly disorderly world could not support empirical experience?) In other words, why should we have confidence in Kant's (or Bhaskar's) assertion of the major premise: [only if X] then Y?

In fact, Strawson's critique of Kant's argument in The Bounds of Sense is precisely that Kant errs in maintaining that spatiotemporal order is necessary for the possibility of empirical experience; he constructs a hypothetical world in which experience is ordered auditorially but not spatially ordered and argues that this is a perfectly coherent basis for ordinary empirical experience.

And this is where the Cruickshank-like argument comes in strongly: Bhaskar’s arguments for the “only-if” statements upon which critical realism depends are: interesting, skillful, determined — and far short of deductively or rationally conclusive.

If Bhaskar is thought to embrace fallibilism to this extent: that his whole construction of the ontological prerequisites of experimentation may be in error; then indeed he is a fallibilist theorist. Ruth Groff indicates that in her opinion this is a possibility: "Bhaskar may or may not be correct, either about what the implicit ontology of the activity of experimentation is, or about whether or not it is consistent with the explicit ontology of Humeanism and Kantianism." But nothing in RTS makes me think that Bhaskar believes this particular form of corrigibility. E-V raises that possibility above ("What is necessary is that IF science occurs THEN the world must be such that science is possible and/or intelligible"). But this is virtually vacuous; it only becomes an ontological statement when one gives arguments about how the world must be. E-V, Hartwig, and Groff are the experts; but when I pick up the thread of A Realist Theory of Science at almost any point, I find Bhaskar making very confident statements about how the world must be, based on the philosophical arguments that he constructs.

Groff seems to slide over the place where some would say that Bhaskar does in fact over-reach philosophically: the complicated reasoning he provides to go from "we acknowledge the overall rationality of the enterprise of science" to "the world must have certain fairly abstract attributes". We don't have to say that "science is irrational" or "experimentation is unintelligible" in order to question Bhaskar's conclusions about ontology; rather, we can question the sequence of inferences he makes from the one "fact" to the other. These inferential steps take place within a philosophical argument, and they are questionable.

This shouldn't be thought to imply that I (or Cruickshank or Kaidesoja, for that matter) doubt that philosophical arguments have any justificatory or clarificatory weight; philosophy is simply careful reasoning and clear analytical thinking, and of course good philosophy can help illuminate how science works. What I do think some of us want to maintain is pretty much what Kant held as well: we can't derive substantive conclusions about the structure of the real world from purely philosophical reasoning. There are no rabbits in that hat!

So it still seems to me -- and now it's me speaking, not Cruickshank -- that Bhaskar relies too heavily and confidently on philosophical methods to arrive at ontological conclusions. Perhaps it is true, as E-V and Hartwig assert, that he also duct-tapes onto his construction some warnings about the overall fallibility of all human knowledge. But I'm still not seeing that this corrigibility extends very deeply when he is actually trying to reach conclusions about ontology. And yet this is precisely where the corrigibility/fallibility warning is most needed: the philosophical arguments offered for the “only-if” statements (the heart and substance of critical realism) fall far short of any kind of certainty. They are suggestive, but they are not rationally compelling. And Bhaskar does not appear to highlight this fact.

In short, Bhaskar does appear to believe that we can arrive at philosophically compelling conclusions about ontology; and those conclusions are drawn through recourse to philosophical arguments. And this does seem to distinguish his general theory of knowledge from coherence theorists (Goodman and Quine) and naturalists (Kaidesoja), who believe that ultimately there is only one kind of knowledge: scientific knowledge at various levels of abstraction.

But it also seems to me that this debate is in some ways missing the most important point: how good is critical realism as a meta-theory of the situation of material human beings acquiring knowledge of the world?  Putting aside the question of whether philosophical theory can shed light by itself on the structure of the world, what should we actually think about the latter topic? Is realism a good way of thinking about the knowledge enterprise? Is the kind of back-and-forth that Bhaskar is so good at, from existing scientific practice to apparent presuppositions about how things work, a good way of leveraging some new thinking about the way the world works? The most interesting thing about critical realism is surely not its philosophical method; it is the set of ideas it brings forward about how science and knowledge progress in giving material human beings a better notion of how the world works. Philosophy is a part of that process, but only a part. And the realist ontology is an important construction no matter what its argumentative origins are.

Is ontology an apriori field of knowledge?

