Metaphysicians, by and large, aim to be scientifically respectable in their theorizing. To what extent do they succeed? That’s an excellent question. But before we can answer it, we need to answer a more basic question: What does it mean to be scientifically respectable in your metaphysical theorizing, anyway?In this important and original book, Andreas Hüttemann puts forward a novel way of thinking about this second question. He puts forward, in other words, a novel methodology for naturalistic metaphysics. He then uses this methodology to generate a number of substantive metaphysical results regarding laws of nature, causation, and fundamentality. Although both aspects of the book—the methodological commitments and the metaphysical conclusions—are innovative, in what follows I will focus mainly on the former. One of the key claims I will make is that this methodology can be developed and deployed in more ways than one might at first recognize. That is to say, one can find inspiration in key aspects of Hüttemann’s methodology without endorsing the details of his metaphysics. (Though one may, of course, wish to endorse his metaphysics too.)Hüttemann’s methodology has two components: (1) a view about how to derive metaphysical commitments from scientific practice; and (2) a view about how to keep one’s metaphysics suitably minimal. The first component involves deploying the following schema:Practice to Metaphysics Schema1. Scientific practice includes X.2. The best explanation of the success of scientific practice, given that it includes X, is that metaphysical thesis M is true.3. So, M is true.As a particularly straightforward example, in chapter 1 Hüttemann argues that scientific practice includes using law statements to explain, confirm, and predict the behavior of systems. And the best explanation of the success of that practice is that systems exist. So, he concludes, systems exist. (Here systems are understood in contrast to the more traditional metaphysical category of substances. Systems can include everything from space-times to economies to hydrogen atoms.)So far, so good. As far as I am aware, the Practice to Metaphysics Schema has seen little if any discussion in the contemporary literature. (Perhaps the closest antecedents are indispensability arguments in philosophy of math a la Quine and Putnam.) The vast majority of discussions of naturalistic metaphysics have focused instead on the extent to which metaphysicians are respecting the content of our best scientific theories. But, as Hüttemann points out in the introduction, there are reasons to be skeptical of this focus. Perhaps most importantly, the content of our best scientific theories often undergoes dramatic changes when new theories are developed. Scientific practice, at least at a certain level of abstraction, appears to be more stable, maintaining the same general features even across periods of dramatic theory change.All of this is to say that the Practice to Metaphysics Schema seems like a fairly compelling route by which to pursue a naturalistic metaphysics. Indeed, I have trouble seeing how someone who calls themselves a naturalist could resist the conclusions generated by this schema, unless they have a good argument against relying on inference to the best explanation in general. At the same time, it seems pretty unlikely that the Practice to Metaphysics Schema as described above will place significant constraints on our metaphysical theorizing. This is because nothing has been said so far about how we are supposed to make judgments about which explanation is best. And as readers familiar with the literature are undoubtedly aware, there are many different ways of understanding what makes one explanation better than another.For this reason, it looks to me like the second component of Hüttemann’s methodology—the claim about minimality—is doing quite a lot of the work in generating the substantive metaphysical claims that he puts forward. And here, things do not seem quite so straightforward.Here is a natural way of understanding how minimality fits into the methodology after reading the introduction and first few chapters of the book. First one deploys the Practice to Metaphysics Schema. This is what makes one’s metaphysics a metaphysics of scientific practice. Then, one then applies the following criterion:This is what makes one’s metaphysics of scientific practice a minimal metaphysics of scientific practice.Here is an example that supports this reading. Later in chapter 1 (after the bit about systems), Hüttemann argues that nomological modality can be analyzed in terms of invariance relations. What it means to say that some pattern holds as a matter of nomological necessity, for instance, is just that it would continue to hold under various kinds of changes or interventions. (I’m rephrasing Hüttemann somewhat here, but I don’t think this matters for the point that I’m making.)Now, anyone who has spent time thinking about the metaphysics of laws, and certainly those of us who’ve been in the trenches of the Humean/anti-Humean debate will read this and think: well, that’s all fine, but now I want to know more about invariance. Hüttemann