In this post, I discuss how philosophy of historiography helps us to understand the estimating of futures. More specifically, by focusing on my forthcoming publication, I discuss why we need a philosophical account of historiographical explanation to understand the use of historiography in scenario-work.
Soon, my paper “What Should We Require from an Account of Explanation in Historiography?” will be published in Journal of the Philosophy of History. In that paper, I argue that
- Accounts of explanations in historiography have not been fully developed and many important notions have not been defined by them.
- A counterfactual account can help us to define the notions.
An account of explanation is a philosophical account which defines the conceptual structure of explanation. For example, Hempel´s famous Covering Law model is an account of explanation that attempts to define the structure of explanations (in the model, an explanation is a deduction of the explanandum from premises describing initial conditions and universal laws). A fully developed account of explanation is an account that defines important explanation-related notions on the basis of the conceptual structure. For example, a fully developed account should tell when one explanation is deeper than another. It is necessary to develop the accounts of explanations to their full extent because we need to understand many explanation-related notions and because we can only assess the relative merits of different account by comparing how well they manage to do clarify explanation-related notions issues and notions.
In the philosophy of historiography, accounts of explanation have attempted to define the conceptual structure of historical explanations. For example, it has been suggested that if Z explains X, Z and X must be connected by universal laws (see Hempel 1942); that the connection is that of reason and action (see Collingwood 1946; Dray 1957); that there is an (ephemeral) mechanism (see Glennan 2010); that the connection is that of difference-making (see Weber 1949; Ben-Menahem 2016); or, as the limiting case, that the connection requires no analysis (Tucker 2004), just to mention some prominent suggestions.
In the forthcoming paper, I defend a version of the difference-making account. I argue that historiographical explanations track down patterns of counterfactual dependencies. In what follows, I introduce the basic ideas of my account and explain why they are important in understanding the future
1. Counterfactual Dependencies
In James Woodward’s account of explanation, “the underlying or unifying idea in the notion of causal explanation is the idea that an explanation must answer what-if-things-had-been-different questions, or exhibit information about a pattern of dependency” (2003, 201). To put it simply, explanations answer questions of the form “Why X rather than Y?” by pointing out factors Z and W such that “had W rather than Z been the case, Y rather than X would have been the case”.[1] Explanations provide information about counterfactual dependencies between explanans and explanandum and are contrastive in nature.[2] An explanation “must enable us to see what sort of difference it would have made for the explanandum if the factors cited in the explanans had been different in various possible ways” (Woodward 2003, 11).[3]
Let’s take an example. We ask
“Why did scientists come to believe that atoms exist?”
The answer to this question is
“Scientists believe that atoms exist rather than believe that atoms do not exist because Einstein formulated an explanation of the Brownian motion and Perrin confirmed this explanation with his experimental work. Had there not been such explanation or experimental work, scientists would not believe in atoms.”[4]
Notice that here we have a very simple explanation. The explanation shows how the belief in the existence of atoms depends on the work of Einstein and Perrin. Of course, such explanation does not tell much. It is rather minimal in the quite literal sense that it cites only two counterfactual situations such that had one of those occurred, the outcome would have been different: one where no explanation of the Brownian motion was formulated, and one where the explanation was not confirmed by experimental work. A much more satisfactory explanation would also describe what the assumptions, methods, and values of the relevant scientific community at the time, and what kind of processing was required before the results of Einstein and Perrin became accepted.
This means that we have a continuum of explanations: A minimal explanation gives information about one counterfactual situation where the outcome would have been different. A complete (or ideal) explanation describes all the counterfactual situations where the outcome would have been different. Complete explanations are extremely complex and even unachievable. Therefore, a historiographical explanation cannot be expected to be identical with a complete explanation. Rather, historical explanations cite some set of counterfactual situations where the outcome would have been different. Historiographical explanations are collections of true counterfactual statements of the form “had W rather than Z been the case, Y rather than X would have been the case,” that is, a historiographical explanation consists of a set of minimal explanations.
First lessons:
Historical explanations require knowledge of counterfactual situations. Both historiographical explanation and scenario-work require that we track down developments of which we do not have direct evidence. Investigation in the structure of counterfactual thinking can be highly fruitful in scenario-work (see the post https://blogit.utu.fi/futuresofscience/2020/05/29/327/).
