Abstract

@schnieder:2020 argues, against @orilia:2009 and @koslicki:2013a, that claims of existential grounding of the form “the fact that x exists is grounded in the fact that y is F” cannot be grounded in claims of existential dependence of the form “x existentially depends on y” and defends the view that the latter claims are grounded in the former, via a definition of existential dependence. I will first argue that Schnieder’s main point against the claim that existential grounding is grounded in existential dependence is not conclusive. I will then put forward a proposal concerning how claims of existential grounding can be grounded in claims of existential dependence. The proposal is a third way between those of Schnieder and Orilia/Koslicki, which, although accepting the former’s definition of existential dependence in terms of grounding, makes room for the latter’s idea that existential dependence does the real job of structuring reality.

According to a broadly Aristotelian metaphysics, reality is structured by relations of ontological priority: many kinds of entities inhabit the world, but they don’t live all on the same floor; some of them are more basic or fundamental than others, and the latter owe their existence to the former. However, even philosophers sympathetic to this layered conception of reality can still disagree on important questions. One of these questions concerns which notion (or notions) is more theoretically fruitful and metaphysically revealing to unfold and make the Aristotelian intuition precise. Starting from the seminal work of K. Fine (1995, 2001, 2012a), the notion of grounding has come to the fore as an appealing candidate for this role: that reality is layered in different levels of fundamentality should be spelled out saying that some facts or propositions are grounded in other facts or propositions, where the notion of grounding points to an objective explanatory connection of a non-causal kind between facts or propositions.

A central question regarding grounding is what, if anything, grounds facts of grounding themselves. In virtue of what, for instance, does the following fact obtain?

$\langle$The fact that the singleton of Socrates exists is grounded in the fact that Socrates exists$\rangle$.¹

A sensible answer to this question seems to be that (1) obtains because the singleton exists in virtue of Socrates (and not the other way around); similarly, one could say that the following fact

$\langle$The fact that fictional character J. Gatsby exists is grounded in the fact that F.S. Fitzgerald wrote the novel The Great Gatsby$\rangle$²

obtains because the character owes his existence to the writer. Generalizing: whenever there is an objective explanatory connection between the fact that a given thing exists and the fact that another thing is a certain way, this is explained by the existence of an objective relation between the entities involved in those facts, namely the relation of existential dependence; facts of existential grounding are in turn grounded in facts of existential dependence. This stance concerning the ground of existential grounding facts seems to be endorsed by philosophers like Orilia (2009) and Koslicki (2013). They should, therefore, be prepared to endorse the following principle by which Schnieder (2020, 114) makes explicit this view:

Dep $\rightarrow$ Ground. For any $x$, for any $y$, if $\exists F(\langle y$ is $F\rangle$ grounds $\langle x$ exists$\rangle)$, then $\langle x$ existentially depends on $y\rangle$ grounds $\langle\exists F(\langle y$ is $F\rangle$ grounds $\langle x$ exists$\rangle)\rangle$.

Those advocating Dep $\rightarrow$ Ground think that existential dependence does the real job of structuring reality and that existential grounding claims simply keep track of it. On the opposite side, people like Schnieder (2006b, 2020) and Correia (2005) think that facts of existential dependence are grounded in facts of existential grounding, therefore advocating the following principle:

Ground $\rightarrow$ Dep. For any $x$, for any $y$, if $x existentially depends on $y$, then $\langle\exists F(\langle y$ is $F\rangle$ grounds $\langle x$ exists$\rangle)\rangle$ grounds $\langle x$ existentially depends on $y\rangle$.

Ground $\rightarrow$ Dep follows from the fact that, according to Schnieder and Correia, existential dependence can be defined in terms of grounding in the following way:

Df.Dependence. $x$ existentially depends on $y=_{df}\Box(x$ exists $\rightarrow\exists F(\langle y$ is $F\rangle$ grounds $\langle x$ exists$\rangle))$.

That is to say, an entity $x$ existentially depends on another entity $y$ when, necessarily, if $x$ exists, there is a way $y$ is such that the fact that $y$ is that way grounds the fact that $x$ exists.

Provided that definitions like the former entail a strict non-causal explanatory relation between the definiens and the definiendum, one can infer Ground $\rightarrow$ Dep from the latter definition: since existential dependence can be defined in terms of grounding, facts of existential dependence are grounded in or metaphysically explained by facts of grounding; the former stay at an upper, less fundamental, level of reality than the latter.

Orilia (2009) and Koslicki (2013) have complained that Df.Dependence turns things upside down: that there is an objective explanatory connection between the fact that a given thing exists and the fact that another thing is a certain way should be explained by some fact concerning the entities involved in those facts and not the other way around. This fact is, according to Orilia and Koslicki, that one of these entities existentially depends on the other. Therefore, both Df.Dependence and Ground $\rightarrow$ Dep should be rejected in favour of Dep $\rightarrow$ Ground.

