Essentialism, Conventionalism, and De Re/De Dicto Necessity

Lowe's metaphysical considerations are that wholesale anti-realism follows from the denial of essentialism and, furthermore, wholesale anti-realism is incoherent. Since these claims may well strike anti-essentialists as implausible—evidenced by the many anti-essentialists who are not wholesale anti-realists and who detect no incoherence in their denial of essentialism—some further defence is in order.

It bears repeating that the route into essentialism is, first, a recognition that the essence of a thing is “what it is to be” that (kind of) thing. The essence of a thing is just its identity. Not to be confused with the identity relation holding between a thing and itself or the property of being self-identical—which, if legitimate, are possessed by every object—an object's identity is some fact of the matter that makes it what it is and not something else.Footnote ¹ To deny this is to claim that there isn't anything that it is to be that (kind of) thing. On all accounts, that is to be denied. Even one who thinks that there is nothing in reality that gives metaphysical priority to some objects over others, or one who thinks that there is nothing in reality that distinguishes genuine from gruesome objects, denies that there's nothing that makes something a fax or a fox. Even a world of gunk that is chunked by us in conventional ways contains chunks of gunk that count as faxes and not as foxes, and vice versa. Unless chunking gunk results in a chunk that is nothing at all, there is something it is to be that thing. It is, furthermore, one thing to chunk a fax and quite a different thing to chunk a fox.

Some find this argument unconvincing, since if it is gunk that we chunk, then there are no faxes and no foxes in themselves, even after chunking. There is only gunk and our selective attention to fax- or fox-shaped gunk. That is not the same as there being, in reality, faxes and foxes. If any position could count as ontological conventionalism, this is it. We do not discover objects; we create them or, perhaps expressed more carefully, we carry on as though there were such objects. Lowe correctly takes this inference, even if warranted, to be insufficient to avoid essentialist commitments. The linchpin of his argument that anti-essentialism entails incoherent global anti-realism is, effectively, that even the various forms of ontological conventionalism tacitly assume that the gunk chunker is something prior to the chunking and must be so in order to be a (potential) chunker. Not only that, but the tools used in the chunking, be they private mental concepts, portions of public languages, or abstract propositions must be things. It is one thing to maintain that all non-personal, non-conceptual, and non-linguistic things depend on the users and uses of concepts or language for their existence; it is quite another to maintain that there is nothing that makes even these things themselves—the chunkers, the conceptual schemers, their tools—what they are and this is just what wholesale anti-essentialism must involve. Conventionalism about some things requires that makers and users of conventions possess non-derivative identities. If for everything there is nothing that it is to be that thing, then there is nothing that it is to be a chunker, a conceptual schemer, a language user, a concept, an expression, a convention, or even something that exists or is the case by convention. At the most general level, the essentialist affirms with Quine, while recognising the essentialist's very unQuinean meaning: no entity without identity. Without cognisers and their tools, nothing else can have its identity dependent upon the process of making the world and its (other) contents. In effect, Lowe's argument entails that there can be no separation between either mind or language on the one hand and the world on the other. Wholesale conventionalism can be maintained only to the extent that a hard and fast distinction between mind and/or language on the one hand and the world on the other is maintained. Certainly none who wish to maintain naturalistic accounts of persons and language can maintain this separation required for exempting persons and language from the force of the essentialist's argument.

The force of my use of ‘certainly’ is merely that things are more obvious in this context, since it is part of the very naturalistic thesis that humans and language use are to be seen as fully embedded in not only The World in some metaphysical sense but also in the natural world as it is ordinarily understood. Other forms of conventionalism are no more able to escape the consequences of the argument, even though it is no part of the explicit statement of those positions that the very same properties, relations, and the laws governing the natural world apply to language users and languages. Independent essence, i.e., essence that is not parasitic upon the existence or activity of something else, must be found somewhere, even if there are various options available to essentialists regarding exactly where essence is best located: in things to be discovered or in the creators of heretofore not existent objects.

For this reason, Lowe is explicit—and quite correct—that de dicto necessity is not free from essentialist implications. Though it is easiest to appreciate on platonistic accounts of concepts or language types, the point applies equally to constructivist views of concepts and language. The mental or linguistic items are objects with characteristics and relations. They are not non-objects, after all. So, if there are concepts suitably related, if there are propositions and languages suitably constituted, then it might well be that there are truths that derive from conceptual or linguistic meaning alone. There are such truths, however, only insofar as there is something that it is to be the concept of bachelor or the meaning or other characteristics of ‘bachelor’ and its relations to that of ‘male’ that make it so, for instance. To ignore for the moment what is distinctive about Lowe's serious essentialism, de dicto necessity is a species of de re necessity. Anyone prone to countenance de dicto necessity must recognise mental and/or linguistic entities, thus counting each of them as a res to which necessity attaches.