thinks we should resist this desire. He writes:As I read this passage, Hüttemann is saying: we have deployed the Practice to Metaphysics Schema, and it has told us to understand nomological modality in terms of invariance relations, but it hasn’t required us to take a stand on any further analysis of those relations. So following the Minimality Criterion, we should refuse to take such a stand. This isn’t to say that there is no further analysis of invariance relations or that they are primitive. Instead, a minimal metaphysics of scientific practice insists on remaining quiet regarding any further analysis of invariance relations.Insofar as this is a correct understanding of the role of minimality in Hüttemann’s methodology, it is clearly a further commitment that is independent of one’s commitment to the Practice to Metaphysics Schema. One could have a metaphysics of scientific practice, in Hüttemann’s sense, without endorsing a minimal metaphysics of scientific practice. And while I think someone who thinks of themselves as a naturalist will have a hard time resisting the Practice to Metaphysics Schema, I also think that they should feel free to resist the Minimality Criterion. One can take science to be a key, or even the best, source of information that we have about what the world is like while also thinking that there are some questions that science leaves open, and that it is fair game for metaphysicians to try to answer those open questions.The reader should note, however, that I am not entirely confident that the Minimality Criterion is the right way to think about the role of minimality in Hüttemann’s methodology. That things are more nuanced becomes clear in chapters 6–8, where Hüttemann considers the role of reductive explanation in scientific practice. (Hüttemann discuss a number of types of reductive explanation, but, as one example, consider the attempt to reductively explain the properties and behavior of thermodynamic systems in terms of the properties and behavior of corresponding statistical mechanical systems.)The key question that Hüttemann asks in these later chapters is which of the following three views best explains the success of scientific practice, given that scientific practice involves reductive explanation.I could write a whole second review on how to think about Ontologically Neutral Monism and the question of whether it is really an ontological or an epistemic thesis, as well as the details of why Hüttemann thinks that Ontologically Neutral Monism is capable of explaining the reductive aspects of scientific practice. But, for my purposes here, let’s just grant his argument that Ontologically Neutral Monism can do the relevant explanatory work, and let’s focus on his further claim that Ontologically Neutral Monism is a better explanation than either Physical Foundationalism or Physical Eliminativism. Here is what he writes in defense of this latter claim:This passage suggests that minimality is not a separate, secondary criterion but rather a key part of the Practice to Metaphysics Schema—minimality plays an important role in determining which explanation counts as the best explanation. The passage suggests, in other words, that we adopt the following criterion:According to this criterion, Ontologically Neutral Monism is a better explanation than either of the two competitors described above.The key point I want to make here, however, is that it still isn’t clear to me that there is any especially compelling reason for thinking that those committed to the Practice to Metaphysics Schema need to understand inference to the best explanation in terms of minimality. The Minimal-Explanation Criterion isn’t unreasonable, but it is also far from obvious. Someone could also reasonably argue that what is important in an explanation is that there is a certain degree of completeness in the information that it provides about the world—that it doesn’t leave obvious questions about the entities that it posits without an answer. One might, in other words, endorse the following criterion:And someone who endorses this criterion should presumably think that both Physical Foundationalism and Physical Eliminativism are better explanations than Ontologically Neutral Monism.So, to recap, although there are different ways of understanding the role that minimality plays in Hüttemann’s methodology. On either of the views discussed above, at least, minimality seems to be a commitment that a naturalist could well choose to resist.The foregoing remarks may sound critical, but I don’t think they should be read that way. The key thing to take away from all this is that a metaphysics of scientific practice, by which I mean a metaphysics that relies crucially on Hüttemann’s Practice to Metaphysics Schema, can be developed in a range of different ways that, almost certainly, will give rise to many different metaphysical views. And this speaks to the fact that this book can—and should!—have both a broad and long-lasting impact. Even those who disagree with the details of Hüttemann’s metaphysical views or his arguments for those views should pay close attention to the innovative methodology he is proposing.

