Introduction

The scientific realism debate is one of the main debates in the philosophy of science. The issue is old, and sometimes it seems difficult even to tell how to distinguish between realism and antirealism.

In a relatively new area of debate, there have been attempts to resolve this stalemate by analyzing whether the positions are committed to different futures of science and whether they make different recommendations regarding which scientific research programs to pursue. The debate centers around whether scientific realism is more conservative in its position towards science than antirealism. The question is whether realism implies that we should not expect radical changes in the future of science and whether it encourages us to dismiss radical alternatives to current best scientific theories.

However, it seems that the debate has reached another stalemate. argued that realism and antirealism can lead us to expect and value the same futures. If this is really the case, then there does not appear to be a significant difference between realism and antirealism, at least in terms of the future.

In this post, I argue that, while it is true that realism and antirealism can commit to and desire the same possible futures, this does not mean that the positions cannot be distinguished from each other by using future-oriented considerations. I show that the positions differ in what they tell us about the possibility, desirability, and trackability of scenarios of the future.

What Is Scientific Realism?

Scientific realism can be characterized in three theses:

The Metaphysical Thesis: The world has a definite and mind-independent structure.

The Semantic Thesis: Scientific theories should be taken at face value. They are truth-conditioned descriptions of their intended domain, both observable and unobservable. Hence, they are capable of being true or false. The theoretical terms featuring in theories have putative factual reference. So, if scientific theories are true, the unobservable entities they posit populate the world.

The Epistemic Thesis: Mature and predictively successful scientific theories are well confirmed and approximately true of the world. So, the entities posited by them, or, at any rate, entities very similar to those posited, inhabit the world. (Psillos 1999, 4).

The main line of the debate, as it currently stands, concerns the epistemic thesis. The issue is whether it is correct to accept scientific theories as approximately true descriptions of the mind-independent world. Everything comes down to the question of whether the arguments for the epistemic thesis are stronger than the arguments against it. As we will see, these arguments and their assessment are closely tied to the assessment of actual scientific work and the historical development of science. The arguments and their assessment are nuanced, and both the supporters and critics of scientific realism have put serious effort into using detailed historical arguments and abstract conceptual reasoning to make their case. The recent form of the debate on which we focus in this post is, therefore, remarkable for its focus on the future rather than the past.

The central argument for scientific realism is the no miracles argument (NMA):

“The positive argument for realism is that it is the only philosophy that doesn’t make the success of science a miracle. That terms in mature scientific theories typically refer …, that the theories accepted in a mature science are typically approximately true, that the same term can refer to the same thing even when it occurs in different theories – these statements are viewed by the scientific realist not as necessary truths but as part of the only scientific explanation of the success of science, and hence as part of any adequate scientific description of science and its relation to its objects.” (Putnam 1975, 73.)

An argument against NMA is a pessimistic induction. Laudan (1981) argues that there have been successful theories whose terms did not refer to anything that exists (according to our best current theories) and, therefore, were not true of the mind-independent world. An example of such a theory is the optical ether theory (see below). Given these theories, we can infer that theories can be successful without being approximately true and that our successful theories might very well not be approximately true. 

A general line of argument that realists use to counter pessimistic induction is the so-called divide et impera strategy. In this strategy, realists identify the theoretical constituents of the past theories that were responsible for their success. The divide et impera move leads to selective realism that characterizes most scientific realists these days. According to selective realism, certain parts or aspects of our current best theories should be taken as at least approximately true.

A related issue is whether past science is a sufficient base for the induction. As Psillos points out,

“one could argue that as science grows, theories acquire some stable characteristics (they become more precise; the evidence for them is richer and varied; they are more severely tested; they are incorporated into larger theoretical schemes and others) such that (a) they can no longer be grouped together with older theories that were much cruder or underdeveloped to form a uniform inductive basis for pessimism and (b) they constrain the space of alternative possibilities well enough to question the extent of the unconceived alternatives predicament” (2009, 73).

The point here is that since (i) science is different today than it was in the past, (ii) most of the science has been produced in recent decades, and (iii) science has been quite stable recently, it seems that the historical challenges do not apply to the current science.