 

The critical realists -- Roy Bhaskar in particular -- attach a great deal of importance to the question of ontology. A theory of ontology should describe the kinds of things, relations, and forces that exist in a realm. So the pre-Socratic philosophers were engaging in ontological theorizing when they asked the question, what does matter consist of -- atoms or a plenum?

The question I am raising here is one of philosophical methodology: what kind of epistemic basis is available for formulating and defending a theory of ontology? How can we claim to know various truths about the nature of reality?

There seem to be three possibilities.

• Apriori philosophical argument: derive conclusions about the necessary structure of the world from apriori philosophical principles. This is traditional metaphysics, and few philosophers would advocate for it today. (foundationalist theory)
• Transcendental philosophical argument: arrive at conclusions about what the world must consist of, in order to make sense of our cognitive abilities. This is Kantian metaphysics, which attempts to do without foundational assumptions and to derive conclusions from the prerequisites of epistemic achievements we are known to have. (internalist theory)
• Generalized empirical theorizing: all substantive representations of the world are hypothetical, justified by the contribution they make to our ability to formulate good, empirically supported scientific theories. This is the approach taken by naturalistic philosophers, who maintain that there are no apriori truths and the only vehicle we have for discovering the nature of the world is through scientific imagination and observation. (coherence theory)

Ontology appears to be about the world; but equally it might be considered to be about a set of particularly fundamental concepts and conceptual structures.  The question, "What does the world consist of?" can also be presented as the question, "What concepts serve best to represent the hypothetical structure of the world underlying observations?" Concepts are the intellectual tools or schemes through which we analyze the world; and if they refer to unobservable entities, they are unavoidably hypothetical constructs. As "knowing beings", it has been necessary for human beings to use their imaginations to come up with concepts in terms of which to analyze the world. Some conceptual systems are defective because they lead to expectations about the world that are not born out; other systems are more complex than necessary; yet others postulate entities or processes that we may have reason to want to avoid: magical forces, divine intervention, action-at-a-distance. And when we arrive at a conceptual scheme that appears to serve well as a durable basis for a range of scientific theories, we may want to conclude that the world actually has the properties attributed to it by the scheme.

Nelson Goodman takes a fairly radical view on this question in Ways of Worldmaking. He takes the example of two apparently inconsistent statements about the world: "The sun always moves" and "The sun never moves." And he points out that the statements must be framed within one or another frame of reference; they are not absolutely true or false, but rather true or false with respect to a frame.

Frames of reference, though, belong less to what is described than to systems of description; and each of the two statements relates what is described to such a system. If I ask about the world, you can offer to tell me how it is under one or more frames of reference; but if I insist that you tell me how it is apart from all frames, what can you say? We are confined to ways of describing whatever is described. Our universe, so to speak, consists of these ways rather than of a world or of worlds. (58)

Here is the conclusion that Goodman reaches that is most relevant to the topic of realism:

Many different world-versions are of independent interest and importance, without any requirement or presumption of reducibility to a single base. The pluralist, far from being anti-scientific, accepts the sciences at full value. His typical adversary is the monopolistic materialist or physicalist who maintains that one system, physics, is preeminent and all-inclusive, such that every other version must eventually be reduced to it or rejected as false or meaningless. If all right versions could somehow be reduced to one and only one, that one might with some semblance of plausibility be regarded as the only truth about the only world. But the evidence for such reducibility is negligible, and even the claim is nebulous since physics itself is fragmentary and unstable and the kind and consequences of reduction envisaged are vague. (59-60)

The philosophical position I am invoking here is also a key part of W.V.O. Quine's approach to empirical knowledge in Word and Object. His phrase, the "web of belief", captures the idea well. All real knowledge falls within that web, and it is held together only by observation (when statements have implications for outcomes that can be observed) and logic. The premises of quantum mechanics are some distance from the observational and experimental sentences that can be examined in the lab; and the premises of metaphysical theory are even more distant. But they are all dependent on the same kinds of requirements: simplicity, coherence, and (when possible), empirical observation. Quine referred to "Neurath's boat" as a way of describing the state of our knowledge of the world -- from the observable properties of coal to the fundamentals of time and space:

Neurath has likened science to a boat which, if we are to rebuild it, we must rebuild plank by plank while staying afloat in it. The philosopher and the scientist are in the same boat. If we improve our understanding of ordinary talk of physical things, it will not be by reducing that talk to a more familiar idiom; there is none. It will be by clarifying the connections, causal or otherwise, between ordinary talk of physical things and various further matters which in turn we grasp with help of ordinary talk of physical things. (3)

...

