A PRIORI AND A POSTERIORI METHODS FOR IDENTIFYING VERBAL DISPUTES IN METAPHYSICS John Fraiser December 2009 Attempts at defining a verbal dispute commonly depend on the following principle: (PVD) If linguistic frameworks are able to equivalently describe the same uncontroversial data, albeit in different ways, the respective parties are engaged in a verbal dispute. 1 Assuming that (PVD) is correct, we still need criteria for determining when (PVD) is satisfied. Methods for determining when (PVD) is satisfied are of two kinds: (1) a priori methods which require us to prove by some a priori principle that a particular debate is verbal; (2) a posteriori methods which require us to test a sufficient number of relevant conceptual scenarios in order to demonstrate that both frameworks have equivalent powers of description, thereby increasing the probability that we have a verbal dispute to a high enough degree to conclude that we most likely have a verbal dispute. As more scenarios are tested, the probability of a verbal dispute continues to increase. I argue that ontological superficialists operating under (PVD) have not given us a satisfactory a priori method and there is good reason to think that they cannot. This leaves us with a posteriori criteria 1 Since this is an interpretation of a rough principal expressed in different ways by several superficialists (and a few non-superficialists), my use of the term data should be understood very broadly. It will encompass physical data, mental data, metaphysical data, actual-world data, possible-world data, conceptual data, etc., so long as the data isn t a first-order part of the dispute. I will focus primarily on the presence of (PVD) in the approaches to verbal disputes set forth by Eli Hirsch and David Chalmers, but it is worth noting that Stephen Yablo and John Hawthorne both offer principles similar to (PVD) (though Hawthorne s is a principle for identifying substantive disputes). Yablo: Are there Xs? is moot iff hypotheses that presuppose Xs are systematically equivalent (modulo ) to hypotheses ( ) about how matters stand Xs aside (Yablo 2009, p. 524). Hawthorne: Let us say that a theorist x intensionally advances over theorist y when there is some true intension that x accepts that y does not (Hawthorne 2009, p. 220). 1
for determining mainstream metaphysical debates to be verbal disputes. I argue that a posteriori methods are more promising but they make most of the superficialist s judgments regarding verbal disputes premature and overconfident. Since, as I will argue, the default attitude toward metaphysical disputes should be to treat them as substantive unless we have sufficient reason to think that they are verbal, we should continue to engage the mainstream metaphysical debates. Definitions of Verbal Disputes There are several definitions of verbal disputes available, with Eli Hirsch s receiving perhaps the most attention. He defines a verbal dispute as a dispute in which, given the correct view of linguistic interpretation, each party will agree that the other party speaks truth in its own language (Hirsch 2009, p. 239). Elsewhere he terms a verbal dispute as: A dispute over the truth of S is verbal if one party uses S as definitionally equivalent to S 1, the other uses it as equivalent to S 2, and the parties agree on the truth of S 1 and S 2 (Hirsch 1993). In similar fashion, David Chalmers tells us that: A dispute over S is (broadly) verbal when for some term T in S, the parties disagree about the meaning of T, and the dispute over S arises wholly in virtue of this disagreement regarding T (Chalmers, ms.). Each of these is more or less the same definition, the central idea being that verbal disputes are not about truth but about the meaning of words. Were the meaning agreed upon, there would be nothing of substance left over to dispute. While these definitions are not without controversy, suppose we accept them. What will follow for the mainstream metaphysical disputes? Under these definitions, they will be verbal if what both sides are disagreeing over is the meaning of words and not about the facts of the matter. How will we know when the dispute arises wholly in virtue of this disagreement regarding T or when each party speaks truth in its own language? We have two options here: an a priori option and a posteriori option. On the a priori option the superficialist would need to give us a way of knowing that in principle the dispute is entirely a matter of 2
disagreement over the meaning of words and not over the facts. That is, we could see that just by looking at the logic of what the parties of the dispute are saying that there is nothing substantive going on. Our a posteriori option would require us to look at particular conceptual scenarios to which the competing frameworks relate and see if each is capable of providing equivalent descriptions of the undisputed data. Another way of characterizing these methods might be to say that the first is deductive, the second inductive. A Priori Methods for Identifying Verbal Disputes Superficialists tend to prefer a priori methods since they are less messy and promises more certainty about verbal disputes. But how do they go about showing that, in principle, nihilists and mereological sum-lovers aren t disputing over the facts of the matter? It certainly seems like they are. The sum-lover looks at a chair and says There are simples and a chair while the nihilist says, No, there are only simples. These are claims about facts, so why