Moreover, historiography can never tell the full story. We need to understand how and why certain factors are chosen in a historiographical explanation. This can reveal the possible relevance of a historiographical explanation in scenario-work by informing us about the inherent “perspective” of the explanation. Let me explain
2. Competition and Goals
As there can exist many historiographical explanations for an explanandum of the form “why X rather than Y”, it becomes important to understand when two such explanations compete. Even though we cannot find the explanation for something, it does not follow that all explanations are compatible. We have to understand when two explanations are competitors.
To clarify explanatory competition, we can first define when two minimal explanation compete. They compete when one tells that the outcome X depended on Z and other one tells that X did not depend on Z. We can continue and say that two historiographical explanations are competitors if they either incorporate competing minimal explanations or if they incorporate different sets of minimal explanation. In the latter case, omitting a minimal explanation does not automatically mean that its explanatory relevance is denied. It can simply mean that the minimal explanation is not relevant from the perspective of an explanatory goal (see below) and we should be careful to recognize when the difference in explanations is due to a difference in explanatory goals and when it is not.
For example, the following historiographical explanations are competitors:
(1) Scientists believe in atoms rather than believe that atoms do not exist because Einstein formulated an explanation of the Brownian motion and Perrin confirmed this explanation with his experimental work. Had there not been such explanation or experimental work, scientists would not believe in atoms.
(2) Scientists believe in atoms rather than believe that atoms do not exist because Einstein formulated an explanation of Brownian motion and Perrin confirmed this explanation with his experimental work. Had someone else formulated the explanation or performed the experimental work, scientists would not believe in atoms.
There are two senses in which the two can be competitors. One is the ‘hard sense’: they disagree whether it was the personal prestige or the details of the work that was explanatorily relevant in the process. In the case of hard competition, the explanation (1) must have the following structure:
(1c) Scientists believe in atoms rather than believe that atoms do not exist because Einstein formulated an explanation of the Brownian motion and Perrin confirmed this explanation with his experimental work. Had there not been such explanation or experimental work, scientists would not believe in atoms. However, had someone else formulated the explanation or performed the experimental work, scientists would still believe in atoms.
Both (1c) and (2) cannot be true and, therefore, there exists a hard competition between them. In such cases, the competition can be resolved only by careful historiographical research and judgement.
The other one is the ‘soft sense’. Assume that it is true that both the details of the work and the personal prestige are explanatorily relevant, that is, had the details or the person who did the work been different, the outcome would have been different. Then the difference between (1) and (2) might reside in a difference in the explanatory goals. We could say, for example, that (1) aims to underline the role of explanations and experiments in the history of science, whereas (2) aims to underline the role of social hierarchies in the history of science. It is important to distinguish these two senses of explanatory competition.
Historians with different explanatory goals, participating in soft competition, can always attempt to find common ground in an explanation that is more complete than the ones the historians initially provide. For example, if the difference between explanations (1) and (2) above is soft and reflects different explanatory goals, the beliefs in atoms can be explained as follows:
Had there been no explanation or experimental work OR had these works been produced by someone else than Einstein and Perrin, scientists would not have come to believe in atoms.
Two historians with different explanatory goals could both accept this claim about the patterns of counterfactual dependencies. However, if they did not, we would have a hard competition between two explanations. This indicates the importance of the distinction between hard and soft competition: the latter allows us to achieve deeper explanations by combining different historiographical results, the former does not.
But how should the notion of explanatory goals be understood? No neat catalogue of such goals can be given. The notion is based on the idea that historians want to highlight certain aspects of the process that led to the explanandum. The explanatory goal is what the explanation seeks to establish in addition to providing information about conditions on which the explanandum depends. An explanation may seek to pinpoint the role of some types of factors (F) in history and this attempt is centered around the (meta)explanatory goal of understanding the role of F in historical processes. In this sense, the choice of the explanatory goal aims at second-order understanding (whereas an answer to a particular explanandum gives us first-order understanding).
There are (at least) two possible approaches to the choice of explanatory goals. First one is axiological. In the axiological approach, the choice of explanatory goals reflects our values and we want to understand the role of an F in history because F is considered important with respect to our values. The second approach is epistemic. In this approach, we attempt to find out a set of factors we can use to effectively explain much of history. For example, we might make the hypothesis “experiments play an important role in scientific developments” and attempt to confirm or falsify this hypothesis by taking a number of historical cases and evaluating the explanatory power of experiments. Notice that sometimes the axiological and epistemic approach coincide. For example, we might attempt to give a marginalized group the recognition it deserves by pointing out how the actions of that group have shaped history.