Serious Essentialism

The serious part of Lowe's serious essentialism is that it is better to account for metaphysical modalities in terms of essence rather than the other way around. For those whose understanding of essentialism derives from some combination of quantified modal logics, possible worlds semantics, and direct reference, this order of analysis seems backwards. Nearly all of us cut our logical teeth on the rudiments of first-order non-modal logics, progressed to modal propositional logics, and then moved to quantified modal logics and their attending formal semantics. That framework admits both ‘□∃xFx’ and ‘∃x□Fx’ as well-formed formulae and accounts for the semantic difference between the two. From a formal point of view, perhaps, there is nothing amiss; the mistake the serious essentialist's critic is prone to make is to confuse the task of formal semantics with philosophical insight. The merits of the familiar extensional semantics are many and well-known, not the least of which is that the introduction of an accessibility relation provided a background against which we could understand and assess the differences between various modal logical systems. It is a mistake, though, to think that there is something sacrosanct about the tacit framework of the semantics for fruitful philosophical theorising.

Consider the metaphysics of concrete particulars, with two main theories being the substratum theory and the bundle theory. The former takes ordinary objects to be composites of a substratum and attributes; the latter takes ordinary objects to be bundles of attributes alone. For the substratum theorist there are sharp distinctions between a thing, its substratum and its attributes. Standard first-order formal languages use quantifiers that take variables ranging over individuals. Predicate letters, then, permit us to make attributions to things. The syntax of such languages is completely blind to the composition of the objects in the domain. It is precisely because the languages are so blind that it is misguided for a substratum theorist to think that there is significant philosophical warrant for the substratum theory and against the bundle theory to be gained from the syntax of formal languages. Similarly, those comfortable with the standard way of characterising essences or essential properties in terms of necessity should take no philosophical comfort in the standard possible worlds semantic frameworks that typically accompany formal modal languages.Footnote ² The serious essentialist asks us not to be blinded by artefacts of the now-standard possible worlds framework.

Let us stipulate that this framework give us de re modality, i.e., it gives us what we can articulate using boxes, diamonds, quantifiers, truth-functional connectives, and the quantification into modal contexts. De re modality, though, is too coarse grained for all purposes. Even before serious essentialism was clearly formulated, there were reasons to think that the possible worlds framework was a tool too blunt to serve as a foundation for a proper philosophical understanding of properties, for instance. Some bi-conditional mathematical theorems are not trivial. All and only equilateral triangles are equiangular. That this theorem of Euclidean geometry required proof, and not merely for academic curiosity or completeness, is grounds for thinking that being an equilateral triangle is not the same property as being an equiangular triangle. For a rough and ready ground for this claim, look in a dictionary. The definition of ‘equilateral’ makes reference to side, equality, and (tacitly) length; no reference to interior angles or degrees appears. Similarly, the definition of ‘equiangular’ makes reference to angles, equality, and (tacitly) degrees; no reference to sides or length appears.Footnote ³ It is a stretch to which some effort and argument is required to embrace the position that, dictionary appearances aside, ‘equilateral’ and ‘equiangular’ express the same property.

One not wedded to the possible worlds framework being a universal solvent for philosophical problems should take these semantic appearances as the basis for a search for a more fine-grained framework that permits the distinction of at least some necessarily co-extensive, properties, thus showing possibilist ontology to be insufficient for meaning. Meinongians and advocates of impossible worlds might champion an ontological solution to this problem by including impossible objects in addition to actual and merely possible objects in their ontologies. Impossibilist ontology, though, is sufficient only to the extent that the ontology incorporates, say, equilateral triangles that are not equiangular or equiangular triangles that are not equilateral, to carry on with Euclidean triangles as our example.