would superficialists think that this is just a dispute about the meaning of words like chair and not about facts? Hirsch employs a principle of interpretive charity (which I will abbreviate PC) as a way to know a priori that both sides are speaking truth. (IC) All else being equal, an interpretation s plausibility is proportionate to the interpreter s effort to make the assertions of a linguistic framework come out true or, at the least, as reasonable as possible (Hirsch 2002, p. 71). Hirsch concludes that out of this principle that we ought to see that in most of the mainstream metaphysical disputes each party is speaking the truth in its own language. Nihilists are speaking the truth in their own language and sum-lovers are speaking the truth in theirs. And if each side employs (IC), they will see that this is the case. Hirsch specifies further how this will work for the endurantism/perdurantism debate (though a parallel argument works just as well for the debate between nihilists and sum-lovers). According to Hirsch, endurantists speak E-English in which talk about objects being wholly present at a point in time comes out true in E-English and perdurantists speak P-English in which talk about objects having 3
temporal parts comes out true in P-English. For the endurantist/perdurantist debate, (IC) yields the translational key: (IC-P) In P-English a sentence of the form a has at time t a temporal part that is F has the same character as the E-English sentence a is F at time t (where F is a term that applies to an object at time t by virtue of how the object is at that time). Other forms of sentences in the language operate in the obvious ways (Hirsch 2009, p. 245). 2 It should be clear that (IC-P) depends upon (PVD). For both sides to speak the truth in their own language there can t be uncontroversial data which the perdurantist can describe in P-English that the endurantist can t describe in E-English. If there is, then the translational key won t work, and it won t turn out that they are both speaking truth in their own language. So how did Hirsch get from (IC) to (IC-P)? (IC) does not entail (IC-P), and Hirsch provides no transitional step. I take it that (IC) is widely uncontroversial, while (IC-P) is widely controversial. Suppose that endurantists accept (IC) and are committed to interpreting P-English sentences as charitably as possible. Why must this guarantee that P-English regarding temporal parts always have the same character as their E-English sentences regarding wholly present objects at t? (IC) falls short as an a priori principle for showing us that both sides are speaking truth in their own language. Still, I think that there are other disputes where (IC) can function as the principle we need to determine a priori that the dispute is verbal. I think this could be done pretty easily in cases where the disputing parties have simply switched their terms. Your friend suffers a head injury and consequently calls dogs cats and cats dogs. Supposing this is the only terminological change, if you employ (IC) you get a translational key from your language to your injured friend s. And so we can say that any dispute you might have with her over whether an Irish Setter is a dog or a cat is a priori a verbal dispute since your terms are entirely interchangeable with hers and you employ (IC). It isn t necessary to test all dog/cat phenomena to see if the language of both parties can account for it. On principle, your term dog is equivalent to her term cat and her term dog is equivalent to your term cat. sentences. 2 As Hirsch points out, there s a parallel translational key for perdurantists when they encounter E-English 4
Hirsch accepts that we can determine a priori that we have a verbal dispute when the terms are interchangeable, and it is clear that the head injury case satisfies this criterion. The meanings of the terms dog and cat have been switched wholesale and so you have the necessary translational key. As long as this switch is employed consistently we know that the disputes satisfy (PVD). It isn t clear, however, that the traditional metaphysical disputes are like this. While Hirsch holds that E-English and P-English sentences are a priori interchangeable, (IC) doesn t reveal that they are. So Hirsch leaves us with a gap between (IC) and (IC-P), an without an a priori method for showing that (PVD) is satisfied. Chalmers in the spirit of Hirsch s definition of a verbal dispute offers us a way to bridge (IC) and (IC-P)-like sentences. To apply this method to a dispute over a sentence S that is potentially verbal with respect to term T, one proceeds as follows. First: one bars the use (and the mention) of term T. Second: one tries to find a sentence S' in the newly restricted vocabulary such that the parties disagree nonverbally over S', and such that the disagreement over S' is part of the dispute over S. Third: If there is such an S', the dispute over S is not wholly verbal, or at least there is a substantive dispute in the vicinity. Fourth: If there is no such S', then the dispute over S is wholly verbal (Chalmers unpublished). Again, I think that it will work in scenarios where the terms are clearly interchangeable such as with your injured friend, but I don t think it will work for common metaphysical disputes. How do you a priori establish a universal negative like: There is no such sentence S' about which the parties disagree nonverbally and is part of the dispute over S? The possibility of false positives seems to be too great. I think we can see this more clearly in cases of scientific debate. 