中文翻译：

---

科学实践的最小形而上学

总的来说，形而上学家的目标是在他们的理论中获得科学上的尊重。他们在多大程度上成功了？这是一个很好的问题。但在我们回答之前，我们需要回答一个更基本的问题：无论如何，在你的形而上学理论中受到科学尊重意味着什么？在这本重要的原创书中，Andreas Hüttemann 提出了一种新颖的思考方式问题。换句话说，他提出了一种新的自然主义形而上学方法论。然后，他使用这种方法产生了许多关于自然法则、因果关系和根本性的实质性形而上学结果。尽管本书的两个方面——方法论的承诺和形而上学的结论——都是创新的，但在下文中我将主要关注前者。我要提出的一个关键主张是，可以通过比人们最初认识到的更多的方式开发和部署这种方法。也就是说，人们可以在 Hüttemann 的方法论的关键方面找到灵感，而无需认可他的形而上学的细节。（当然，人们也可能希望认可他的形而上学。）Hüttemann 的方法论有两个组成部分：（1）关于如何从科学实践中得出形而上学承诺的观点；(2) 关于如何保持形而上学适当最小化的观点。第一个组件涉及部署以下模式：实践到形而上学模式 1。科学实践包括 X.2。科学实践成功的最好解释，假设它包括 X，是形而上学论点 M 是真的。 3．所以，M为真。作为一个特别直接的例子，在第 1 章中，Hüttemann 认为科学实践包括使用定律陈述来解释、确认和预测系统的行为。对这种做法成功的最好解释是系统的存在。因此，他得出结论，系统是存在的。（这里的系统被理解为与更传统的形而上学物质类别形成对比。系统可以包括从时空到经济再到氢原子的一切。）到目前为止，还不错。据我所知，形而上学图式的实践在当代文学中几乎没有讨论过。（也许最接近的前因是数学哲学中必不可少的论点，如奎因和普特南。) 绝大多数关于自然主义形而上学的讨论都集中在形而上学家尊重我们最好的科学理论内容的程度上。但是，正如 Hüttemann 在引言中指出的那样，我们有理由对这一重点持怀疑态度。也许最重要的是，当新理论发展时，我们最好的科学理论的内容往往会发生巨大的变化。科学实践，至少在一定的抽象层次上，似乎更稳定，即使在理论发生巨大变化的时期也能保持相同的一般特征。追求自然主义的形而上学。的确，我很难理解自称博物学家的人如何能抵制这种图式产生的结论，除非他们有很好的论据反对依赖一般最佳解释的推理。同时，上述形而上学图式的实践似乎不太可能对我们的形而上学理论化施加重大限制。这是因为到目前为止，我们还没有说过我们应该如何判断哪种解释是最好的。熟悉文献的读者无疑都知道，有许多不同的方式可以理解是什么让一种解释比另一种解释更好。出于这个原因，在我看来，Hüttemann 方法论的第二个组成部分——关于极简主义的主张——在产生他提出的实质性形而上学主张方面做了很多工作。在这里，事情似乎并不那么简单。在阅读本书的介绍和前几章后，这是一种理解最小化如何融入方法论的自然方式。第一个将实践部署到形而上学模式。这就是使一个人的形而上学成为科学实践的形而上学的原因。然后，然后应用以下标准：这就是使一个人的科学实践形而上学成为科学实践的最低限度形而上学的原因。这里有一个支持这种阅读的例子。在第 1 章的后面（在关于系统的部分之后），Hüttemann 认为法则模态可以根据不变关系进行分析。例如，说某种模式作为法则必然性而成立的意思只是它会在各种变化或干预下继续成立。（我在这里稍微改写了 Hüttemann，但我认为这对我要表达的观点来说并不重要。）