Into the Future

Here we get to the crux of the issue. Stanford (2006) argues that science in the past has been unable to conceive all alternatives to its theories and, therefore, there probably are unconceived alternatives to our current science. This would undermine scientific realism. Recently, Stanford (2015; 2019) has argued that the modern trends in science have, in fact, made it more vulnerable to the problem of unconceived alternatives. These trends include “the professionalization of science in the middle decades of the nineteenth century, the shift to peer-reviewed funding of academic science by the state following World War II, and the ongoing expansion of so-called big science” (2015, 868). Given these trends, it has become increasingly difficult for new research programs to disrupt the existing programs and, therefore, the development of radically different theories is also more difficult. According to Stanford, science has become more conservative.

This is where things get interesting, as the conservativeness of science is an important issue in the philosophy of science independently of its role in the problem of unconceived alternatives. It is here that the recent line of the realism debate we are focusing on takes a future-oriented tone.

This conservativeness is widely considered a problem. As Stanford puts is

“[there is] evidence of a widespread and growing conviction among the members of those very communities that the resulting institutional apparatus of contemporary scientific inquiry has erected severe and/or unprecedented obstacles to the pursuit of genuinely revolutionary, transformative, or unorthodox scientific theorizing. Moreover, in recent decades a wide and growing range of writers on science policy have either expressed mounting concern regarding what they see as the excessive and/or increasing intellectual and theoretical conservatism engendered”. (Stanford 2019, see citations in the paper).

However, Stanford suggests that  “the classical scientific realist can afford to be cavalier or even enthusiastic about evidence of increasing theoretical conservatism in science. After all, she thinks that contemporary theories have things sorted out at least roughly right and that our remaining errors are simply errors of detail.” (2015, 873-874). On the other hand, “the historicist critic of scientific realism thinks that one of the most important ambitions of the scientific enterprise should be identifying and developing the fundamentally distinct and even more powerful successors that will ultimately replace even the most impressive theories of the present day” (2015, 874-875).

Similar conclusion has been made by Fine:

“Suppose that the already existing theories are themselves approximately true descriptions of the domain under consideration. Then surely it is reasonable to restrict one’s search for successor theories to those whose ontologies and laws resemble what we already have especially where what we already have is well-confirmed. And if these earlier theories were approximately true, then so will be such conservative successors.” (1984, 87.)

It seems that there might be a difference between realism and antirealism that can be made visible in their verdicts concerning how the future of science could develop and what we should do in science. However, the issue is not quite so simple. Stanford’s notion of the classical realist should be read in contrast to the selective realist (see above) who accepts that important parts of current theories may be overthrown in the future. Stanford argues that the selective realists and their historicist critics have a “shared vision of the future of science” where both continuities and discontinuities characterize the development of science (2015, 875). It seems that another stalemate has been reached.

Stanford goes on to suggest that the locus of disagreement can be understood along the lines of the clash between catastrophism and uniformitarianism in nineteenth-century geology:

“Uniformitarians argued that the broad topographic and geographic features of the Earth were produced by [–] natural causes acting consistently over long periods of time at the same frequencies and magnitudes we now observe. By contrast, their catastrophist opponents held instead that such causes had operated in considerably stronger degrees in the past [–] and that the earth has steadily and progressively quieted down over the course of its history” (Stanford 2015, 876).

In the case of science, uniformitarians would argue that the future of science resembles its past when it comes to theory-change, while catastrophists would argue that all the major upheavals are in the past and our current theories will not face equally fundamental and radical changes as the past theories. In short, catastrophists think that the future of science is different from its past, while uniformitarians think that the future of science is similar to its past.

Stanford goes on to suggest that scientific realism should be associated with catastrophism in order to maintain coherence (2015, 876-877). Stanford adds that catastrophists do not need to worry about conservatism in science because fundamental changes is theories are not to be excepted (2015, 877). Given this, realism and antirealism differ in practice: Realism endorses the conservative approach in science while antirealism opposes it.

In a later work, Stanford slightly modifies the argument. Stanford (2021, 218) argues that uniformitarian realism is a position that a selective realist may hold. Stanford argues that both realists and antirealists can accept that there are constant theory-changes in the future and that these changes are not as radical (e.g., involving incommensurability) as the more traditional antirealists thought. However, a selective uniformitarian realist is still inclined toward a more conservative approach in science. The realists expect there to be continuity in “whatever it is that they are realists about” (Stanford 2021, 224 emphasis original) and the expectation of such continuity guides the choices of research programs. For example, a structural realist believes that we should commit to the mathematical or structural content of our theories (Worrall 2009) and, therefore, expects that the structure is preserved through theory-change. A similar point has been made by Shaw: “Clearly the kind of continuity we are seeking will differ depending on which [type of realism] is adopted” (2018, 88). According to Stanford, the difference between realism and antirealism lies in that realists assume that we can identify “general or categorical features of scientific theories and/or their supporting evidence from which their approximate truth (or some analogue) can be reliably inferred” (2021, 227). Antirealists do not need to deny that there will be continuity. According to Stanford, antirealists only claim that we are not generally able to predict which parts or aspects of the theories are preserved. (Stanford 2021, 227). Antirealism would claim only that different parts or aspects of theories are preserved in different cases of theory-change. If antirealism was to deny all prospects of continuity, it would face the opposite problem of conservatism, “absurd permissiveness with respect to the alternative theoretical possibilities” (Stanford 2021, 226).