3 In the Newtonian-Einsteinian debate over the 3 I think we are at liberty to draw from scientific disputes to make conclusions about metaphysical disputes. I don t think this is a problem for two reasons: 1) we often employ some of the same criteria for theory choice in science as in metaphysics: explanatory power, simplicity, fruitfulness, theoretical fit, etc.; 2) in a dispute whether scientific or metaphysical if we deem it a verbal dispute, the same reasons we consider the dispute to be verbal in one discipline will also apply in the case of disputes in the other. At the very least, if the criteria for a verbal dispute in one is relevantly dissimilar to the dispute in the other, we have to have a reason to say so and I am not aware of any reason to think that there is relevant dissimilarity. See Hawthorne 2009 for a similar questioning of why conditions for verbal disputes in metaphysics wouldn't apply to various scientific disputes. Clearly there are metaphysical questions that are not scientific questions and there are scientific questions that are not metaphysical, but the disciplines of both metaphysics and science will often count progress the same way. One theory progresses over another when there is something that we need to explain or be able to talk about and there are things that one theory can explain in a way that another cannot. 5
definition of gravity, we would have to bar the term gravity. Next we would have to look for a new sentence that doesn t mention gravity, yields a non-verbal dispute, and is semantically equivalent to our disputed gravity sentence. If there isn t such a sentence then the original dispute was verbal. If there is, it was substantive. Now suppose a Newtonian and an Einsteinian are presented with a sentence that doesn t use gravity, and is semantically equivalent to the gravity sentence. Living between the years 1905 and 1916 when Einstein had proposed his general theory of relativity but its implications were not yet known, they might think that Chalmers method would reveal a verbal dispute. Even late into the 20 th century it was widely agreed that Newton s theory of gravity was a special case of Einstein s, since Einstein's theory is actually exactly equivalent to Newton's theory for very weak gravitational forces of normal slow-moving bodies. 4 As regards disputes over slow-moving bodies under weak gravitational forces, it is possible that we could bar talk about gravity and not find a sentence in which they have nonverbal disagreement, but the conclusion that the dispute is verbal would be premature. The possibility of misdiagnosis points to what seems to me to be a permanent worry with Chalmers method: there will inevitably be cases in which we falsely consider a substantive dispute to be a verbal dispute when we can t find a sentence of non-verbal disagreement even after barring the disputed term even though such a sentence is still possible. In cases where we cannot find S' we need a way of determining whether it is because of a failure of our knowledge or because of a genuine verbal dispute. Despite his claim to be the way forward in sorting out many philosophical disputes, I don t see how Chalmers method can deliver regularly as an a priori method for deciding them. The success of Chalmers method for identifying verbal disputes depends on our ability to find S' not on the existence of S' as he claims. Like Hirsch s (IC) I think the method will work for certain verbal disputes but they will be disputes which were quite obviously verbal to begin with. 5 4 As late as 1962, Thomas Kuhn claimed that this was the majority opinion among scientists (Kuhn 1996, p. 98). 5 I also have some tangential concerns about Chalmers method. I think in most cases of philosophical debate disputing parties will already have tried to get clarity about the linguistic differences between them. Vocabulary exhaustion will usually be the norm rather than the exception. Without employing Chalmers method 6
Chalmers method relies on finding an alternate linguistic framework into which we can convert the frameworks of the disputing parties. And like others, I question the ready abundance of alternate linguistic frameworks (Manley 2009, p. 13). To identify a substantive dispute we must reach bedrock, the point at which there are no S'. Chalmers notes that sometimes bedrock is reached because we have a substantive dispute and other times we will simply reach a point of vocabulary exhaustion. Chalmers offers us a crude method for deciding between substantive disputes and vocabulary exhaustion. Regarding the mind physicalism debate, he writes, A proponent could try suggesting that physicalism is bedrock, and that no further progress can be made by barring the term. But for the reasons given in the last section, this move is implausible. It seems that if two parties agree on all the truths in non- physicalism - involving language, then any further dispute over physicalism will be broadly verbal (Chalmers, unpublished. Emphasis added). Notice, however, that Chalmers way of distinguishing a substantive dispute from vocabulary exhaustion terminates in what is essentially (PVD). 