For the longest time, metaphysics was seen as a pseudo-science within scientific and philosophical circles, but ‘metaphysics made a spectacular comeback’ (Humphreys, 2013) following the failure of logical-empiricist and pragmatist attempts to do away with metaphysics.
There has been a new wave of philosophers since the late 20th century, who are focused on the sciences, yet distance themselves from the naive tropes of scientism and ‘post-metaphysical’ pretensions of pragmatism. Today we see a proliferation of metaphysical ideas closely linked to scientific theories and developments, often recognized under many new names like analytic metaphysics, scientific metaphysics, naturalist metaphysics and so on. It is worth exploring some of the important historical precedents leading to the resurgence of metaphysics within the philosophy of science.
The much-revered philosopher of empiricism and one of the greatest skeptics in modern philosophy, David Hume, was particularly known for his hostility towards metaphysical speculations and grand systems built on hypotheses. He suggested that while scientists have managed to overcome the temptations of metaphysical thinking, philosophers haven’t yet “purged themselves of their passion for hypotheses and systems” (A Treatise of Human Nature). He further opined that in order to make any serious progress in resolving the interminable disputes in philosophy, one must “reject every system, however subtle or ingenious, which is not founded on fact and observation” (An Enquiry Concerning Human Understanding).
However, many of these interminable philosophical conundrums still exist today, despite all the radical attempts by empiricists both in philosophy and natural sciences – for some of these problems, we have a different, refined nomenclature today (like the hard problem of consciousness in place of the classical mind-body problem) while other problems remain more or less the same (the problem of free-will, for example). As a reaction to this impasse, modern empiricists and pragmatists within the philosophy of science (starting with the logical empiricists) claimed that we have to give up on quests for ontological closure (concerns about understanding the fundamental/ultimate structures of reality) and distill complex questions into simpler, solvable problems that are amenable to scientific procedures of inquiry.
This has often morphed into forms of hostility towards philosophy as such or defeatism about the efficacy of philosophical inquiries – Stephen Hawking famously announced the ‘death of philosophy’, while philosophers in the analytic tradition had proclaimed the death of metaphysics (Bas Van Fraassen), the death of epistemology (Richard Rorty) and the like. However, there has been a recent revival of interest in philosophy and particularly metaphysics within the sciences. This was a result of the theoretical and empirical crises and obstacles to progress in many areas of science, as well as the realization that science always presupposes metaphysics and that “the role of philosophy is quite as much normative as descriptive—with everything, including science, coming within its critical purview” (Lowe, 1998).
The return of metaphysics to the mainstream is evident today in frontier areas of cosmology, foundations of physics, biology, cognitive sciences, etc., where speculations and hypotheses are buttressed by their respective theoretical edifices and empirical findings. This comeback is also accompanied by several polarizing tendencies, marked by defensive claims about conventions and importance of scientific orthodoxy (often associated with commitments that are defended using other tentative, often honorific, terms such as naturalism, physicalism, etc.) on the one hand and openly contrarian, heterodox systems of thought on the other (among which we can include recently popular and unorthodox metaphysical schemas like psycho-physical monism, process-ontology, property-dualism, panpsychism, etc.).
Let us look at the background to some of the meta-theoretical disputes concerning the philosophy of science today.
The famous Humean problem of induction is a major concern to much of philosophy of science. To put it simply, Hume argued that our knowledge of the world is limited to discrete impressions (via our sense-experience) of events in space and time, and that our accounts of general laws are derived from our observation of particular instances of ‘constant conjunction of events’ (eg., everytime I light a match, fire is produced). He denied that such connections between events could be the result of any metaphysical necessity. According to him, we cannot justify the necessity of causal relations, since our belief in causality itself is a habit of thought based on repeated observations of event-conjunctions.
The implication of this is that we have no real ground to assume that necessary connections extend into the future (that is, I have no solid ground to claim that the next time I light a match, fire will appear). However, within Hume himself we see a dislocation when he argues that scientists, when faced with exceptions to established generalizations search for “secret operation of contrary causes” (Treatise), thereby admitting implicitly that causality is not merely reducible to contingent event-regularities that we observe. So, if I light a match and fire is not produced, I can reasonably infer that there was something that hindered the production of fire in that particular action/instance (maybe the match or the box got wet, for example).