Second lesson
Historical explanations often seem to compete. We should not fall in despair (“everyone says different things”) in such cases but analyze whether the competition is hard or soft. This can be achieved by analyzing the possible goals behind historiographical studies. Soft competition is beneficial in that (i) it reveals how history can be seen from different perspectives, and still we can (ii) achieve deeper knowledge by connecting soft competitors together. Hard competition is beneficial in that it forces us to think seriously about our commitments to certain historical narratives in our thinking of the world. The use of historiography in thinking about the future is both a constructive and a destructive process. Historiography is neither a collection of views that can be fit together in a scenario-structure nor a field of never-ending disagreement about the past. The surface features of historiography (“every generation has its own explanation”) easily mislead us about the real continuities and discontinuities in the field. A philosophical account of historiographical explanation must go beneath these surface features and explicate notions such as explanatory competition and goals. Such work makes it possible to use historiography carefully in other epistemic practices, such as scenario-work
3. Explanatory Depth
The notion of explanatory depth has gone mostly unnoticed in the philosophy of historiography and, unfortunately, the lack of understanding about the notion has sometimes led to the ridiculing of important historiographical issues. For example, it has been argued that
”The best explanation for the independent diaries of soldiers of the same unit that state that on a certain date their unit came under heavy bombardment and therefore panicked and retreated is that indeed they came under heavy bombardment and therefore panicked and retreated. There is no need for further knowledge of psychology or human nature under fire. Had this explanation of the retreat depended on psychological theories, it would have been indeterminate, since under fire soldiers are known to retreat out of fear, become paralyzed with fear and stay put, or become emboldened with rage and charge forward. The mere description of the stimulus together with background conditions and contemporary psychology is insufficient for explaining the actual retreat. But psychology is redundant here”.[5]
The passage is correct in claiming that sometimes providing a (rather minimal) explanation is a rather simple task, given the evidence. However, it seems that “knowledge of psychology and human nature” is too quickly dismissed. For the very reason that “under fire soldiers are known to retreat out of fear, become paralyzed with fear and stay put, or become emboldened with rage and charge forward” it would be interesting to know why they did retreat in this case. In other words, we would like to know what properties of the group of soldiers would have led to a different behavior under the bombardment. For example, perhaps less experienced soldiers would not have retreated because they would not have understood the danger. It seems that if we knew that both the heavy bombardment and the experience of the soldiers made a difference in the outcome, we would have achieved a deeper explanation. If psychology can provide us with such explanatory knowledge, it is far from redundant. An account in explanation in historiography should be able to dissect such dimensions of explanatory practice.
I have followed Woodward and Hitchcock (2003) in defining that a historiographical explanation H is deeper than H* when H is better at answering what-if-things-had-been-different questions (what-if questions) that are important to us (to simplify the matter). For example, assume that the following explanations are soft competitors:
(1) Scientists believe in atoms rather than believe that atoms do not exist because Einstein formulated an explanation of the Brownian motion and Perrin confirmed this explanation with his experimental work. Had there not been such explanation or experimental work, scientists would not believe in atoms.
(2) Scientists believe in atoms rather than believe that atoms do not exist because Einstein formulated an explanation of Brownian motion and Perrin confirmed this explanation with his experimental work. Had someone else formulated the explanation or performed the experimental work, scientists would not believe in atoms.
Then it follows that (1) and (2) together (i.e. “Had there been no explanation or experimental work OR had these works been produced by someone else than Einstein and Perrin, scientists would not believe in atoms”), provide a deeper explanation than either of them alone. It answers more what-if questions. It tells what would have happened (a) if there was no explanation, (b) if there was no experimental confirmations, (c) if the explanation was not provided by Einstein, and (d) if the experimental confirmation was not provided by Perrin.
Lesson three
Once we define explanatory depth in terms of ability to answer what-if questions, the search for deeper explanations becomes directly relevant to the estimating of futures. The deeper the explanation, the more we know about different counterfactual situations. Some of these situations sometimes resemble our present conditions or conditions we can create. Therefore, deeper explanations increase our chances of understanding what will happen in the future. For example, if we know what would have happened without there being the explanation of the Brownian motion, we can tell that explanatory work could change science in the future. However, if we also know what would have happened without experimental confirmation, then we can tell that the future might depend on the existence of both a novel explanation of some phenomena and an experimental work supporting the explanation. If we did not know the relevance of experimental work, we could be misled to expect that a mere explanation can have a decisive effect on the future of science. We would have located ourselves in wrong area of the modal space: one where only the explanation for some phenomena is missing, while in reality we would be in an area of the space where both the explanation and the explanatory work are missing.