6 Chalmers gives up on a translational key at this point, and opts for (PVD) as a way of identifying verbal disputes. It is important to notice, however, that Chalmers gives us no method for determining when (PVD) is satisfied. Rather, (PVD) is put forward as the final way to tell when disputing parties are engaged in a verbal dispute. As I have argued, the only clear a priori way of determining when (PVD) is satisfied is in cases of a mere switch of terminology, but the common metaphysical debates are not like this. In addition to a difference in terminology, there are, as John Hawthorne notes, attending non-linguistic attitudes that disputing parties already try to state there differences in other terms and will have reduced them as far as they can by their lights. This means that we should expect that alternate frameworks have already been explored. Furthermore, in cases in which an alternate language can be found, I wonder why Chalmers considers it a neutral party? If it has the distinct advantage of revealing whether we have a verbal or substantive dispute, why isn't it the superior language? Why do we not have a three-party dispute here with the neutral language winning out? Is the third language only good for settling disputes? Presumably we are turning to the third language because of their inadequacies in clarifying the dispute, but if the third language can accommodate the language of both of them and reveal the kind of dispute we have it should whether the dispute is verbal or substantive be the preferred language since it has a virtue or clarifying a dispute. 6 I think (PVD) is a more careful articulation of Chalmers statement. Each disputing parties could agree on the truth of non- physicalism -involving language even though the language of one party is incapable of describing it as well as the language of another party. So I think what we want to see in cases of a verbal dispute is not just agreement on the truth but the capacity of each language to equivalently describe the undisputed data. 7
separate the disputing parties (Hawthorne 2009, p. 216). 7 In cases of a vocabulary switch there will presumably be no implications for other beliefs or utterances beyond those employing the switched vocabulary. At present we do not have a strong a priori method for determining when (PVD) is satisfied. A Posteriori Methods for Identifying Verbal Disputes As we have seen, Hirsch s and Chalmers methods terminate in (PVD). But given the attending non-linguistic attitudes of disputing parties and the ethical implications of the differences in the linguistic frameworks, employing an a priori method for determining when (PVD) is satisfied does not seem to be the way forward. A posteriori methods don t have the promise of certainty that a priori methods do, but I think that they will be more effective in more complex cases of disputes such as our common metaphysical debates. (PVD) demands that we identify a verbal dispute as a case in which the languages of disputing parties are equally capable of organizing the rest of the data which isn t being disputed. Conversely, if the language of one party leaves them incapable of explaining or describing phenomena which the other can, then we have a substantive dispute between them. A posteriori methods will require testing of various conceptual scenarios to see if they are capable of equivalently accounting for the undisputed data. We might identify many scenarios in which the languages are equivalent, but this does not mean that we have determined that there is a verbal dispute. Unless we have tested all scenarios between them there remains the possibility that one language will be able to account for some data that another cannot. This means that a posteriori methods will only be able to deal in probabilities when determining whether (PVD) is satisfied. The more phenomena to which competing language frameworks relate, the more difficult it will be to identify whether the parties are engaged in a verbal dispute. Consider Karen Bennett s example of a dispute between the sorority girl and the purist over what constitutes a Martini. The sorority girl considers 7 Peter van Inwagen presents numerous ethical questions that relate to disputes about mereological sums, questions regarding when life begins, when death occurs, and organ transplants (van Inwagen 1990). Also, John Hawthorne claims that anecdotally he finds that perdurantists are more likely to think abortion disputes are shallow since perdurantism allows for multiple candidates for the referent of I of varying temporal lengths, while standard versions of endurantism repudiate multiple candidates (Hawthorne 2009, p. 216). 8
any mixed alcoholic beverage poured into a V-shaped glass to be a martini. The purist on the other hand denies this and considers a martini to be a drink made of vodka or gin, dry vermouth and perhaps an olive set (Bennett 2009, 50). This strikes us immediately as a case of verbal dispute. But why? I contend that it is not because we know a priori that it is verbal. This dispute is not like disputes involving vocabulary switch. I think this dispute strikes us as verbal for two reasons: (1) We are able to gather rather quickly that (PVD) is satisfied in the dispute and so it strikes us as verbal. They can both account for the same phenomena, just in different ways. Both sides agree on the undisputed relevant facts and are able to describe them under different linguistic frameworks. The dispute has a small range: alcoholic drinks. It will be easier to conclude that small range disputes are verbal compared to broad range disputes with attending non-linguistic attitudes and many implications for nonmetaphysical issues. The more phenomena to which competing language frameworks relate, the more difficult it will