A question to be answered, though, is why would an impossibilist maintain that there are such objects as equilateral but not equiangular or equiangular but not equilateral Euclidean triangles and, furthermore, why would an impossibilist maintain that such objects are impossible? Note, first, that impossibilists are not typically in the business of simply taking standard assertions and playing with the syntax to produce new assertions and, when the assertion seems odd, declaring the assertion to be necessarily false and the relevant objects impossible. Such a procedure would be exceedingly inefficient, since merely mixing up the syntax of an assertion that is acceptable to ontological traditionalists often results not in an assertion about an impossible object but in no assertion at all. The result would frequently be nonsense. An impossibilist might, though, be prepared to maintain that there are impossible triangles of the sort mentioned precisely because being an equilateral triangle and being an equiangular triangle are different properties and the difference between them is sufficiently evident in spite of possibilist accounts of properties. It is the ability to discriminate these properties that permits us to see the impossibilist's claim as a claim and not nonsense.

The serious essentialist, of course, has no interest in embracing impossible objects, but that the impossibilist's claim is not mere gibberish is the very same basis from which the serious essentialist proposes that we not be blinded by artefacts of the possible worlds frameworks and that we not, for that reason, think in the course-grained way that it requires. Antecedently, we took some properties to be distinct, even if necessarily co-extensive. That possibilist accounts of properties make our discriminatory abilities illusory contributes to the implausibility of those accounts. That impossibilist ontologies could embrace the impossible triangles as impossible objects contributes to the plausibility of any theory which distinguishes these properties; otherwise, by their own lights, there would be no reason to think that there were objects with incompatible properties rather than no such objects at all.Footnote ⁴

If the framework within which we define de re modality is too coarse, then the only way to permit ontology to be the basis for properties is to accept that the ontology of worlds comprises impossible worlds as well as possible. For those who want no truck with impossible worlds or objects, the limits of the philosophical usefulness of ontology have been exposed. Helping ourselves to Quine's distinction between ontology and ideology [Quine, Reference Quine1951], the essentialist can maintain that when the limits of the usefulness of ontology have been exposed, philosophical progress is achieved only with developments in ideology. Linguistically, further expressive resources are required. If those expressive resources are deemed to be fit for purpose only if connected with non-linguistic features of reality, then those features are not objects but characteristics of objects or they are ways in which objects can exhibit those characteristics.

Thus we have well-motivated grounds for ceding to Fine what he wants when he urges us to distinguish between Socrates necessarily being a member of {Socrates} and Socrates being a member of {Socrates} essentially [Fine, Reference Fine1994]. Even though it is not possible for Socrates to fail to be a member of singleton Socrates, Socrates is not essentially a member of singleton Socrates, or as Lowe would put it, it is not part of Socrates' nature to be a member of singleton Socrates, or any other set for that matter.Footnote ⁵ It is only an artefact of trying to treat the possible worlds framework with ontological seriousness and to put it to work regarding the nature of properties that blinds us to distinctions that we had been able to make among distinct properties. Similarly, treating that framework in that way makes us less able to distinguish what we should be able to distinguish: what is necessarily true of Socrates and what is true according to his nature.

The rough and ready way of making the distinction vivid is the same here as it was with equilaterality and equiangularity. The metaphysical dictionary, as it were, for Socrates might have in its entry that Socrates is a human person, perhaps of specific parentage, etc. What it will not contain, though, is any reference to the sets of which he is a member. The omission results not from limitations of space, but from the recognition that being Socrates is not a matter of being a member of any sets, even if the entry for {Socrates} is very much a matter of containing Socrates and essentially so. The point generalises. Imagine that some metaphysical issue apparently has been settled. The final accounting of events, say, is said to be complete. E is an event iff it satisfies conditions C₁, C₂, …, C_n. One would be perverse to insist that the analysis must also specify, for the sake of completeness, that a further condition, C_n+1 is required to state that each event must also be a member of its singleton, its singleton's singleton, the set containing only it and Socrates, and so infinitely on.Footnote ⁶ Only in philosophical contexts in which explicit attention to the possible worlds framework limits the subtlety of our thought does anyone find this final imposition on the nature of a (kind of) thing at all reasonable. Cavilling at the essentialist's subtlety on possibilist grounds, then, is unwarranted.