While Stanford’s analysis of the relationships between realism, the future of science, and conservatism are highly nuanced, the overall lesson it has been taken to indicate is that realism and conservatism are closely connected and that realists and antirealism have different expectations towards the future. These different expectations have different consequences for how to select research programs. The recent critiques of Stanford have argued against this association between realism and conservatism.

Dellsén argues that, while Stanford is right in claiming that realists have reasons to believe that radical theory-changes are less probable, “Stanford’s analysis overlooks the impact one’s stance in the debate over scientific realism will have on the value of different outcomes of the relevant type of search” (2019, 33). If we wish to decide what to do, we cannot only focus on the probability of different outcomes but we also have to take into account the value of different outcomes. Dellsén uses expected utility theory to show how Stanford’s considerations are lacking in this regard. The crux of Dellsén’s argument is that, while realists assign a lower probability to the finding of a radical theoretical alternative than antirealists, they value the outcomes of the search for such alternatives more:

First, realists value the finding of a radical theoretical alternative more:

“According to the realist, having identified and developed a radically distinct alternative that is in fact epistemically superior to currently accepted theories has clear scientific value in that it makes it much more probable that what we end up with is a theory or theory-part that is at least approximately true.” (Dellsén 2019, 35).

In contrast “the anti-realist should be even more pessimistic at that point than she was before about the prospects of a given accepted theory being approximately true” (Dellsén 2019, 35). Given that realists find the radical alternative as more likely to be true than antirealists (given the success of the theory), they also find the theory-change as more valuable because it promotes the goal of science, discovering truths. I return to this argument in the next section.

Second, realists value more a failed search for radical theoretical alternatives. Dellsén argues that, from the perspective of realists, the failures provide evidence for the approximate truth of the accepted theories (2019, 36). Since antirealists believe that the current theories are less likely to be true than realists, this type of evidence is not as convincing to them as it is for the realists. Again, the outcome of searching for radical alternatives is valued more by realists.

Given that realists place greater value on the potential outcomes of the search for radical alternatives, assigning a lower probability to finding such alternatives does not determine what one should do in science. It seems that both realism and antirealism can be motivated to oppose conservatism in science.

There Is a Difference in What They Say about the Future

In futures research, possible, probable, and desirable futures are studied (Amara 1974; Bell, 2009). An essential component in the mapping of futures is the critical study of our own conceptions that ground different scenarios of the future (Bell, 2009; Inayatullah, 1998). The interplay between understanding possible futures and the critical study of our conceptions about the future is captured best by the notion of alternative futures that has been a guiding concept in futures research (see Slaughter, 2020).

Futures research can be understood as centering around scenarios of the future:

“The goal of scenario writing is not to predict the one path the future will follow but to discern the possible states toward which the future might be ‘attracted.’ [–] If a prediction is a definitive statement of what the future will be, then scenarios are heuristic statements that explore the plausibilities of what might be.” (Staley, 2002, 78).

While there are different definitions of a scenario and subtle differences between the definitions, in this post, we can consider a scenario simply as a “description of a future situation and the course of events which allows one to move forward from the actual to the future situation” (Amer et al., 2013, 23 emphasis added).

While there are several methods to create scenarios, in order to achieve the goals of the post, we can focus merely on certain overall features that scenarios can have. For the purpose of distinguishing between scientific realism and antirealism in terms of scenarios of the future, it is enough that we identify the following aspects of a scenario:

1. Is the scenario a probable, possible, or alternative future? These are vague notions but, roughly, a probable future is a future that we consider as likely above a certain threshold; a possible future is a future that can happen, given what we believe about a domain; and an alternative future is a future that follows when we question our assumptions about some issue in the domain (Inayatullah 1998) for some understandable reason.