be to identify whether the parties are engaged in a verbal dispute. (2) I think we make connections between trivial disputes and verbal disputes. Verbal disputes seem trivial or at least less important than substantive disputes. While verbal disputes may make the dispute insignificant (for certain purposes), this is only one way for a dispute to be insignificant. Insignificance, however, does not entail a verbal dispute. The choice between linguistic frameworks can be trivial for several reasons, one of those being that the difference between them is only terminological, another being that the practical value of either choice is very low, such as a substantive dispute over what constitutes a martini. So, suspend your intuition for a moment, and suppose it could be shown that by calling sweet alcoholic fruit drinks poured into martini glasses "martinis" the sorority girl's language is left without a way to account for certain other undisputed libatious phenomena. Maybe when sitting with a British secret agent whose agent number is 007, and upon hearing him order a martini shaken not stirred, she exclaims Sounds good. I ll have the same. When the drink arrives she is disappointed that it is not a delicious fruity substance. If she has no other way of distinguishing a classic martini from the fruity substances she drinks from a martini glass, then the fault belongs to her for broadly grouping all alcoholic drinks poured into V-shaped glasses as martinis. In this modified case, I think you ll see that we would have a substantive dispute albeit a trivial one but substantive nonetheless, since (PVD) is not satisfied. While I think we can safely conclude that the unmodified martini case is a verbal dispute, we do so 9
because of the high probability that (PVD) is satisfied, not because we employ some a priori principle for satisfying (PVD). 8 So when it comes to satisfying (PVD) under a posteriori methods we can increase the probability that a dispute is verbal as more conceptual scenarios are tested where both sides, within their respective language frameworks, are able to equivalently describe the undisputed phenomena. Often the phenomena to which a language framework relates will be enormous. It will include both actual and possible metaphysical phenomena and all the implications of the framework for non-metaphysical issues. Since testing conceptual scenarios is so massive, we will usually have to hold back from considering broadrange metaphysical disputes to be verbal. There may yet be a set of untested phenomena about which there is a question regarding each framework s ability to organize in a way equivalent to the other, and until it is tested, we cannot know that (PVD) is satisfied. So long as this possibility remains, we cannot be certain that we have a verbal dispute. I don t think, however, that this invites a deep skepticism about identifying verbal disputes. Progress in identifying them can still be made. Most everyday verbal disputes will be like Bennett s Martini example they have a small range and so we can usually intuit the probabilities and see whether the dispute is substantive or verbal. The major metaphysical disputes will of course be more difficult to decide on, since they relate to broader phenomena and have many implications. But the more we test the phenomena to which each language relates and find that each language has a way of organizing it, the greater the probability that it is a verbal dispute. This means that some of our current disputes have a greater probability of being verbal disputes than others. So I see no reason to deny that a posteriori methods can help us in considering whether some of the mainstream debates are verbal disputes when we've been testing phenomena against the languages for a while, and there is nothing that presents itself to us as a substantive dispute. Rather 8 I think we can still hold out a modicum of possibility that it could turn out that one side will be able to account for some undisputed phenomena that the other cannot. 10
than giving up on identifying verbal disputes, we should feel free to go on identifying them, but I recommend that we put most of our talk about them in terms of probabilities. How do we test conceptual scenarios? I think we do this constantly as we go about our daily business. The nihilist goes about her day talking about chairs, rocks, and trees finding that her nihilism allows her to talk about these things without problem. As she encounters something that both she and the sum-lover agree requires description, and her framework is incapable of describing it, it is most likely a substantive dispute. Even possible-world phenomena get tested on a regular basis. The testing of some scenarios will be more intentional and rigorous than others. Hawthorne hones in on several scenarios which he claims are substantive metaphysical disputes since one side can describe something that the other cannot. If Hawthorne is right then we have likely identified a substantive dispute. If not, then we have increased the likelihood that the dispute is verbal. In broad-range disputes, a substantive dispute will usually be easier to identify than a verbal dispute. We have to test a lot of phenomena to which each language relates before we can conclude that we have a verbal dispute. So long as there is untested phenomena to which these claims relate, the probability of a substantive dispute can diminish, but some probability will always remain. Identifying a substantive dispute will not