Essentialism and Essences

Lowe is right to argue that essences are not things and he is correct to reason as he does. Essences are not things to be perceived. They are not objects that are components of other objects, and so on for the essence itself and its essence, etc. There is, though, slightly more point than Lowe sees to finding locutions that can serve as alternatives to those that give the impression that essences are things, i.e., alternatives to locutions that involve singular terms that seem to refer and to existential quantification apparently over essences. Essentialists certainly affirm the sentiments behind “Socrates' essence contains only properties that he possesses whenever and wherever he exists” and “The essence of Socrates' is not the same as The Taj Mahal's”. ‘Socrates' essence’ functions all too similarly to ‘Socrates' beard’; ‘the essence of Socrates’ functions all too similarly to ‘the beard of Socrates’. It would be somewhat more useful to have ways of speaking that dispel the faulty ontological impression that Lowe wishes to avoid. The significance of such locutions would be to enable essentialists to express more perspicuously where the metaphysical fundamentals are located.

Lowe opts for a sentential operator “‘it is part of the essence of X that’ — where ‘the essence of X’ is not taken to make an independent contribution to the meaning of the operator, which might be represented symbolically by, say, ‘E_X’ in a sentential formula of the form ‘E_X(p)’.” As the creator of the formal locution, Lowe is certainly its master and is free to stipulate how it will function. If he says that ‘the essence of X’ makes no independent semantic contribution, then so it does not. The problem to be avoided at this stage, however, is not so much a mistaken doctrine held by essentialists or their critics; it is a mistaken impression to which our standard uses of various linguistic constructions give rise, including the locution form ‘the essence of X’. Not only do constructions like ‘Socrates' essence’ or ‘the essence of Socrates’ give the appearance of referring to some thing, even talk of essential properties gives rise to apparent ontological commitments on nearly all accounts of properties. If the essentialist urges us to make philosophical progress by expanding our ideology rather than our ontology, then it would be good to avoid the appearance of ontology whenever possible.

To be explicit, the point of locating or constructing alternative locutions is to express what is true without giving any mistaken impressions. It is not, as Lowe worries it must be, to paraphrase ontologically committing locutions. If paraphrases maintain both meaning and commitments, then no progress can be made paraphrastically; if paraphrases fail to maintain both meaning and commitments the essentialist invites the complaint that the paraphrases are inadequate. What serious essentialists need is a way to express what they take to be the sober truth. To that end, I propose not that we ban all such appearance of ontological commitments to essences as things, but only that we be clear about what the fundamentals are. One way to deal with the appearance is to maintain that any such appearance is just talk; it arises only out of locutions of convenience and not from any insight into what there is and how it is. Maintaining this is not difficult, if there are ways of saying what we want to say without the appearances. Essentialists use such constructions regularly. It remains only to mark them as expressing most clearly what the essentialist wants to express and all others being legitimate only if they are deemed to be parasitic upon the preferred expressions. The parasitic expressions need not be equivalent in meaning or even quite good paraphrases; they must be deemed to be apt only on the basis of the ontology to which one is committed when using the preferred expressions. If we think that it is part of the essence of Socrates that he is a human being, we simply say that Socrates is essentially human or that he is human by nature. If we think that it is part of his nature that he was the son of Sophroniscus and Phaenarete, we say that he was essentially their son. Saying these things does not give the impression that there is a thing, or even a property, had by Socrates that is the special concern of the essentialist. If anything, this way of speaking expresses what is surely initially more natural, i.e., that there is something special about the way in which Socrates is human or that there is something special about how he is related to Sophroniscus and Phaenarete. The box and diamond approach to such matters is one way to articulate what is special. Once that approach ceases to be normative, other ways of understanding essentialist claims come to the fore. Here, McGinn's copula modifier suggests itself [McGinn, Reference McGinn2000]. Whereas the box and diamond approach takes modal notions to be primitive, the serious essentialist takes as primitive what it is to be a (kind) of thing or what it is to be a certain way in a certain way. If the metaphysical fundamentals are to do with essence rather than de re necessity as the serious essentialist maintains, and if de dicto modality is a special case of de re, and if the grounds for maintaining the first of these is the fineness of discrimination we were warranted in believing we possessed prior to the advent of theories framed in terms of possible worlds, then taking the locutions I propose to be the most preferred is quite natural and motivated. If the philosophical case I have made here on behalf of the serious essentialist is correct it would be quite surprising were we constrained to use boxes and diamonds to express the metaphysical fundamentals, since the serious essentialist urges on us a more fine-grained framework on the basis of pre-possible-worlds understandings of at least some things. Since we may speak of the ways Socrates is essentially, the serious essentialist has no need for special sentential operators when metaphysically serious.