Analyze theory-building how we will, we all must start in the middle. Our conceptual firsts are middle-sized, middle-distanced objects, and our introduction to them and to everything comes midway in the cultural evolution of the race.... We cannot strip away the conceptual trappings sentence by sentence and leave a description of the objective world; but we can investigate the world, and man as a part of it, and thus find out what cues he could have of what goes on around him. (4-5)

(Quine's participation in the Boolos panel above is a very good exposure to some of his thinking about meaning and concepts.)

It is perhaps surprising to invoke Goodman and Quine in the context of reflections on critical realism, since their philosophies are anti-realistic (or at least agnostic between realism and anti-realism), and the logical-positivist background of much their thinking is anathema to the critical realists. Moreover, both lend support to a certain kind of conceptual relativism: Quine through his arguments about the indeterminacy of translation and ontological relativity (Ontological Relativity), and Goodman through his view of "many worlds" in Ways of Worldmaking. This perspective doesn't necessarily commit one to anti-realism; in fact, Hilary Putnam's effort to create a defensible formulation of "internal realism" indicates one possible direction of argument towards realism from these premises. (Maria Baghramian's discussion in "From Realism Back to Realism" of Putnam's various positions on realism is very good; link.) But it is difficult to see how one could arrive at a strong philosophical realism within these epistemic constraints.

However, if these arguments on the limits of metaphysical reasoning are valid, then we need to acknowledge these limits and move forward. Fundamentally, the core of their position seems unassailable: there is no epistemic foundation possible outside the loose constraints of empirical observation and logic that can justify a set of beliefs about the fundamental structure of the world. There is no secret recipe for arriving at metaphysical knowledge through purely philosophical pathways.

The statement of realism in which I have the greatest confidence is this: we are justified in acknowledging the reality in the world of the things, processes, structures, and forces that are postulated or implied by the best scientific theories we have to date. And we acknowledge that these beliefs, like all scientific and empirical beliefs, are fallible and correctable.

(An upcoming post will discuss Tuukka Kaidesoja's very interesting critique of critical realism and his advocacy of "naturalized critical realism" in Naturalizing Critical Realist Social Ontology.)

Cruickshank's central critique

Justin Cruickshank is a friendly critic to critical realism, not a hostile one. He criticizes the philosophical method but supports many of the substantive conclusions about realism. (Here is a prior post on the gist of Cruickshank's criticisms.) Cruickshank provides a useful analysis of critical realism and its anti-matter double, social constructivism, in “Positive and Negative,” a working paper at the International Migration Institute in 2011 (link).

It seems that Cruickshank's most basic concern in "Positive and Negative" is that CR posits a particular ontology as the unique precondition of all science (a move familiar from transcendental metaphysics), and that philosophical reasoning can tell us what that ontology involves. So Bhaskar places excessive reliance on an apriorimethod of philosophical reasoning in constructing his metaphysical ideas. By contrast, in "Positive and Negative" Cruickshank favors a more malleable and fallibilist stance on ontology.  Here is one fairly clear statement of his view of the central shortcoming of CR:

Finally it is suggested that the correct way to construe the post-positivist problem-situation is to focus on the fallibility of knowledge, as critical realism does and, unlike critical realism, argue that fallible knowledge claims should be revised and replaced through criticism, with the focus being on theories’ ability to solve explanatory problems rather than their adherence to a set of ontological assumptions that are posited as the condition of possibility of the social sciences. (4)

What is implied here is the view that all knowledge — both ordinary scientific knowledge and ontological knowledge — is fallible and revisable. This means that ontology should not be treated as part of a priori philosophy but rather as the more abstract end of the spectrum of scientific theorizing about the world. Bhaskar errs, then, in asserting that ontological knowledge is different in kind from ordinary scientific knowledge; it is transcendental knowledge — knowledge based on rigorous analysis of the necessary preconditions of ordinary scientific knowledge.