2. Is the scenario a desirable future? In other words, do we find the future ethically sound and in accordance with our values and goals? (See Bell 1997.) Values and goals are the main dimensions that matter in our analysis below.

3. Is the scenario trackable, i.e., can we form a path to the scenario and can we continue further into the future from the scenario? In other words, does the description of the scenario imply that we cannot know or explain happened before or after?

4. Why do we answer questions 1-3 in a certain manner with respect to the scenario, or why we cannot explain our answer?

There are several other features that we can use to characterize a scenario, but the four features above are general enough to be widely applicable and they are also sufficient for the modest purpose of this post: I am not attempting to exhaust all the scenarios that realism and antirealism create. Rather, I show that there are differences in how realism and antirealism understand the scenarios, and this is enough to show that realism and antirealism can be distinguished on the basis of their implications for the future.

There are several ways of formulating scenarios of the future on the basis of scientific realism and antirealism (see Virmajoki 2023). However, given the focus on the issue of conservatism in science in the recent debate, we may take it as a starting point. We begin by creating two scenarios:

1. Scientists follow a research program that is not radically distinct from the current empirically successful and mature program.

2. Scientists follow a research program that is radically distinct from the current empirically successful and mature program.

That both realists and antirealists equally accept these scenarios as possible/probable is not controversial. We need to elaborate.

1.1 The program leads to empirical success.

1.2 The program is not empirically successful.

Within this step, there are some differences. First, realists will take 1.1 as more probable than antirealists, and antirealists will take 1.2 as more probable than realists. After all, antirealists can more readily allow that the current science will face a crisis as far as they are not committed to the approximate truth of our theories. These different judgments of probability are widely shared and, for example, Dellsén’s (2019) argument takes them for granted. Yet, there is no controversy about whether it is a possible – or at least an alternative – future where success is no longer achieved. Both realists and antirealists can accept this. In general, realism is not committed to an infallibilist position (Psillos 1999, 78). Moreover, both realists and antirealists can agree that 1.1 is a scenario of a desirable future, at least for the instrumental success that follows. Yet, one might argue that is somewhat more desirable for realism, as it increases the possibility that we can achieve success on the basis of current theories that realists also consider as epistemically highly valuable discoveries of truth.

In 1.1, realists have an explanation for the success: it is due to the parts that were preserved in the choice of the research program. More importantly, this enables realists to infer that, in scenarios further in time, these parts will be preserved. Scenario 1.1 is trackable to the realists. Realists can follow the scenario further into the future. Of course, different types of realists will make different conclusions about which parts or aspects of theory will be preserved, but this does not alter the fact that the scenario is trackable for all types of realists, as they all will find some part or aspect that they can project onto the future. In contrast, even the most sophisticated antirealist á la Stanford cannot track the scenario. Remember that this sophisticated antirealist can accept that parts or aspects of a theory are preserved but it changes from case to case which parts or aspects are preserved (Stanford 2021, 227; see the previous section for discussion). Given that the antirealist does not have the general categories that realists can use to track the scenario, the scenario is untrackable to the antirealist.

I leave scenario 1.2 as an exercise to the reader, as I will deal the details in a paper.

Next, take the following scenarios:

2.1 The radically different research program leads to success.

2.2 The radically different research program is unsuccessful.

In this case, realists will take 2.2 as more probable than antirealists, and antirealists will take 2.1 as more probable than realists. Moreover, as we have seen, Dellsén (2019) has argued that realists find both outcomes more desirable than antirealists. According to Dellsén, in 2.1 realists would believe that the new, fundamentally different but superior, theory would be more likely to be true. However, I think we should resist this conclusion. The description of the scenario involves a change that would undermine, in the spirit of pessimistic induction, the realists´ idea of the connection between success and truth. After all, the scenario shows that a successful theory – the current theory that has been replaced in the scenario – can be fundamentally wrong. Even if the realists would hold the new theory as true, they would still lose their ability to track what happens in the future, as there would be no way of excluding futures with radical alternatives by using the existing knowledge at any point in time. History would have shown that radical alternatives can be successful and pursuitworthy. It does not seem that any features of scenario 2.1 enable realists to resist conservatism. Why put effort to pursue a future that is not probable, desirable, or even trackable?