usually require us to test as many scenarios. Once we hit on phenomena that one language cannot organize as well as the other, we have likely identified the dispute as substantive. It could turn out that we have a verbal dispute after all, but only if it is later shown that the language does have a way of accounting for the phenomena. Treating All Metaphysical Disputes as Substantive I think that the burden of proof will largely be on the superficialist to show that the major metaphysical debates are really just verbal disputes. We naturally treat disputes as substantive unless we have reason to think otherwise - not the other way around. If the disputes are truly verbal then each party should, on the advice of Hirsch, no longer knock themselves out trying to come up with fancy arguments that might allow them to persevere in speaking their specialized language. To identify that a dispute is 11
verbal is to take away much, if not all of the motivation for disputing. So if we assume that every metaphysical dispute is verbal until shown to be substantive, we will likely lack the motivation to do the mental work of showing it to be substantive. On the other hand, if we treat metaphysical disputes as substantive, the testing of conceptual scenarios goes forward by the disputing parties. In scientific debate, the attitude of treating disputes as substantive by default is commonly understood. When questions remain about the capacity of even a long-standing, repeatedly-tested theory to account for a particular set of phenomena, a new theory may come along which can account for the phenomena just as well. Yet, because the new theory has not been tested to the extent that the older theory has, scientists do not (or at least should not) dismiss the disputes between the two theories as verbal since the new theory might explain the phenomena better than the old theory. In the meantime, we are required to treat the disputes between the two theories as substantive, even though at present the new theory explains as much as the older theory, and the dispute between them they may in fact turn out to be verbal. The default attitude works in metaphysics, too. Take the endurantism/perdurantism debate, for example. Endurantism is certainly the older, more established theory. Yet, right or wrong, superficialists have not given perdurantism much of a chance to explore whether it can account for certain agreed-upon data better than endurantism. Ted Sider and others have argued that four-dimensionalism is more compatible with the scientific data on Minkowski spacetime. 9 Maybe it is maybe it isn t, but if we follow the superficialist conclusion on the endurantism/perdurantism debate we will likely never know. In other academic disciplines and in commonplace public dealings, we naturally treat disputes as substantive until we have reason to think that they are verbal, not vice versa. This same attitude has the promise of benefitting metaphysical dispute. In principle, Hirsch agrees. He thinks that we should not consider a dispute to be verbal until all is said and done and each position has achieved a state of equilibrium (Hirsch 2009, 241). As we have seen, however, he is overly confident about getting to this point, and doesn t provide us a method for determining when we ve reached it. 9 See: Sider 2009, pp. 79-87, and Balashov 1999, pp. 644-62. 12
Treating metaphysical disputes as substantive by default has important implications for superficialism. We have no clear a priori method for determining when (PVD) is satisfied, and, when using a posteriori methods, we must speak in terms of probabilities of satisfying (PVD). The broader the range of the dispute the more difficult it will be to safely conclude that (PVD) has been satisfied. Until we can safely conclude that (PVD) is satisfied, we should follow the default of treating metaphysical disputes as substantive and move forward debating them as we have. 13
Bibliography Balashov, Y. (2000). Enduring and Perduring Objects in Minkowski Space-Time. Philosophical Studies, 99, 129-166. Balashov, Y. (1999). Relativistic Objects. Noûs, 33, 644-662. Bennett, K. (2009). Composition, Colocation, and Metaontology. In D. J. Chalmers, D. Manley, & R. Wasserman (Eds.), Metametaphysics: New Essays on the Foundations of Ontology (pp. 38-76). New York: Oxford University Press. Chalmers, D. (2009). Verbal Disputes and Philosophical Progress. Unpublished manuscript. Hawthorne, J. (2009). Superficialism in Ontology. In D. J. Chalmers, D. Manley, & R. Wasserman (Eds.), Metametaphysics: New Essays on the Foundations of Ontology (pp. 213-230). New York: Oxford University Press. Hirsch, E. (2009). Ontology and Alternative Languages. In D. J. Chalmers, D. Manley, & R. Wasserman (Eds.), Metametaphysics: New Essays on the Foundations of Ontology (pp. 231-259). New York: Oxford University Press. Hirsch, E. (2005). Physical-Object Ontology, Verbal Disputes, and Common Sense. Philosophy and Phenomenological Research, 70, 67-96. Kuhn, T. S. (1996). The Structure of Scientific Revolutions (3rd ed.). Chicago: The University of Chicago Press. Sider, T. (2001). Four-Dimensionalism: An Ontology of Persistence and Time. New York: Oxford University Press. van Inwagen, P. (1990). Material Beings. Ithaca, NY: Cornell University Press. Yablo, S. (2009). Must Existence-Questions Have Answers? In D. J. Chalmers, D. Manley, & R. Wasserman (Eds.), Metametaphysics: New Essays on the Foundations of Ontology (pp. 507-525). New York: Oxford University Press. 14