1 Schnieder’s Point against Dep $\rightarrow$ Ground

Schnieder’s (2020) main objection to Dep $\rightarrow$ Ground is, in a nutshell, that claims of existential dependence cannot ground claims of existential grounding such as

$\langle$Fitzgerald wrote The Great Gatsby$\rangle$ grounds $\langle$The fictional character J. Gatsby exists$\rangle$,

since the notion of existential dependence is too coarse-grained to sustain such an explanatory role.

The reason lies in the predicative content of the grounding facts involved in such claims. In fact, what grounds the existence of something in such claims is that something else is a certain way. However, the bare existence of a thing is consistent with and is actually accompanied by many different ways this thing is, many of which have no explanatory role for the existence of the grounded entity. So, pointing to the bare fact that $x$ existentially depends on $y$ gives no hint about which of these ways $y$ explains why $x$ exists.Thus, an explanation such as

[$\langle$Fitzgerald wrote The Great Gatsby$\rangle$ grounds $\langle$The fictional character J. Gatsby exists$\rangle$] because [the fictional character J. Gatsby existentially depends on Fitzgerald]³

suffers from an explanatory gap that depends on the genericity of the explanans in respect of the explanandum. Since the bare existence of Fitzgerald is consistent with his being in many different ways, many of which have no role in grounding the existence of the fictional character J. Gatsby, in order to explain why a fact like (3) obtains something more must be said than simply pointing to the fact that the existence of one of the entities involved in it (J. Gatsby) depends on the existence of another one (Fitzgerald) (see Schnieder 2020, 116). This limit is similar to that of an explanation such as “This rose is scarlet because it is red.” Since a red rose can be, for instance, also crimson or cardinal red, something more must be said in order to account for a rose being scarlet, rather than simply pointing to the fact that it is red.

Another related point made by Schnieder is that different properties of the grounding entities will be involved in different cases of existential grounding (for instance, the fictional character J. Gatsby exists because Fitzgerald wrote a novel, but a philosophy conference exists because some people behave in certain specific ways). But according to the advocates of Dep $\rightarrow$ Ground, in all these different cases, there is always the same kind of fact serving as ground, namely that a given entity existentially depends on another. So, explanations like (4) and a fortiori Dep $\rightarrow$ Ground lack systematicity: “There seems no systematic way in which the bare fact that $x$ existentially depends on $y$ could serve as an appropriate basis for such a variety of predications, and in particular on how it should pick out the relevant predications in the different cases” (Schnieder 2020, 116).⁴

Notice that, while according to Dep $\rightarrow$ Ground, what is grounded in a fact of existential dependence is a general, quantified fact (namely $\langle\exists F(\langle y$ is $F\rangle$ grounds $\langle x$ exists$\rangle)\rangle)$, what I have said so far concerns the grounds of specific instances of this general fact, such as (3).

This follows Schnieder’s line of reasoning since he accepts Fine’s (2012b) logic of ground, according to which an existentially quantified fact is immediately grounded only on its instances and, mediately, by whatever grounds them. Therefore, since

$\exists F(\langle$Fitzgerald is $F\rangle$ grounds $\langle$The fictional character J. Gatsby exists$\rangle)$

is immediately grounded in (3), the only sensible way in which Dep $\rightarrow$ Ground can be justified is by claiming that the relevant instance of $\langle x$ existentially depends on $y\rangle$ (i.e., $\langle$The fictional character J. Gatsby existentially depends on Fitzgerald$\rangle$) grounds (3), and this is tantamount to arguing in favour of (4). Once this has been done, it will follow, by the transitivity of grounding or “because,” that

[$\exists F(\langle$Fitzgerald is $F\rangle$ grounds $\langle$The fictional character J. Gatsby exists$\rangle)$] because [the fictional character J. Gatsby existentially depends on Fitzgerald],

which is the relevant instance of Dep $\rightarrow$ Ground.

What Schnieder wants to point out is precisely that the crucial step in this line of reasoning (that is to say (4)) is, for the reasons given above, flawed.

Although I agree with this line of reasoning, I will try to show in the following paragraphs that a modified version of Dep $\rightarrow$ Ground can successfully survive it and that, therefore, the advocate of existential dependence still has space to argue that claims of existential grounding, such as (3), are in fact grounded in claims of existential dependence. Since the core idea of my argument will be that what the advocate of existential dependence must do in order to get her point is to choose the right kind of dependees in the relation of existential dependence, I will call it “the ontological challenge” to Schnieder’s criticism of Dep $\rightarrow$ Ground.

2 The Ontological Challenge

Consider the following explanation:

[$\langle$Fitzgerald wrote the Great Gatsby$\rangle$ grounds $\langle$The fictional character J. Gatsby exists$\rangle$] because [the fictional character J. Gatsby existentially depends on Fitzgerald qua author of The Great Gatsby].

