Bryant’s Epistemology for Scientific Metaphysics

Intro

In response to Chakravartty’s defense of a scientific metaphysics, Amanda Bryant believes that there is a need for an epistemological infrastructure for this field of philosophy. She then goes through options for constructing this infrastructure. First, she gives the option of using an empiricist epistemology, saying “we should tie metaphysics to science because the road to knowledge is empirical, and because science seems to be a paragon of empirical investigation.” However, this would not be a great option, as mixing metaphysics and scientific evidence would not call for the need of scientific evidence. Also, science isn’t always purely empirical. Bryant uses the examples of evolutionary biology, string theory, and cosmology to support her point. 

Another option she goes through is one in which scientism is the only way of knowing: “if it were the case that science should guide all inquiry, then of course it should follow that it should guide metaphysics.” However, this view quickly becomes difficult to defend, as it is far too strong. Just as we would expect a philosopher to appeal to the sciences to prove their points regarding the nature of reality, we would not expect a philosopher to appeal to science when investigating subjective things, such as justice, beauty or morality. I did, however, take issue with the idea that beauty is not something that can be scientifically proven, as there are ratios and aesthetic patterns that have proven attractive to the human eye. Because of the flaws of these two prima facie options, Bryant believes that a weaker epistemic infrastructure can do the job of supporting naturalistic prescriptions and establishing scientific metaphysics as the better option. 

Scientific Metaphysics

Bryant then begins to explain what she means when she says “scientific metaphysics.” She uses its contrast, “free range metaphysics,” which is only nominally constrained by science and focuses on “simplicity, consistency, intuitive plausibility, and explanatory power” (4) to highlight the fact that scientific metaphysics is closely associated with science. Scientific metaphysics is, in every way, engaged with science. Basically, Bryant believes scientific metaphysics to allow the philosopher to behave like a scientist, incorporating scientific data into metaphysical theories, drawing philosophical conclusions on the basis of scientific data, and other things. Any metaphysics that does some of these things that Bryant prescribes is a scientific metaphysics. Bryant believes that, while there are numerous ways that people can perceive scientific metaphysics, the common ground they can all agree on is that metaphysical speculation should be reigned in (4).

Epistemic Principles

Bryant then goes into the epistemic principles of a scientific metaphysics. It is at this point where  there could be a split in how Bryant’s scientific metaphysics could be perceived. Much of science is done in an exploratory manner. In order to understand the world, scientists need to be exposed to everything that is in this world, even if they do not officially understand it. However, the principles that Bryant lays out are not conducive to discovering new things. Instead, they are focused on justifying, and certifying, the things that scientists have already been exposed to. Her goal for scientific metaphysics is a “convergence towards a theory consilient with and strongly confirmed by the best available evidence at the time” (5). This is where Bryant and Chakravartty most strongly agree. While scientists might not understand the nature of a thing, dispositions allow them to understand what a thing does, which is better than having nothing at all.

            In order to achieve this goal, Bryant believes that science needs theoretical constraints. Theories should be adequate with, and be able to explain, the data. She also acknowledges that science sometimes requires that the theories need to be in line with political or ideological goals, which does not seem very scientific, at least to me.

            Bryant posits that these constraints fall on a spectrum of strength, with data being the affecting variable. The reason being, the more data there is, the less theoretical content there will be. Having a robust theory with fewer atomic propositions (less permissive) will lead scientists to develop stronger theories. Bryant uses the metaphor of a bouncer at a club, saying that a club that lets anybody in is not a good club. Basically, she says that by limiting the scope of inquiry, science can develop stronger theories. However, she does not claim that these more constrained theories are superior. She acknowledges that extremely constrained theories “cannot say much at all” (7). Bryant just wants there to be limits for what a theory should include; while metaphysical claims should be allowed, science should lean more towards aligning with the available data.