A caveat on ‘ways’ is in order. The serious essentialist should maintain that even though we can talk of the ways Socrates is and how some of them concern the way he is essentially and some do not, such talk of ways does not introduce ontology just where the serious essentialist hoped to avoid it. Talk of ways is just shorthand, just as talk of possibilities is. There is a fork in the road, one being the High Road and one being the Low Road. In virtue of there being both the High Road and the Low Road, there are two ways to Scotland from here. Likewise, since there are the two roads, there are two possible routes from here to there, i.e., two possibilities for getting from here to there. Generalise this lesson for ways and it is evident that there is nothing standing between a serious essentialist and even a rather austere form of nominalism. As the point of constructions involving ‘routes’ and ‘possibilities’ is quite sensibly seen to be parasitic upon the existence of roads, for instance, so constructions in terms of ‘ways’ and ‘properties’ provides convenient shorthand for what does not seem ontologically committing in the first place. Two spheres, one round and red the other round and green, share a property and differ on a property. The use of ‘property’ and the talk of sharing, differing, and counting merely gives us a way of expressing how this situation is similar to two cubes one of which is blue and other of which is yellow and infinitely many other kinds of cases. In none of these cases need one be committed to some previously undreamed of entities be they ways, possibilities, or even properties.

Freed from the demand of meaning-preserving paraphrases, the serious essentialist can maintain that a metaphysical basement is more clearly articulated when one determines that Socrates is essentially the son of Sophroniscus and Phaenarete, say, and not when one determines that it is part of his essence that he is their son or that necessarily he is so or that it is necessarily true that he is so. The less-than-perspicuous expressions give the impression that the essence is a thing with parts, encourage a lack of subtlety in our thinking, or invoke truth bearers and meta-linguistic considerations where they are both unhelpful and extraneous. Given the grammar with which we are so familiar and have such facility, these constructions are useful, even though they are not, the serious essentialist can maintain, a straightforward route to metaphysical insight. Thus, is the serious essentialist freed not only from the commitment to essences as things, but also from other commitments.

Essence and Knowledge

According to critics, the thorniest problem for essentialism is the question of our knowledge of essence. It is usually at this point that terms of abuse such as ‘dark’, ‘mysterious’, and ‘occult’ are wheeled out. Lowe goes some way in undercutting the appropriateness of these terms by the form of his epistemological argument. If this argument succeeds, then the essentialist's problem is a problem for everyone. Unless we have at least partial knowledge of the essence of a thing we, quite literally, know not of which we speak. To know the essence of a thing is to know what it is to be that thing which, where appropriate, is to know the kind of thing it is which, in turn, involves knowing what it is to be a thing of that kind. Without this knowledge already in hand, we cannot even begin empirical investigations. We cannot discover the existence of a fax or a fox if we have no idea what it would be to be a fax or a fox. This priority of knowledge of essence over empirical knowledge is the priority of the a priori over the a posteriori.

It is no criticism of Lowe's claim that empirical research and discovery presupposes knowledge of essences that this knowledge is typically, if not always, incomplete. Requiring complete knowledge is much too demanding in other contexts, even by anti-essentialist lights. In ordinary contexts, we distinguish objects not by their essences but by their attributes. For no ordinary object does anyone possess complete knowledge of that item's attributes and subtle differences between two concrete objects can make distinguishing them impossible in practice. Nevertheless, when our task is merely to distinguish an object from those with quite different characteristics, we do not take our lack of comprehensive knowledge of an object to undermine the contention that we distinguish objects by way of their attributes. Likewise, our lack of complete knowledge of what it is to be a human does not prevent us from knowing when there are people in the vicinity and our lack of complete knowledge of the particulars of Socrates' nature need not prevent us from knowing when Socrates is among us. As in cases with which the anti-essentialist is comfortable, only enough knowledge is required. So long as I know enough about what it is to be a hole in a crystal sphere and what it is to be a planet, I can make the appropriate discriminations in both modal and non-modal contexts. It is precisely knowing enough about their respective natures that permits us to know that some evidence sufficient for us to judge that Hesperus and Phosphorus are one if they are planets, is insufficient for us to judge that they are one if they are holes.