This is a serious criticism of the method that Bhaskar uses in developing his theory of critical realism; but it does not imply disagreement with the key substantive conclusions of Bhaskar’s developed theory. Cruickshank remains a realist, and leaves it open that we can coherently maintain the key substantive ideas of critical realism: Ontology is important, bad assumptions about ontology can lead to bad scientific theory, and we can indeed regard the statements of an ontological theory as being referential to the “real” structure of the world. (In Bhaskar’s terminology, we can look at ontology as being and account of non-transitive knowledge.) Here is the brief summary that Cruickshank offers in his introduction to CRITICAL REALISM: THE DIFFERENCE THAT IT MAKES:

Critical realism is realist because it holds, contra postmodernism and social constructionism, that research is about gaining knowledge of a reality that exists independently of our representations of it.... Critical realism is critical, as regards methodology, because it holds that the concepts which inform the meta-theory that defines structure and agency can only be developed via a critical dialogue with alternative social ontologies. (kl 568)

Cruickshank situates critical realism and its opposite, social constructivism, in terms of their different efforts to reject "positivism”. So positivism is the foil against which critical realism is unfolded. (Ruth Groff makes a slightly different choice in her choice of Humean causation theory is the foil. Humeanism and positivism have various similarities, but they are not identical doctrines.) According to Cruickshank, Bhaskar rejects positivism for several reasons; but its commitment to the idea of unified deductive theories underlying the full range of empirical observations is at the top of the list. He characterizes this approach to the relation of science to the world as one involving a “closed systems ontology” (7); whereas he believes an ontology that looks at the world as a “stratified open system” is preferable. The “stratified” part of the concept refers to the idea that causal mechanisms are “emergent” from the lower-level things of which they are composed; and the “open” part of the concept refers to the idea that there are substantial dimensions of contingency involved in any complicated social process. “The ontology [of a stratified open system] is held to be one of open systems because the underlying causal laws interact in contingent ways to produce change at the level of observable events” (7).

Cruickshank believes that this distinction between open and closed systems leads eventually to Margaret Archer’s concept of “morphogenesis” (8); link.

JC also offers an answer to a question posed in an earlier post: what is “critical” about critical realism (link)? Here is the heart of his answer:

For those critical realists who regard critical realism as a form of neo-Marxism, the task of social science is not just that of explaining how structures and agents interact but also that of criticism (see Bhaskar 1998 and Collier 1998). Their argument runs thus. Any scientific account of how the capitalist structure works will show how it is oppressive and exploitative. It will also show how this structure needs to generate ideological beliefs to mask its nefarious character. Ideological beliefs here are defined as beliefs which are not only false but caused by a structural need for obfuscation and which serve the interests of the capitalist class by obfuscating oppression and inequality.  (10)

So “critical” here has two meanings: exposing of undesirable characteristics of some social entities and exposing how certain formulations of ordinary belief have the effect of concealing those characteristics.

(Incidentally, Cruickshank explains one of Bhaskar’s most basic ideas, the notion of an “intransitive domain of reality”, in terms that suggest the terminology itself is misleading. Theory is transitive because “fallible theories are open to change” (8). This sounds like a definition of “transitory” or "changing" rather than “transitive” ("allowing inference from one proposition to another: if A is longer than B and B is longer than C then A is longer than C"). Here is the place where Bhaskar introduces the distinction in A Realist Theory of Science:

Any adequate philosophy of science must find a way of grappling with this central paradox of science: that men in their social activity produce knowledge which is a social product much like any other, which is no more independent of its production and the men who produce it than motor cars, armchairs or books, which has its own craftsmen, technicians, publicists, standards and skills and which is no less subject to change than any other commodity. This is one side of 'knowledge'. The other is that knowledge is 'of' things which are not produced by men at all: the specific gravity of mercury, the process of electrolysis, the mechanism of light propagation. None of these 'objects of knowledge' depend upon human activity. If men ceased to exist sound would continue to travel and heavy bodies fall to the earth in exactly the same way, though ex hypothesi there would be no-one to know it. Let us call these, in an unavoidable technical neologism, the intransitive objects of knowledge. The transitive objects of knowledge are Aristotelian material causes. They are the raw materials of science—the artificial objects fashioned into items of knowledge by the science of the day. They include the antecedently established facts and theories, paradigms and models, methods and techniques of inquiry available to a particular scientific school or worker. The material cause, in this sense, of Darwin's theory of natural selection consisted of the ingredients out of which he fashioned his theory. Among these were the facts of natural variation, the theory of domestic selection and Malthus' theory of population. Darwin worked these into a knowledge of a process, too slow and complex to be perceived, which had been going on for millions of years before him. But he could not, at least if his theory is correct, have produced the process he described, the intransitive object of the knowledge he had produced: the mechanism of natural selection.... In short, the intransitive objects of knowledge are in general invariant to our knowledge of them: they are the real things and structures, mechanisms and processes, events and possibilities of the world; and for the most part they are quite independent of us. (Kindle Locations 749-764) (italics mine)

It would appear that Cruickshank's interpretation is consistent with these remarks by Bhaskar.)

