This modification provides a great addition to the current body of scientonomic knowledge. Once we accept that the current zeroth law is a tautology, it becomes clear that we need a new law explaining the mechanism of compatibility, i.e. a law stating how a pair of elements comes to be considered compatible. Fraser and Sarwar's formulation, I believe, succeed in doing precisely that.
In its essence, the law mimics the current formulation of the second law, by connecting assessment outcomes with their respective courses of events. Because the law forbids a number of logically conceivable scenarios, it is evident that the law is not a tautology. Thus, it is provides a great improvement over current understanding of compatibility.
Thus I believe the modification is to be accepted.
I agree that since this law is non-tautological, it is a significant improvement over the previous understanding of the way compatibility works in scientific change. Thereofore, I recommend that this modification should be accepted.
I believe this modification should be accepted. It acknowledges that there are compatibility assessments done by epistemic agents, and knowledge of compatibility is not assumed for the agent, but is something arrived at by the agent. It also allows for compatibility assessments to change over time in the face of new information, e.g. figuring out there is a contradiction between two theories potentially long after they have been proposed. Overall, this new law of compatibility is a substantial addition to the theory of scientific change that allows for a diachronic study of compatibility.
The zeroth law of scientific change; the law of compatibility, is flawed in several respects, which Fraser and Sarwar have identified. Unlike other scientonomic laws, the old law of compatibility (or zeroth law) is stated from a static perspective, invoking a hypothetical moment in time. There are many reasons why this is problematic. One is that it does not allow for the possibility that, human cognitive abilities being limited, unrecognized incompatibilities might lurk, undetected, within the mosaic. A dynamic account of compatibility, couched in terms of changes to the mosaic made by agents, avoids this shortcoming and many others. Absent a definition of compatibility, it also appears to be tautological, lacking any empirical content. This is, Fraser and Sarwar argue, a serious shortcoming for an empirical theory of scientific change. Their paper recommended a set of changes. One change was a new definition of compatibility, which was accepted in 2020. Another change they advocated was the formulation of a new dynamic law of compatibility, which is currently at issue.
The main issue, to my mind, is whether a new law of compatibility is actually needed. Is assessment for compatibility with other elements of the mosaic really conceptually distinct from the process of assessment for theory acceptance, which is already covered by other scientonomic laws? Occam’s razor requires that we ought not to multiply theoretical entities unnecessarily. The new law of theory compatibility is patterned after the law of theory acceptance, which is already an accepted part of scientonomy. Isn’t the work of the proposed law of theory compatibility already done by the law of theory acceptance?
Fraser and Sarwar argue that theory compatibility is conceptually distinct from other stances, such as those pertaining to theory pursuit, scientificity, or, most notably theory acceptance. It is possible, they assert, for unaccepted theories to be assessed for their compatibility with the mosaic. However, they fail to offer or explore any specific examples of this. String theory is perhaps the sort of example that warrants such exploration. It is currently pursued but not accepted by the physics community. Physicists pursuing the development of this theory clearly do care about its compatibility with accepted theories like quantum mechanics and general relativity theory. It is unclear, however, whether they care about such compatibility as a distinct conceptual issue apart from their struggle to garner acceptance for the theory. Under the current method of theoretical physics, a theory must account for the explanatory successes of its predecessors in order to win acceptance. Supporters of string theory might wish to argue that it offers a more fundamental explanation of the successes of quantum mechanics and general relativity, in what they might claim are limited domains of application. So far as I can see, however, they would seek to make such a compatibility argument for the sake of garnering acceptance of the theory, rather than as a separate conceptual issue in its own right.
Are issues of compatibility conceptually separate from issues of theory acceptance when the theories in question deal with two differing domains of phenomena, as is the case, for example, when one considers the compatibility of string theory with evolutionary theory? Here too, it seems that the two are inextricably intertwined. Consider a hypothetical example in which string theory predicted that the nuclear reactions in the core of the sun could not be sustained for more than 2000 years. This would render it incompatible with accepted elements of evolutionary theory and geological theories which claim that the Earth is 4.6 billion years old and has supported life for at least 3.8 billion years. This incompatibility would clearly pose a serious bar to the acceptance of string theory. The fact that string theory and evolutionary theory are apparently compatible therefore simply tells us that evolutionary theory poses no bar to the acceptance of evolutionary theory. Again in this instance, there seems no need to consider compatibility and acceptance as two conceptually distinct issues.
Suppose we ask the question of compatibility regarding two theories that are currently both rejected, such as Aristotelian natural philosophy and Ptolemaic astrology. Historical agents such as Aristotelian scholastic philosophers accepted both theories, and considered them to be compatible elements of the same mosaic. For these agents, the issue of acceptance and compatibility would have been inseparable. But suppose we consider the issue from the point of view of a modern historian. For the historian, admittedly, the issue of the compatibility of the two theories might be considered conceptually separable from that of their acceptability. This might imply that they are conceptually separable, in principle, for all cases. Fraser and Sarwar did not make this argument.
If one accepts Fraser and Sarwar’s premise that theory compatibility and theory acceptance are, indeed conceptually distinct issues, then Fraser and Sarwar’s new law of theory compatibility otherwise seems a reasonable consequence. Fraser and Sarwar envision theory compatibility and theory acceptance as following parallel laws, and the new law of theory compatibility is patterned directly after the law of theory acceptance.
With strong reservations, I’m willing to endorse the acceptance of the law of theory compatibility since the dynamical approach to compatibility that it embodies seems superior to the old static approach. However, I believe that the issue of the conceptual separability of theory compatibility and theory acceptance, and thus the need for two parallel laws, remains an open question that warrants further investigation.
Enable comment auto-refresher
Hakob Barseghyan
Ameer Sarwar
Kye Palider
Paul Patton