The natural tendency among anti-essentialists is to think that while it may be that empirical investigation demands that we know what it is to be some kind of thing before we can determine whether there is anything of that kind around, this knowledge is just conceptual or linguistic knowledge and not especially deep metaphysical knowledge. The mistake, by the serious essentialist's lights, is not that knowledge is conceptual or linguistic; it is that such knowledge is allegedly not knowledge of essence, permitting the knowledge to be at best contingent since the relevant concept(s) apply only contingently to what falls under those concepts or, worse, the knowledge is only apparent knowledge of reality since the concepts are merely our own impositions on what there is and not a proper characterisation of it. Given that serious essentialism extends to language users and languages, a central motivation for insisting that the epistemic prerequisites are conceptual is undercut. So long as knowledge of essence is part of the content of an item of conceptual knowledge, though, the serious essentialist should drop all resistance to the conceptual character of the knowledge. We have serious essentialism, whether the epistemically prior knowledge is conceptual or linguistic or not. The serious essentialist can allow that conceptual knowledge is epistemically prior to empirical investigation, so long as what it is to be a (certain kind of) thing is part of the content of the relevant concept(s).

Though there is no insurmountable barrier to serious essentialism involving conceptual knowledge in the first instance, the problem of a priori knowledge that is peculiar to strong essentialism is accounting for how it is a priori while also being about reality that is not of our making. If knowledge of essence is what the serious essentialist claims it is, how do we obtain it? There are two relevant questions to be addressed. How can the content of a concept contain such information? Are conceptual beliefs reliable?

There is the option of innate ideas. This option permits at least two sub-options. First, a theistic account according to which God created us and implanted the innate ideas that are grist for our conceptual mills. As those mills whirr away, we acquire conceptual knowledge. After arguing for essentialism, when accounting for the relevant knowledge of essence the essentialist is entitled to assume that things indeed do have natures and, so, a theistic essentialist is entitled to assume that God knows those natures and can pack the relevant content into what is innately given. What remains to be given for an account of this conceptual knowledge and how it is employed is grounds for maintaining that we are reliable when employing concepts. The short story for the essentialist is that this is a matter for the cognitive psychologists and not a special problem for the philosopher, much less a special problem for the serious essentialist.

With no recourse to omniscience behind our conceptual knowledge, non-theists must take a different tack, which I suggest to them here. If Lowe is correct that essentialism is warranted because empirical investigation demands it, then post-experiential investigation of the world cannot be the basis for the acquisition of the most basic conceptual knowledge. Hypothesis testing is out. A conceptual tabula rasa cannot gain conceptual knowledge of essence empirically via hypothesis testing. Any hypothesis would be framed in terms of concepts and any test would require that one know what it is to be the kind of thing about which the hypothesis hypothesises. There is something in the neighborhood, though. Hypothesis testing and variations on that theme rest naturally in internalist contexts regarding both conceptual content and epistemic justification. Conceptual externalism, though permits conceptual content to be subject to environmental facts not transparent to the concept user and epistemic externalism permits one to be warranted in believing something while not warranted in believing that one is so warranted. Conceptual externalism is not helpful to the essentialist who argues for the essentialist prerequisites of empirical knowledge, but epistemic externalism is. It matters not how one comes to conceptualise the world in a given way and how one comes to judge that the nature of a hole in a crystal sphere is this and the nature of a planet is that. What matters is that one does and that one is correct in doing so and reliable in wielding those concepts to make empirical judgements. The non-theistic externalist may have questions to answer regarding our warrant for thinking that we are correct and reliable in the requisite ways that the theist does not have, but the concern for the essentialist is not answering questions of how we are warranted in believing that we are warranted, since that is not a problem peculiar to the essentialist; everyone must deal with that. Naturalistic accounts of how cognisers possess the cognitive abilities to conceptualise and perceive the world as they do simply substitute natural processes for Divine activity above. Once the abilities are in place and the cognitive architecture can function as it does, that essentialist content is part of our conceptual content is not especially mysterious.

What any form of externalism provides the essentialist, whether theistic or not, is a way of making sense of how we could come to have the knowledge that the essentialist says we have. Everyone must make sense of first concepts in the externalist's way; all conscious mental activity that could count as reasoning, deciding on conventions, etc. must come after a basic conceptual framework is in place. Exactly how we manage to acquire initial concepts and employ them usefully, again, the essentialist can defer to the cognitive scientist without embarrassment. A nativist externalism permits the essentialist to find the a priori/a posteriori distinction to be a false dichotomy. There is certainly a sense in which there must be experience of the world to get the cognitive machinery going on the naturalist version of externalism, but the experience is not the rich, conscious experience that we so cherish. No matter. The essentialist's problem turns out not to be peculiarly difficult.