Put Your Money Where Your Mouth Is (PYMOWYM) principle: All research on historical phenomena both assumes and makes conclusions about the workings of the relevant phenomena. The commitments limit and shape the possible structures of the phenomena in the future.
How can a historian (or a historiographical community) identify a chunk of actual history as prima facie relevant to an outcome? It seems that there must be some basic framework in use that tells us what kinds of factors that might be connected to each other. For example, the release of A Trip to the Moon most likely had nothing to do with the observation of the gravitational deflection whereas the theoretical background in physics and international politics are serious explanatory candidates. This might sound trivial because of course no one would suggest that the movie belongs to the history of the observation. However, this seeming triviality is only a result of our enormous confidence that we understand many things about the workings of the world. But the question remains: how can we know this?
Imre Lakatos notoriously argued that we should rationally reconstruct histories of science in terms of philosophical accounts of scientific development. Lakatos’s idea is somewhat ridiculed these days because it is often summarized by its claim that the best historiography of science deems most of the history of science rational. However, it is a bit unfair to hint that Lakatos was building a grotesque philosophy of history in order to save the rationality of science by any means possible. It is more charitable to notice that Lakatos was aware of something that many of his colleagues were not, i.e. that all historiography requires some framework and that the merits of the frameworks cannot be assessed simply by comparing them to how the past really went because we have no direct access to the explanatory order of the past. Kuukkanen has pointed out that
“All history writing includes a theoretical basis of some kind and is indeed normative, implying selectivity and emphases on what is important and explanatory in history. [–]. [Lakatos unlike others] explicitly accepted that the same historiographical data can be brought into several alternative accounts, and he formulated some viable options using schemes and ‘philosophies of science’ of his time.” (2017, 91.)
It is difficult to say whether we should consider Lakatosian framework-driven reconstructions as explanatory or not. Lakatos did not inject any explicit modal dimension to the idea of rational reconstructions – something we will come back to later – and therefore it does not fit any model that connects explanation with modality. However, the reconstructions clearly add something to the history, namely a coherent (in terms of a framework) sequencing of events. They also are some sorts of idealizations, as Lakatos was explicit about the fact that the actual history might diverge from the reconstructed one. These aspects of Lakatosian reconstructions imply that their goal is not merely to describe the past as well as possible but also to provide additional understanding. They sacrifice the perfect match with the facts for coherent sequencing. What’s going on here?
One option is to argue that it is possible to find the disagreeing facts (i.e. facts that should not be there according to a framework) only in relation to a framework in the same way as it is possible to identify shadows only against lighted areas. Only after a reconstruction in terms of a framework, we are able to judge the significance of facts outside the reconstruction. For example, consider a falsificationist historian of science. She reconstructs a historical development in terms of the formulation of a theory and subsequent tests that attempt to refuse the theory. She finds the steps involved in the theory formulation and a set of subsequent tests. Many of these tests are in accordance with the theory but some are not. The historian concludes out that the theory was accepted because it survived many serious tests but also points out that there were tests that, according to (naïve) falsifactionism, should have led to the refutation of the theory. The historical significance of the contradictory tests is revealed in relation to a framework-driven reconstruction. The historical account can, therefore, go beyond its own limits in determining the significance of historical facts without making any fact whatsoever significant. For example, some supposed tests can be so defective that they do not serve to be mentioned in the context of theory acceptance and thus lack any historical significance. While I find this line of reasoning somewhat compelling, it cannot be the final word. Even if it is true that we can separate significant and insignificant facts only on the basis of a framework, the question still remains why we are not satisfied with any listing of facts whatsoever as historical insight. If someone was to suggest that history of science proceeds because theories are made acceptable by interventions on our cognition by an evil demon, this historiography would be completely distorted even if it allowed us to identify facts that are against its main assumption. A framework cannot be a mere magnet that captures facts from the darkness of history.
It seems much more fruitful to say that the frameworks enable us to make sense of the history.Even though different frameworks assume different workings of the world, they all make sense: the historical developments that are in the accordance of a framework make sense. Lakatos’s initial examples of frameworks of rationality from philosophy of science are all logical and prima facie plausible. The evil demon framework is not understandable or plausible and a historical development that accords it would presumably not make sense. A mere listing of facts does not make those facts understandable; neither does an account that captures the fact with some obscure framework. However, a plausible and logical framework that captures historical facts does make them understandable. This is why we have to sacrifice the perfect match between a historical account and historical facts. If we wish to understand history and if we do not have a perfect framework, we have to be settled for reconstructions that do not match the facts perfectly. We have to admit that not all things in history make sense to us; and it might be that some things just do not make sense simpliciter.
Notice that the frameworks cannot be logical and plausible only in relation to some particular historical context, but they have to be logical and plausible to us in general. History has to make sense to us. Moreover, to assume that we kind find a context-dependent logicality and plausibility by studying a historical sequence is to assume that the relevant sequences can be identified directly, without frameworks. Our initial problem was, exactly, that this seems impossible. Of course, we can test in a Lakatosian way how well a framework fits with the facts of a certain historical context by estimating how many significant facts it ignores and, in this sense, we can judge how well the framework captures the contextual parameters. We may prefer one framework over another due to its better ability to accommodate significant facts. However, this does not mean that the framework make sense only in a certain context. Rather, it means that we can make especially good sense of a context in terms of the framework.
We arrive in the one-dimensional (i.e. without modal or axiological loading) PYMOWYM principle:
To select some set of historical facts or a sequence of events as significant or relevant requires a logical and plausible framework that selects and orders those facts and or events. Making sense of history requires frameworks. The logicality and plausibility of these frameworks is not context-dependent but general for us. Therefore, a possible future set of facts or sequence of events can only make sense as long as it is in accordance with a framework. Any effort to make sense of the past generates a set of possible futures that make sense. Outside this set, there are disagreeing future details and incomprehensible futures.
However, there is a complication to this issue. Perhaps one could argue that different frameworks make sense of different contexts, and this means that there is no need to choose one. Perhaps the different degrees of applicability of a framework on different contexts is due to the underlying changes in the history itself. If we accept the plurality of frameworks for the purposes of historiography, we do not have to commit to the set of possible futures that one particular framework generates.
We have to distinguish between the weaker claim (a) that there might be different patterns of events in different historical contexts and that we can make sense of these in terms of different frameworks, and the stronger claim (b) that the patterns change and are rarely repeated and, therefore, the future requires a framework not applicable to past. When it comes to (a), we need to notice that PYMOWYN is formulated in terms of possible futures. Even if we do not commit to one framework and one pattern, our pluralistic selection of frameworks generates a set of possible futures and leaves other futures as incomprehensible. On the other hand, (b) amounts to saying that the future cannot be made sense of. Notice that this is also a commitment: it says that the set of possible futures does not contain any of the futures that the existing frameworks generate. It is not just there will be some anomalies to a pattern projected by a framework, rather, there will be no pattern at all that makes sense to us. While such a position surely is possible, it seems difficult to understand how one could consistently think that the past must make sense and thus reconstruct it in terms of frameworks and that the possible futures make no sense at all. If one’s sense-making forces on to abandon all sense at all, something is wrong with the person’s commitments.