In 2.1, antirealists have at least one reason more reason to pursue the scenario: they consider it more likely. However, the scenario is no more trackable to the antirealists and many versions of antirealism would, again, lose their explanation for the success of a theory. Moreover, a sophisticated antirealist would be equally staggered as the realist. Again, the sophisticated antirealists’ claim that there are different types of continuities that hold in different situations of theory-change would be undermined. When the sophisticated antirealist keeps the possibility of scenario 1.1 alive by allowing continuities, this fires back in making scenario 2.2 less likely and more awkward for her. A sophisticated antirealist who wishes to understand futures with continuities struggles to understand futures with radical alternatives. Softening the antirealist position does not seem to enable the antirealist to orient themselves towards more future possibilities.

In 2.2, realists have a clear path to follow: simply return to the earlier theory. For a realist, scenario 2.2 leads back to scenarios 1.1 and 1.2. In fact, as Dellsén (2019) pointed out, the realist might have even better reasons to believe that the earlier theory is true in this case. If Dellsén is correct in this, scenario 1.1 which is favorable in the trackability for the realist is even more likely, and deserves even more attention according to them.

Scenario 2.2 is desirable for realists both in confirming their convictions and in allowing the realists to navigate the scenarios of the future better. On the other hand, antirealists need to fall back to the earlier theory without the prospects of drawing a lesson about the failure. Moreover, the scenario seems to provide evidence for realism and therefore against antirealism. The so-called put-up-or-shut-up argument (see Soler 2015) becomes even stronger: Either you create a radically different science or you quit claiming that such science is possible.

This brief and crude study of the features of scenarios that antirealism and realism imply shows that, while the positions can agree on many issues about the future of science, there are subtle differences in what they consider likely, possible, desirable, and trackable futures and how they explain their considerations. In fact, the brief survey indicates that while antirealism may have fewer possible surprises in the future, it remains more ambiguous about which futures are desirable. It does not explain (or explain better than realism) why certain futures should be valued beyond improving, maintaining, or restoring empirical and practical success in science. Moreover, antirealism does not seem to have any scenario where it can say, “be aware, if this happens, only I can tell you what will happen,” because there does not seem to be any scenario that is more trackable to antirealism than realism. One might even wonder why to keep antirealism on the table when discussing the future, as it does not enable us to think about the scenarios of the future any further than realism. I think there is a truth in this worry, but we must remind ourselves that adequate futures research remains critical to our conceptions of the world. Moreover, there might be cases where antirealism fares better. Our survey above was anything but exhaustive. Even if realism provides better ground for scenarios, antirealism needs to remain on the table to remind ourselves that realism and its future scenarios are not the only ones we can have. It might be that antirealism is true. If so, building mere realist scenarios would lead us astray about what could happen.

Who Cares about Philosophical Positions?

In the previous section, I pointed out that realism and antirealism, even in their sophisticated forms, differ with respect to what they say about the future and what they enable us to infer about the future. Once we evaluate (i) which scenarios are probable, possible, or alternative futures, (ii) which scenarios are desirable futures, (iii) which scenarios are trackable, and (iv) why (or why not) we answer the questions the way we do, then we can see differences between realism and antirealism. One of the main differences between the positions that explain the differences in their scenarios of the future seems to be that realism allows more continuity than antirealism. However, we have seen that it can be denied that realism implies such continuity. As we saw, Tambolo and Cevolani argue that

“sophisticated realists claim that our best theories are the best relative to our current knowledge, but do not need to be the best in some absolute sense. This makes them reasonably well-equipped with a sophisticated assessment of the historical track record of science: nothing, in their position, implies that theories, or parts of them, are ‘untouchable’ in the sense that they cannot be improved even in a radical way. Also on this count, then, we fail to see the essential difference between realists and antirealists that Stanford’s account seems to imply.” (2023, 11.)

It is not quite obvious whether realism can, in fact, fall back to such a position without losing its basic conviction that our best theories are approximately true descriptions of the mind-independent world and that we can trust that certain parts of them are survive theory-changes. However, we can leave that issue aside. Maybe such a position can count as realism, maybe not. The important thing to notice is that, if the notion of continuity is dropped from realism, then Stanford’s sophisticated antirealist with the idea that, while there are continuities, we cannot know on a general level which parts of a theory will be preserved, and the sophisticated realist á la Tambolo and Cevolani do come rather close to each other.  In this case, the realists´ and antirealists´ futures do come quite close to each other.

However, it seems difficult to understand why there should be any debate about distinguishing between realism and antirealism if their core claims come together. The problem of distinguishing between realism and antirealism was supposed to be that, despite antirealism and realism making fundamentally different claims about science and its achievements, these claims do not seem to make any concrete difference in how we look at the workings of science and its history and future. In a situation where the core claims are almost exactly the same, the problem of distinguishing the positions becomes trivial.