Kant’s response to this and the general spirit of skepticism was with the concept of synthetic a priori knowledge, which are necessary truths that inform our experiences but are not derived from experiences. Mathematical truths, principle of causality, space-time etc. are examples of synthetic a priori knowledge. The revolutionary implication in Kant was that our experiences do not conform to objects in the world, rather the objects in the world conform to our innate categories and forms of intuition (like space-time, causality, etc.).
Kant thus provided a robust framework for how knowledge is both possible and structured. This was supposed to have collapsed a Cartesian dualistic framework in which the external world was seen as distinct in kind (substantially different) from minds. However, even if we accept that mind is not substantially different from the physical world (monism), the question of whether the categories themselves are internally constructed through experience or whether they are derived from the real world still remains an open one, which is often associated with the realism vs irrealism debate (concerning categories of human thought). Such a dislocation is also visible within Kant between his Critique of Pure Reason and his Metaphysical Foundations of Natural Science.
Kant and Hume thus undermined traditional speculative metaphysics since we cannot talk of an external world independent of our internal constructions and categories thereof. Gradually, the entrenchment of both the traditions of pragmatism/instrumentalism (where the utility and testability of statements took precedence over its veridicality) and phenomenalism (wherein physical objects in the external world are reduced to bundles of sense-data, acquired through experience of phenomena) in philosophy and sciences followed, influencing figures like Ernst Mach to Bertrand Russell, A.J Ayer and others in the empiricist tradition.
During the early 20th century, around the time when the Newtonian paradigm was dethroned by Einstein’s revolutionary achievements, logical empiricism emerged under the auspices of a group of philosophers in Vienna (known as Vienna Circle), thus giving birth to philosophy of science as a distinct discipline. They tried to work out a robust criterion to demarcate genuine scientific propositions from unscientific ones (which they called “metaphysical”). This was the criterion of verifiability, which essentially stated that genuine scientific theories have to be testable against the empirical world.
The backbone to this mission was the philosophical positions of atomism (or reductionism) and phenomenalism (following Mach’s empiricism, who was a major inspiration for Einstein’s work), inspired by Wittgenstein’s work concerned with logical syntax and also made possible by Russellian logic. In simple words, their mission stated that our scientific knowledge of the world should be expressed in terms of sense-experience by the employment of elementary propositions. If no observation was relevant to the determination of the truth-value of such propositions, they were deemed unscientific and meaningless.
The verifiability criterion was unsatisfactory to Karl Popper, who was much impressed by Einstein’s upending of the Newtonian mechanics which had been the philosopher’s paradigm of knowledge for the longest time (for figures like Kant, Locke, Hume and others). Popper came up with the criterion of falsifiability according to which the mark of a genuine scientific theory is not whether it would be verified by a finite number of observations, but whether it will specify in advance under what conditions the theory itself can be falsified (something absent in theories like Marxist theory of history or Freudian Psychoanalysis).
Popper agreed with the Humean problem of induction, but disagreed with Humeans (like when he rejected Rudolph Carnap’s attempt to give law-like propositions/generalizations a probabilistic basis) by stating that science does not proceed through inductions, but through hypothetico-deductive procedures. That is, science starts with conjectures and hypotheses, which are then subjected to empirical tests and generalizations until it is refuted, upon which a new, bolder conjecture is proposed. Popper thus displaced the classical empiricist paradigm in philosophy by accepting fallibility as a core feature of scientific theorization and progress.
Popper’s successor Imre Lakatos famously questioned his theory by asking “under what conditions would you give up your demarcation criterion?” (Lakatos, 1970). Lakatos challenged the sufficiency of falsification, pointing out that scientists do not always throw away their theories upon refutation or contrary empirical evidence. They typically update some parts of the theory (auxiliary hypotheses) while preserving a core foundation (which he calls the negative heuristics) that cannot be subjected to refutation by any methodological standards (meaning, they are only abandoned eventually and not entirely refuted, like in the case of aether theories or vitalist theories). According to him, there are scientific theories that lag behind or are degenerating (which are unable to make predictions about novel facts) and those that are progressive (which make novel predictions).