However, there are some complications. The notion of “important what-if question” is discipline-dependent. For example, in historiography the historical context (whatever this means) where the explanandum is embedded needs to be somehow taken into an account when the explanans is formulated. We cannot build every feature of a historical situation into the explanans and treat them as equally changeable. This would, in a sense, tear down the historicity of these situations. For example, it might not seem right to assume that the ethos of scientists in some era could have been different. The fact that scientists had that ethos might seem to be constitutive of that era. If we wanted to understand why something happened in the context of that era, it would be a distortion to provide a scenario where the ethos would have been different. For example, if we added to our explanation above that
“Had the scientific community lived in an Aristotelian framework, Einstein’s and Perrin’s work would not have had any effect.”
we arguably would not have achieved a deeper explanation than the ones we already have despite our being able to answer one more what-if question.
However, even if a change in explanans is missed as historically irrelevant, it might be still be relevant to the future. For example, we might be interested in scenarios where different context has been created and it would highly relevant to know how certain types of factors would work in those contexts. Therefore, historians’ judgement of explanatory depth (as well as similar judgement by any discipline) must be taken with a pinch of salt in scenario-work
4. Intervention, Counterfactual History and Contingency
It is a well known fact that not just any counterfactual dependency is explanatorily relevant. For example, a storm would have occurred, had the reading of a barometer been low; however, the reading of the barometer is not explanatorily relevant. Woodward’s account of explanation is often called ‘interventionist’ because it requires that explanatorily relevant counterfactual dependencies must remain invariant under an intervention.
“The intuitive idea is that an intervention on X with respect to Y changes the value of X in such a way that if any change occurs in Y, it occurs only as a result of the change in the value of X and not from some other source.” (Woodward 2003, 98).
Consider that we ask, “Why did a storm occur?” We could answer: “Because the barometer reading fell. Had it not fallen, there would not have been a storm.” This answer does not provide us with an explanation even though it answers a what-if-things-had-been-different question. The relevant answer would be: “Because the atmospheric pressure fell. Had it not fallen, there would not have been a storm.” The difference between these answers is that the second one is true if an intervention on the atmospheric pressure was performed, while the first one is false if an intervention on the barometer was performed: if we hold the needle of the barometer fixed, the storm would have occurred anyway (as the atmospheric pressure fell).
Notice the following:
(1) Interventionism allows us to be precise in the way in which the antecedent of a counterfactual conditional is changed but does not require us to be too precise. The only requirements for the change are that (i) the change satisfies the definition of intervention, (ii) the change in the antecedent is the one that is supposed to be explanatory according to the one who formulates the explanation. This also means that interventionism provides tools to clarify obscure explanatory claims. For example, Lebow and Stein (1996) argue that the truth-value of “had the United States attacked the missile bases, the Soviet Union would have responded to an attack on Cuba with military action of its own” can be assessed only after we specify what kind of attack is in question. Interventionism gives us the tools to clarify the claim: the one who thinks that the counterfactual is true and explanatory must, in the interventionist account, provide a scenario where intervention I is performed and tell what kind of attack was launched as the consequence of I. For example, someone could say that “had the United States decided to launch a nuclear attack…” is the intervention she thinks makes the counterfactual true. If this is the case, then one is committed to a very specific explanatory claim “had there been a nuclear attack… then the Soviet Union…”, not the claim that any attack whatsoever would have led to a Soviet response.
It is very important to notice that one does not need to care whether a nuclear strike was really considered as an option by those in charge (a common mistake in discussions about counterfactual histories, see this post). As long as it is true that an intervention causing a nuclear strike would have led to an attack by the Soviet Union, we have an explanation for why the Soviet Union did not attack: The Soviet Union did not attack rather than attacked because there was no nuclear strike rather than there being a nuclear strike.
(2) The so-called “minimal-rewrite-of-history” rule says we should avoid far-fetched counterfactuals in historiography and consider only those counterfactual situations that could have happened, given the actual history. In the interventionist account, this rule is rejected when explanatory relevance is the concern. “X rather than Y because Z rather than W” is explanatory if W would have led to Y no matter how far-fetched W is. If a (conceptually possible) intervention makes W to be the case and if Y follows, we have an explanation.