Cruickshank also edited a useful volume, CRITICAL REALISM: THE DIFFERENCE THAT IT MAKES, to which he contributed a useful introduction and an applied article, "Underlabouring and unemployment: Notes for developing a critical realist approach to the agency of the chronically unemployed". Another very useful collection edited by Margaret Archer, Roy Bhaskar, Andrew Collier, Tony Lawson, and Alan Norrie, Critical Realism: Essential Readings (1998), provides an excellent selection of readings on critical realism.

 

Bhaskar's core ideas

Critical realism has become an important topic within sociological theory, and several prior (and upcoming) posts have addressed the theory. As a point of reference for this ongoing discussion, consider a few key statements by Roy Bhaskar about transcendental [critical] realism in A Realist Theory of Science. Here is a simple and clear definition of Bhaskar's theory of realism:

The third position, which is advanced here, may be characterized as transcendental realism. It regards the objects of knowledge as the structures and mechanisms that generate phenomena; and the knowledge as produced in the social activity of science. These objects are neither phenomena (empiricism) nor human constructs imposed upon the phenomena (idealism), but real structures which endure and operate independently of our knowledge, our experience and the conditions which allow us access to them. Against empiricism, the objects of knowledge are structures, not events; against idealism, they are intransitive (in the sense defined). (p. 15)

And here is Bhaskar's statement of how he views the cognitive status of the theory of transcendental realism:

It is not necessary that science occurs. But given that it does, it is necessary that the world is a certain way. It is contingent that the world is such that science is possible. And, given that it is possible, it is contingent upon the satisfaction of certain social conditions that science in fact occurs. But given that science does or could occur, the world must be a certain way. Thus, the transcendental realist asserts, that the world is structured and differentiated can be established by philosophical argument; though the particular structures it contains and the ways in which it is differentiated are matters for substantive scientific investigation. (p. 19)

This passage makes it clear that Bhaskar believes the statements of ontology are philosophical statements, and they are established with a kind of necessity that differentiates them from ordinary empirical statements. This indicates Bhaskar's adherence to a philosophical method of discovery, inquiry, and justification.

Here is an example of Bhaskar's transcendental reasoning, applied to the analysis of experimentation.

The intelligibility of experimental activity presupposes not just the intransitivity but the structured character of the objects investigated under experimental conditions. Let me once again focus on the empiricist's favourite case, viz. causal laws, leaving aside for the moment such other objects of investigation as structures and atomic constitutions. A causal law is analysed in empiricist ontology as a constant conjunction of events perceived (or perceptions). Now an experiment is necessary precisely to the extent that the pattern of events forthcoming under experimental conditions would not be forthcoming without it. Thus in an experiment we are a causal agent of the sequence of events, but not of the causal law which the sequence of events, because it has been produced under experimental conditions, enables us to identify. (p. 23) (italics mine)

Essentially Bhaskar is making a classic Kantian move here: he is arguing that we cannot make intellectual sense of a scientist's use of experiment without presupposing that there are underlying objects and causal laws governing them which are the subject of the experiment. The phrase in italics identifies the necessary presupposition of the experiment: the presence of objective, theory-independent causal laws governing the objects of the experiment. And, as the subsequent sentence in the text makes clear, the causal laws in question are of a different ontological order than the events that manifest them. Here is how Derk Pereboom summarizes Kant's transcendental argument against Hume in his contribution to the Stanford Encyclopedia of Philosophy (link):

In the Metaphysical Deduction (A66–83, B92–116) Kant intends to derive the categories [including causation] from the specific modes or forms of any human thought about the world, the logical forms of judgment. The Metaphysical Deduction has an essential role to play in the Transcendental Deduction, and we will discuss this argument at an appropriate juncture (when we reach §19 of the B-Deduction).