现在，任何花时间思考法律形而上学的人，当然还有我们这些曾经在 Humean/anti-Humean 辩论的战壕中会读到这篇文章并思考：好吧，这很好，但现在我想了解更多关于不变性的知识。Hüttemann 认为我们应该抵制这种欲望。他写道：当我读到这篇文章时，Hüttemann 在说：我们已经将实践部署到形而上学模式，它告诉我们根据不变关系来理解法则模态，但它没有要求我们对这些关系的任何进一步分析采取立场。因此，根据最小标准，我们应该拒绝采取这样的立场。这并不是说没有对不变关系的进一步分析或者它们是原始的。相反，科学实践的最小形而上学坚持对不变关系的任何进一步分析保持沉默。只要这是对最小化在 Hüttemann 方法论中的作用的正确理解，它显然是一种独立于对不变性关系的承诺的进一步承诺。实践形而上学模式。在 Hüttemann 的意义上，一个人可以拥有科学实践的形而上学，而无需认可科学实践的最低限度的形而上学。虽然我认为自认为是博物学家的人很难抵制形而上学模式的实践，但我也认为他们应该可以自由地抵制最小性标准。人们可以将科学视为我们掌握的关于世界是什么样的信息的关键甚至最佳来源，同时也认为科学留下了一些未解决的问题，并且形而上学家试图解决这些问题是公平的游戏回答那些悬而未决的问题。然而，读者应该注意，我并不完全相信最小性标准是思考最小性在 Hüttemann 方法论中的作用的正确方法。在第 6-8 章中，事情变得更加微妙，Hüttemann 在其中考虑了还原解释在科学实践中的作用。（Hüttemann 讨论了许多类型的还原解释，但是，作为一个例子，考虑根据相应的统计力学系统的属性和行为来还原解释热力学系统的属性和行为的尝试。）Hüttemann 提出的关键问题在后面的章节中，以下三种观点中哪一种最能解释科学实践的成功，因为科学实践涉及还原解释。我可以写一整篇关于如何思考本体中立一元论及其是否真的存在的问题本体论或认知论题，以及为什么 Hüttemann 认为本体中立一元论能够解释科学实践的还原方面的细节。但是，为了我的目的，让我们只承认他的论点，即本体中立一元论可以做相关的解释工作，让我们关注他进一步的主张，即本体论中立一元论是比物理基础主义或物理消除主义更好的解释。以下是他为后一种主张辩护所写的内容： 这段话表明，最小化不是一个单独的、次要的标准，而是形而上学图式实践的关键部分——最小化在确定哪种解释被视为最佳解释方面起着重要作用. 换句话说，这篇文章建议我们采用以下标准：根据这个标准，本体论中立的一元论是比上述两个竞争者中的任何一个更好的解释。然而，我想在这里提出的关键点是，是我仍然不清楚是否有任何特别令人信服的理由认为那些致力于形而上学模式实践的人需要理解对最小化最佳解释的推理。最小解释标准并非不合理，但也远非显而易见。有人也可以合理地争辩说，解释中重要的是它提供的关于世界的信息有一定程度的完整性——它不会留下关于它所提出的实体的明显问题而没有答案。换句话说，人们可能会认可以下标准：并且认可这一标准的人应该认为物理基础主义和物理取消主义都是比本体论中立一元论更好的解释。所以，回顾一下，尽管有不同的方式来理解极简主义在 Hüttemann 的方法论中所扮演的角色。至少，根据上面讨论的任何一种观点，极简主义似乎是博物学家很可能选择抵制的承诺。上述评论可能听起来很批评，但我认为不应该那样理解。摆脱这一切的关键是科学实践的形而上学，我指的是一种主要依赖于 Hüttemann 的形而上学图式实践的形而上学，可以以一系列不同的方式发展，几乎肯定会产生许多不同的形而上学观点。这说明这本书可以——而且应该！——产生广泛而持久的影响。