Nevertheless, the rule serves two different functions in the counterfactual account of historiographical explanation.
First of all, the minimal-rewrite rule is a useful methodological rule. It might be too difficult to evaluate a counterfactual citing a far-fetched change in history because we have no cases to which we could compare the counterfactual.
Secondly, the minimal-rewrite rule is directly connected to the issue of historical contingency. If a historiographical explanation is requested to show the contingency of events, as is sometimes argued, then the minimal-rewrite rule guarantees that we are able to show exactly that. For example, Beatty argues that
“Narrative explanations are more demanding than counterfactual, difference-making explanations more generally, in that they go beyond “what-if-things-had-been-different questions” to indicating that things could very well have been different.” (2017)
To see this, consider my (2018) suggestion on how to define historical contingency (in the context of contingency of science):
“It could have been the case that science has the feature F* rather than the actual feature F, where the difference between F and F* is considered interesting in the given context of discussion.
We can accept that F* could have been the case if and only if (a) someone points out a counterfactual past event Z that would have led to F*, and (b) it is shown that the occurrence of Z in the past is not an impossible (or extremely far-fetched) scenario.
Feature F can be judged to be (a) inevitable if and only if the occurrence of any Zi is seen as impossible and (b) a truly chancy feature if Z is a part of the actual history. Moreover, the more far-fetched the occurrence of Z is judged to be, the more inevitable feature F is.”
Given this definition of contingency, an explanation showing that W was not far-fetched implies that X was a contingent event. This means that if we only consider historical situations that are close to the actual history and still find when Y rather than X would have been the case, we have shown the contingency of X. The minimal-rewrite rule therefore makes the phenomena of historical contingency and historiographical explanation inseparable. It seems that the alleged plausibility minimal-rewrite rule is based on its connection to historical contingency. If (and this is a big IF) we want our explanations to reveal the contingency of an historical event, then the rule is justified within an account of explanation.
However, it seems that these two phenomena, historical contingency and historiographical explanation, should be kept separated. First, it seems arbitrary to limit the explananda of historiography to contingent events and processes. Some interesting events and processes are probably rather inevitable but we would nevertheless like to have explanations for them.
The minimal-rewrite rule therefore serves important historiographical functions, but we should not make it a defining feature of historiographical explanations. The (interventionist) counterfactual account of explanation is able to incorporate both these facets of the minimal-rewrite rule: it defines the notion of explanation without a reference to the rule but still recognizes its methodological and contingency-indicating roles.
Fourth lessons
A historical explanation must tell when things would have been different under an intervention. Tracking down counterfactual developments that take place after a hypothetical intervention enables us to steer clear on the nature of the scenario we are considering. It could be a fruitful heuristic also in the creation of future scenarios. If we want to know what would happen if W happens in the future, the scenarios will be cleared if (a) we specify how W would take place in a scenario, and (b) assume that W did not happen as a result of a change that affects everything in the scenario. For example, if we build scenarios where an even more deadly pandemic took place, we should specify with respect to each scenario (a) how the pandemic spreads (recognized vs. unrecognized), and (b) assume that there are not other great changes that made the pandemic possible, like a total war. The point (b) is important when we want to know what would happen if another even more deadly pandemic took place; scenarios for total wars require distinct treatment. The point is that it is useful to focus on different scenarios at different times and an interventionist account of historiographical explanation can help to clarify the issue.
However, there is a difference between explanatory counterfactuals and “scenario-relevant” counterfactuals. Sometimes we would like to know whether some Z could lead to some X even without an intervention (and without Z being a cause of X). For example, Z could serve as evidence for the future, in the same way as a falling barometer serves as evidence for a storm. Still, it is important to distinguish between “causal scenarios” that develop under intervention and “evidential scenarios” which we infer on the basis of some piece of information that does not have causal effect on the future. Again, this is important because the distinction enables us to focus on one well-defined scenario at a time.