It is evident that Bhaskar's style of argument here parallels that of Kant. However, Kant's transcendental method is not in fact satisfactory. We have the example of non-Euclidean geometries to provide a reminder that Kant's reasoning fails. Kant used the same kind of transcendental argument from the possibility of experience to arrive at the conclusion that space is necessarily Euclidean, but the discoveries of the consistency of non-Euclidean geometry (and the physical geometry of general relativity theory) show that this conclusion is incorrect. This example reminds us that transcendental reasoning is not truth-preserving; we can proceed from a transcendental argument to a false conclusion.

Now back to Bhaskar and the transcendental conclusion that he draws from the argument concerning experimentation:

The intelligibility of experimental activity presupposes then the intransitive and structured character of the objects of scientific knowledge, at least in so far as these are causal laws. And this presupposes in turn the possibility of a non-human world, i.e. causal laws without invariances and experiences, and in particular of a non-empirical world, i.e. causal laws and events without experiences; and the possibility of open systems, i.e. causal laws out of phase with patterns of events and experiences, and more generally of epistemically insignificant experiences, i.e. experiences out of phase with events and/or causal laws. In saying that the objects of scientific discovery and investigation are 'intransitive' I mean to indicate therefore that they exist independently of all human activity; and in saying that they are 'structured' that they are distinct from the patterns of events that occur. The causal laws of nature are not empirical statements, i.e. statements about experiences; nor are they statements about events; nor are they synthetic a priori statements. (pp. 25-26)

So here Bhaskar pulls the rabbit from the hat: he argues that we can conclude that we must presuppose intransitive and structured objects subject to causal laws if we are to make sense of the intelligibility of experimentation. Here he repeats the finding:

In §3 I argued that only if causal laws are not the patterns of events that enable us to identify them can the intelligibility of experimental activity be sustained. But causal laws are, or have seemed to philosophers to be, pretty mysterious entities. What can it mean to say that they have a real basis independent of events? The answer to this question will be seen to necessitate the development of a non-anthropocentric ontology of structures, generative mechanisms and active things. (p. 35)

So philosophy allows us to conclude something substantive about metaphysics, according to Bhaskar: (if science exists) that there are real independent causal laws. Science does in fact exist; therefore there are real independent causal laws.

Finally consider Bhaskar's notion of things and powers:

The world consists of things, not events. Most things are complex objects, in virtue of which they possess an ensemble of tendencies, liabilities and powers. It is by reference to the exercise of their tendencies, liabilities and powers that the phenomena of the world are explained. Such continuing activity is in turn referred back for explanation to the essential nature of things. On this conception of science it is concerned essentially with what kinds of things they are and with what they tend to do; it is only derivatively concerned with predicting what is actually going to happen. It is only rarely, and normally under conditions which are artificially produced and controlled, that scientists can do the latter. And, when they do, its significance lies precisely in the light that it casts on the enduring natures and ways of acting of independently existing and transfactually active things. (p. 41)

So things (objects) possess powers, and we explain the behavior of objects (and ensembles) as a consequence of the operation of their powers. And powers and causal laws are linked; powers generate laws:

There is nothing esoteric or mysterious about the concept of the generative mechanisms of nature, which provide the real basis of causal laws. For a generative mechanism is nothing other than a way of acting of a thing. It endures, and under appropriate circumstances is exercised, as long as the properties that account for it persist. Laws then are neither empirical statements (statements about experiences) nor statements about events. Rather they are statements about the ways of acting of independently existing and transfactually active things. (pp. 41-42)

These statements and assumptions by Bhaskar illustrate a fairly clear philosophical methodology. It is a method that derives from Kant's transcendental metaphysics. And Bhaskar seems to be confident in arriving at definite and assertoric conclusions based on this method. Ontology is not an empirical discipline, according to Bhaskar; instead, it is a philosophical reflection on the preconditions of science, and it is grounded in philosophical arguments rather than empirical, scientific, or experimental arguments.

This implies that Bhaskar adheres to the idea that there are at least two kinds of knowledge that we can be interested in -- philosophical and empirical-scientific. He therefore plainly rejects the coherentist and general scientific view (espoused by W.V.O. Quine and the pragmatists) that all defensible beliefs are eventually empirical, whether more directly connected to experience or more distantly so.

This feature of Bhaskar's method lays him open to the kind of criticism that is offered by Justin Cruickshank and others.