It is interesting in its own right if realism and antirealism have reached such consensus. Given the a consensus, it would become ever more easier to provide scenarios of the future of science – realists and antirealists would agree on everything. Difficult questions concerning which of the two competing sets of scenarios we should commit to would no longer arise. I think this would be a remarkable achievement. Of course, this situation would be bad news to those who commit to either realism or antirealism for some reason that stems from their fundamentally different understandings of science because such a difference would not exist. I think this is a small price for much easier scenario-work.

But who cares about philosophical positions anyway? Well, we all should. The consensus between realism and antirealism has not been, in fact, reached, and there are several positions in the debate that one can adopt and defend. All these positions do have different things to say about the future. Especially many realists do, in fact, defend the idea that there are identifiable continuities in science. Given the realists’ commitment to continuities, there are differences between realists and antirealists that make a difference. The fact that the most watered-down versions of antirealism and realism come together does not change this fact.

To put it simply: If we are interested in the future, then the realism vs. antirealism debate makes a difference, as the positions in the debate have different consequences for the scenarios of the future. On the other hand, if we are interested in sticking with philosophical labels and protecting them from all possible counterarguments, then we might end up in a consensus position that makes no real commitments to anything and, therefore, remains mostly silent about the future. But who cares about the labels? I think the future is by far the more interesting topic, and it is here that realism vs. antirealism makes a great difference, given the richness of positions in the debate. The richness enables us to build a rich set of different future scenarios.

References:

Amara, R. (1974). The futures field: Functions, forms, and critical issues. Futures, 6(4), 289–301.

Amer, M., Daim, T. U., & Jetter, A. (2013). A review of scenario planning. Futures, 46, 23–40.

Bell, W. (2009). Foundations of futures studies volume 1 (5th ed.). Transaction Publishers.

Bell, W. (1997). Foundations of Futures Studies II: Values, Objectivity and the Good Society. Transaction Publishers.

Dellsén, F. (2019). Should scientific realists embrace theoretical conservatism? Studies in History and Philosophy of Science, 76, 30–38.

Fine, A. (1984). The natural ontological attitude. In J. Leplin (Ed.), Scientific realism. 83–107. University of California Press.

Inayatullah, S. (1998). Causal layered analysis: Poststructuralism as method. Futures, 30(8), 815–829.

Laudan, L. (1981). A confutation of convergent realism. Philosophy of Science, 48(1), 19-49.

Psillos, S. (1999). Scientific Realism: How Science Tracks Truth. Routledge.

Psillos, S. (2009). Knowing the Structure of Nature: Essays on Realism and Explanation. Palgrave Macmillan.

Putnam, H. (1975). Mathematics, matter and method. Cambridge University Press.

Soler, L.É. (2015). Why Contingentists Should Not Care about the Inevitabilist Demand to “Put Up-or-Shut-Up”: A Dialogic Reconstruction of the Argumentative Network. In E. Trizio & A. Pickering (Eds.), Science As It Could Have Been. Discussing the Contingency/Inevitability Problem. 45-98. University of Pittsburgh Press.

Staley, D. J. (2002). A history of the future. History and Theory, 41(4), 72-89.

Stanford, P. K. (2006). Exceeding Our Grasp: Science, History, and the Problem of Unconceived Alternatives. Oxford: Oxford University Press.

Stanford, P. K. (2015). Catastrophism, uniformitarianism, and a scientific realism debate that makes a difference. Philosophy of Science, 82, 867-878.

Stanford, P. K. (2019). Unconceived alternatives and conservatism in science: The impact of professionalization, peer-review, and big science. Synthese, 196, 3915-3932.

Stanford, P. K. (2021). Realism, instrumentalism, particularism: A middle path forward in the scientific realism debate. In T. D. Lyons & P. Vickers (Eds.), Contemporary scientific realism: The challenge from the history of science. 216-238. Oxford University Press.

Tambolo, L. & Cevolani, G. (2023). Realism, Antirealism, and Theoretical Conservatism. Synthese 201(1):1-18.

Virmajoki, Veli (2023). Pursuitworthiness in the scheme of futures. European Journal for Philosophy of Science, 13(1). 1-26.

Worrall, J. (1989). Structural realism: The best of both worlds? Dialectica, 43, 99–124.