Secondly, it could be useful to find clues for the future from those histories that satisfy the minimal-rewrite rule. As we noted, these are methodologically better behaving than other scenarios. Moreover, it might be useful to focus on such counterfactual histories because they are connected to the contingency of events; and the more contingent a future event or process is, the easier it is to achieve or avoid. However, sometimes minimal-rewrite scenarios can mess up important future scenarios. For example, the battle against climate change will probably involve dramatic changes in our lifestyles. Understanding the possible futures involving those changes is important but that understanding cannot be achieved through scenarios that focus on minimal changes in our current conditions. Sometimes we need to know what would have happened or what will happen, given some drastic changes. These surely are methodologically difficult questions, but their importance must make us rethink issues like the minimal-rewrite rule and other “safe-bet-schemas of reasoning”
5. What to Explain?
Jo Guldi (2019) has pointed out that the choice of explanandum in historiography has not received the attention it deserves. Guldi points out that these choices are often made on the basis of canons of the given historiographical field but adds, correctly it seems, that “questions of significance [of explananda] are far from being settled.” (2019, 351.) Next, I discuss which considerations are relevant in the choice of explananda.
First, it might be argued that we must explain those historical events and processes that are the most relevant in the historical developments. However, the choice of explananda on the basis of their causal significance does not get off the ground before we decide which of the later developments require an explanation. Every historical event belongs to some causal chain and is therefore relevant to some historical development. If every historical event is part of some causal chain, then being a part of a causal chain cannot be the criterion for significance. We need to ask further which of those causal chains are significant. It seems that this choice must, ultimately, be based on our values.
How such judgements based on values can be meaningfully made cannot be discussed here.[6] However, it must be noted that, when evaluating the significance of “why X rather than Y”, both the choice of X and the choice of the contrast must be carefully examined. It is not enough that X is significant but Y must also be a somehow interesting alternative to X. For example, of the two questions
(I) Why did scientists come to believe that atoms exist rather than have no beliefs at all?
(II) Why did scientists come to believe that atoms exist rather than believe that atoms do not exist?
the latter seems more significant. One could argue, for example, that we are more interested in knowing how we could have had different science than we are in knowing when there would not have been science at all. In such cases, the difference in significance lies in the contrasts.
Finally, one could perhaps argue that the whole question about significance is mistaken since it has been claimed that historiography mainly explains evidence, not events (Tucker 2004, 186). Even though evidence undoubtedly is the basis of all historiography and inference to the best explanation an important epistemological tool, this approach puts the cart before the horse. In most cases, the evidence is gathered in order to find an explanation for something. The notion of evidence makes sense only if one has a particular problem in mind, given the number of texts, artifacts, natural objects and so on that could serve as evidence in some possible historiographical inquiry. Moreover, it would be an extremely sorry state of affairs if historians were to focus on explaining their evidence, (i.e. if evidence was the most significant thing to be explained). Surely, historiography can answer (and in fact answers) many questions that stem from our concerns as human beings. What to explain cannot be based on evidential relations. There is no escape from value-laden judgements of significance in historiography.
Fifth lesson
When we choose what to explain, the choice must be based on values. This does not mean that the explanandum must be morally significant or anything like that. It simply means that when we explain some X, we want to know when some alternative Y, considered interesting for some reason, would have been the case instead of X.
Making the choice of explanandum and the alternative Y transparent is necessary in order to understand why historiography explain certain things and not others. It is also highly useful in evaluating which futures are desirable/undesirable or at least interesting. Given that the historiographical explanations reflect our values in the choice of explanandum, investigating historiographical explanations can reveal a lot about how to think about interesting futures.
Moreover, a historiographical explanation tracks down counterfactual histories that would have led to Y instead of X. There are two ways of thinking about the structural similarities between historiographical explanation and scenario-work when it comes to explanandum.
First, we can take both X and Y be features of two different possible futures. Then we could track down when X or Y is the case. The choice of the futures on which we focus, i.e. the choice of X and Y should both be interesting futures (good or bad) and they should be alternatives to each other. We should reflect on how the interesting futures X and Y are chosen.
Secondly, we can consider our present condition as X and a future condition as Y. We could ask what kind of future could be an interesting alternative to our current situation and then attempt to track down what kind of changes could lead to Y. This way of thinking could be preferred on the basis that X’s significance is easier to evaluate when it is present in our world. In essence, we would be reflecting how something of significance could be changed in the future (or preserved, if X is desirable).
Finally, it must be underlined that the targets of study (i.e. what to explain and what to estimate) cannot be chosen without value-based judgements. In historiography, we cannot explain everything that has been causally relevant in the past since there is too many causal chains to begin with. In scenario-work, it is impossible to focus only on scenarios that are outcomes of causally relevant features of the current world because everything is. Moreover, we make choices that affect the future. These choices shape the futures and if we attempted to simply focus on the causal factors currently on work, we would ignore ourselves as factors. While the presence of values is more explicitly present in the futures studies than in historiography, it is still of utmost importance to notice the inevitable role that values play in historiography. An investigation on academic and non-academic debates on historiographical studies and their values can teach scenario-work indirectly how values are debated and what the limits and problems in the debates are.