            Bryant’s gripe with permissive theories is that they have very little explanatory power. There is too much going on. Contrast that with a robust theory: given the fewer things going on, the theory can focus on its most relevant aspects. Bryant does allow that simply being robust is not a good reason to adopt a theoretical constraint. Neither should we give up on a constraint only because it is permissive. This stipulation brings up an important ethical question: if the selection of theoretical constraints should be well-motivated, how can the scientific community ensure that the motivations they have are good ones? This issue is immediately confronted when Bryant says that a theory that allows “too many claims for the purposes of explanation” is not strong enough. That seems to be a dangerous proposition. There should be room for every kind of reaction — the good ones should get the most attention.

Constraint Principles

Bryant posits three constraint principles: weak, moderate, and strong. Explaining the three theories, Bryant says “the weak principle makes robustly constrained theories preferable ceteris paribus, the moderate principle makes them preferable full-stop; the strong principle makes robust constraint a necessary condition of what I call epistemic adequacy” (8). In order to be epistemically adequate, people should be able to rationally adopt a policy concerning the theory or its component parts. Bryant explains that this is why her scientific metaphysics would not be conducive to discovery. Regarding justification, Bryant states that robust constraints might lead to the discrediting of some theories. As the amount of available data grows, the theory might get disconfirmed. Due to this fact, Bryant worries that constraint might not be the best way to go about justifying theories. This makes no sense to me, as justifying a false theory seems to be an ethical issue in an of itself. Thankfully, Bryant does want to allow disconfirmation to be a part of the process, which takes care of that issue. 

Epistemic Principles Defended

Much like Okham’s Razor, Bryant believes that by limiting the theoretical contents in a theory, the more likely the theory is to be true.

Argument for Statistical Likeliness

P1) Ceteris paribus, relatively simple theories are statistically likelier to be true than their more complex rivals

P2) Ceteris paribus, statistically likelier theories are epistemically preferable

P3) Robustly constrained theories are relatively simple

P4) Robustly constrained theories are statistically likelier to be true than their more complex rivals

C) Ceteris paribus, robustly constrained theories are epistemically preferable

Bryant understands that there are trade-offs for her theory. While there are many desires for a theory, “it is better to have a descriptive and explanatory theory this is only somewhat likely than to have a theory that is not particularly descriptive or explanatory, but a good deal likelier” (12). Instead, the desire is to have a principle that leads us to align with the theory that is simpler, all other things being equal.

Argument from Agreement

P1) Ceteris paribus, relatively simple theories are more conducive to agreement

P2) Ceteris paribus, theories that are more conducive to agreement are epistemically preferable

P3) Robustly constrained theories are relatively simple

C) Ceteris paribus, robustly constrained theories are epistemically preferable

By following this argument, Bryant sees two valuable outcomes: greater statistical likeliness and greater consensus. Given that these values are important for epistemic aims, it follows that robustly constrained theories should be preferred, ceteris paribus.

Argument from falsehood avoidance

P1) The more content we exclude from a theory and the less we permit, the more likely we are to avoid substantial falsity

P2) Ceteris paribus, theories that are more likely to avoid substantial falsity are epistemically preferable 

P3) Robustly constrained theories exclude relatively many putative theoretical contents and permit relatively few

C) Ceteris paribus, robustly constrained theories are epistemically preferable  

By saying less, one is less likely to say something false. Theories aim to acquaint their audiences with facts, and it is more efficient to randomly state things in the hope that they will turn out to be true.

Argument from methodological expediency

P1) Theories that better target the relevant facts are epistemically preferable 

P2) Ceteris paribus, robustly constrained theories better target the relevant facts

C) Ceteris paribus, robustly constrained theories are epistemically preferable

By adopting this argument, this allows scientists to more effectively target facts. However, not looking at certain things does not in and of itself lead to more efficient targeting. Bryant uses the metaphor of shining a flashlight in a dark room to illustrate her point.

Scientific Metaphysics and Robust Constraint

By aligning itself more with science, scientific metaphysics naturally adopts more robust constraints. These constraints do not rule out crank hypotheses, but that does not mean it encourages them. Instead, they are simply not paid any attention, or at leas they should not. It also discourages science to engage with claims that cannot be empirically proven. Now that scientific metaphysics is robustly constrained, there is now an epistemic infrastructure that allows us to properly pursue it.

Leave a Reply