Lakatos thus shifted the focus of philosophy of science to sequences of theories or research programmes, which became the subject of normative appraisal (Bhaskar, 1975) in rationally choosing between different available theories. In discussions today within the philosophy of science, the rational grounds for preferring a scientific theory over others are several (extending from practically-successful conventions to those of predictive power, simplicity, heuristic values, empirical adequacy, etc.) and yet, there is no clear consensus as to which of those supplementary strategies are to be applied and under what circumstances.
Along with the acknowledgment of the shaky grounds of demarcation criteria, there were also other controversies within philosophy of science. The problem of underdetermination, famously formulated by Duhem and later, Quine, was a well-known cause for controversy. It was pointed out that in many cases of testing scientific hypotheses, more than one theoretical account could adequately account for empirical data available at any given time, thereby introducing an uncertainty as to what theory to adopt when they are empirically equivalent with respect to experimental results.
A famous example is from the foundations of physics where theories like Bohmian mechanics, Many Worlds and GRW Theory all make the same predictions that are successful, and yet each of them propose radically different physical ontologies (Maudlin, 2017). Since the emergence of such impasses within science, the discourse within the philosophy of science increasingly began to shift towards normative concerns and has led to further combative and controversial stances regarding the problems associated with the scandal of underdetermination. What remains clear, however, is that the demarcation criteria which enjoyed a stable reception for a while has now collapsed within philosophy of science and things have started to look much more tentative, calling forth a new generation of philosophers who are struggling to clarify the entire gamut of this scandal using a plurality of conceptual and practical strategies.
Broadly speaking, there are two ways in which philosophy of science typically proceeds today: One, the Feyerabend way, which is to be a methodological anarchist adhering to his famous dictum “anything goes”. This is essentially a way of recognizing the radical dislocations and discontinuities within the history of scientific progress and embracing the chaos by unleashing metaphysical constraints and constructing tentative theories unfettered by the weight of orthodoxy and conventions. Two, to consistently work within orthodoxies and conforming to established procedures of scientific practice and methodological standards.
While the first way embraces truly radical change and methodological pluralism while remaining metaphysically ambivalent, the second way is more cautious, remaining rooted in conventions and upholding community principles associated with orthodox metaphysical commitments implicit in mainstream, hitherto-successful scientific theories. In any case, the flourishing of scientific enterprise and cultivation of critical dialogues within sciences demand a balance of both and a recognition of the areas in which they are mutually beneficial while also diagnosing and distinguishing between healthy and unhealthy forms of dogmatism.
An important task for philosophers is to taxonomize existing theories based on their core metaphysical commitments and to distinguish them from its auxiliary components (which are more provisional and subject to revisions). Another crucial task is to account for the explanatory potentials of each theory, by drawing out its implications as well as locating its relevance and historical situatedness, thereby recognizing the dialectical or ‘deeply social and cooperative nature of scientific inquiries’ (Massimi, 2022).
What works against the favor of scientists and philosophers alike, however, is premature and dogmatic skepticism towards heterodox research programmes and dismissive attitude towards metaphysics or the philosophy of science, particularly visible among science-communicators and popularizers. The primary hindrance to building a bridge between the ‘generalists’ and ‘specialists’ in both philosophy and sciences is the empiricist and pragmatic skepticism towards the re-emergence of metaphysics (particularly the speculative kind) and a denialism that characterizes substantive problems as mere semantic confusions that science had already or will soon overcome.
Against conventionalists, one must argue that science is not merely ‘a set of institutional error filters for the job of discovering the objective character of the world’ (Ladyman & Ross, 2007), and against the skeptics, one must also argue that the scientific process is not an irrational and erratic one entirely driven by intuitions, dogmas, and cultural biases of its time. Both the gatekeepers and radicals have their place within the philosophy of science, but the real task of metaphysics is, and has always been, about building coherent explanatory schemas with a proper place for speculations (as Whitehead had rightly noticed) that seeks to integrate the scientific and manifest images (to use Wilfred Sellars’ terminology) in order to conceptually furnish a unified account of the world.