We can see a repeating pattern here. In a previous post, we noted that mistakes and errors in futures studies and historiography are not qualitatively different. Now we have noted that the role of values is no less central in historiography than futures studies. A closer look on disciplines reveals that their borders do not lie where they were supposed to.
References
Beatty, John (2017). “Narrative possibility and narrative explanation”. Studies in History and Philosophy of Science Part A 62. 31-41.
Ben-Menahem, Yemima (2016). “If Counterfactuals Were Excluded from Historical Reasoning”. Journal of the Philosophy of History 10 (3). 370-381.
Collingwood, R. G. (1946). The Idea of History. Oxford: Clarendon Press.
D’Oro, Giuseppina (2009) “Historical Understanding”. In Tucker (ed.) A Companion to the Philosophy of History and Historiography. Wiley-Blackwell.
Dray, W. H. (1957). Laws and explanations in Histor. London: Oxford University Press.
Ereshefsky, Marc & Turner, Derek (2019). “Historicity and explanation”. Studies in History and Philosophy of Science Part A.
Garfinkel, A. (1981). Forms of Explanation. Yale University Press.
Glennan, Stuart (2010). “Ephemeral Mechanisms and Historical Explanation”. Erkenntnis 72 (2). 251-266.
Guldi, Jo (2019). “The Modern Paradigms of Explanation: Significance, Agency, and Writing History in the Era of Climate Change”. Isis 110 (2). 346-353.
Hempel, Carl G. (1942). “The function of general laws in history”. Journal of Philosophy 39 (2). 35-48.
Hitchcock, Christopher & Woodward, James (2003). “Explanatory generalizations, part II: Plumbing explanatory depth.” Noûs 37 (2). 181–199.
Lebow, R. N. & Stein, J. G. 1996. “Back to the Past: Counterfactuals and the Cuban Missile Crisis.” In Counterfactual Thought Experiments in World Politics. In Tetlock, P. & Belkin, A. (eds.). 119–48. Princeton University Press.
Lipton, P. (1990). ”Contrastive Explanations”. In Explanation and its Limits. Knowles (ed.). Cambridge University Press. 247–266.
Psillos, Stathis (2011). “Moving Molecules Above the Scientific Horizon: On Perrin’s Case for Realism”. Journal for General Philosophy of Science. 42 (2). 339–363.
Renn, J. (2005). “Einstein’s invention of Brownian motion”. Annalen der Physik 14. 23–37.
Tetlock, P., and Belkin, A. (1996). Counterfactual Thought Experiments in World Politics: Logical, Methodological and Psychological Perspectives. Princeton University Press.
Tucker, Aviezer (2004). Our Knowledge of the Past: A Philosophy of Historiography. Cambridge University Press.
van Fraassen, B. (1980). The Scientific Image. Oxford: Oxford University Press.
Virmajoki, Veli (2018). “Could Science Be Interestingly Different?”. Journal of the Philosophy of History.
Virmajoki, Veli (2019). Cementing Science. Understanding Science through Its Development.
Weber, Max (1949). On the Methodology of Social Sciences. (Translated and edited by Shils, Edward A. and Finch, Henry A). The Free Press of Glencoe.
Woodward, James (2003). Making Things Happen. A Theory of Causal Explanations. Oxford University Press.
[1] Woodward’s account defines causal and explanatory relations as holding between variables (i.e. the value explanandum-variable depends on the value of explanans-variable). I simplify the discussion and therefore I do not write in terms of variables. How the notions here can be written in such terms see [omitted for review].
[2] The idea of contrastive explanations is not a novelty of Woodward’s account. There has been a lot of discussion about contrastive explanations (see e.g. Hart and Honoré 1959; van Fraassen 1980; Garfinkel 1981 and Lipton 1990).
[3] The idea that historiographical explanations are based on difference-making is not a novelty by any means, see Weber (1949) and Ben-Menahem (2016).
[4] See e.g. Renn (2005) and Psillos (2011) discussing this topic.
[5] Tucker, Our Knowledge of the Past, 187.
[6] I have discussed the issue in Virmajoki 2019, Chapter 4.
2 thoughts on “Understanding Explanation and Its Relevance for the Future”