(7) doesn’t suffer from the explanatory gap and the lack of systematicity that affect (4). At the same time, the explanans of (7) is a sentence asserting that an entity (the fictional character J. Gatsby) exists in virtue of another entity (Fitzgerald qua author of The Great Gatsby), so it is a claim of existential dependence. What happened is that I substituted for the singular term “Fitzgerald” a different singular term in which the information that was missing in the explanans of (4) is now embedded. What changed, as far as the objects referred to by the sentence are concerned, is that the singular term “Fitzgerald qua author of The Great Gatsby” refers to what Fine (1982) called a qua-object, where qua-objects are objects that differ from their ordinary twins for some of their essential properties (the main difference between $a$ and a qua F, with $F$ a contingent property of $a$, being that a qua F, but not $a$, is essentially, so necessarily, $F$).

If the advocate of existential dependence as the ground of facts of existential grounding takes this route, he must adopt a modified version of Dep $\rightarrow$ Ground. In fact, according to Dep $\rightarrow$ Ground, the fact that there is a certain way $y$ is such that $x$ exists because $y$ is that way is grounded in the fact that $x$ existentially depends on $y$. But according to the picture sketched above, the ontological ground of $x$, in the claim of existential dependence, is not the same entity $y$ which is involved in the grounding fact to be explained: Fitzgerald and Fitzgerald qua author of The Great Gatsby are in fact two different entities that coincide in the actual world. So, Dep $\rightarrow$ Ground must be substituted by

Dep $\rightarrow$ Ground*. For any $x$, for any $y$: if $\exists F(\langle y$ is $F\rangle$ grounds $\langle x$ exists$\rangle)$ then $\exists z(\langle x$ existentially depends on $z\rangle$ grounds $\langle\exists F(\langle y$ is $F\rangle$ grounds $\langle x$ exists$\rangle)\rangle)$,

where $z$ is an entity suitably related to $y$ and its being $F$.⁵

It should be noted that someone accepting Dep $\rightarrow$ Ground* is not forced to admit qua-objects in his preferred ontology; in fact, qua-objects are not the only entities that can do the desired job. What is needed is, in fact, an entity that is as fine-grained as an individual having a property and referred to by a singular term embedding the same information as the sentence saying that the individual at stake has the relevant property. A linguistic device by which one can obtain this kind of singular term from sentences is nominalization. For instance, by nominalizing the sentence “Fitzgerald wrote The Great Gatsby,” we obtain singular terms such as “Fitzgerald’s writing The Great Gatsby” or “the fact that Fitzgerald wrote The Great Gatsby,” which refer, if they refer at all, respectively to an event and a fact. Events and facts are not the only entities that can do the job. For instance, starting from “this rose is red,” we can obtain “the redness of this rose,” which, presumably, refers to a trope.

3 Too Easy?

The advocate of the priority of grounding has an easy rejoinder to the ontological challenge. She could say, in a nutshell, that the challenge is based on a linguistic trick that should be banned from serious metaphysics. The trick in question is the linguistic transformation based on the nominalization device by which one can obtain singular terms (or, better, expressions functioning grammatically as singular terms) from sentences. But the fact that we can legitimately transform sentences such as “the fictional character Jay Gatsby exists because Fitzgerald wrote The Great Gatsby” into sentences such as “the fictional character Jay Gatsby exists because of, or in virtue of, Fitzgerald’s writing The Great Gatsby” or “the fictional character Jay Gatsby exists because of, or in virtue of, Fitzgerald qua author of The Great Gatsby” is no real advance as far as metaphysical explanation is concerned.

Consider the sentence: “It is true that nobody is in the room because nobody is in the room The fact that we can transform it, by nominalization, into”it is true that nobody is in the room in virtue of nobody’s being in the room” is not a sufficient reason to claim that there is an entity, referred to by the expression “nobody’s being in the room,” which makes the proposition that nobody is in the room true, which is, so to speak, the ontological ground of its truth. In fact, first of all, it is not granted that the expression “nobody’s being in the room,” although functioning grammatically as a singular term, is semantically a singular term, that is, a denoting term. And second, provided that the expression refers to something, it should be demonstrated that the existence of this something explains why the proposition is true. In the same way, in order to justify the claim that the purported entity referred to by the expression “Fitzgerald’s writing The Great Gatsby” is the existential ground of the fictional character J. Gatsby, one should, in the first place, argue that such an entity exists (that is, that the expression “Fitzgerald’s writing The Great Gatsby” is a genuine singular term), and, in the second place, that its existence has explanatory force towards the existence of the fictional character. Concerning the first point, it should be noted that the mere fact that a term $a$ functions grammatically as a singular term, for instance, allowing existential generalizations (inferences from “$a$ is $F$” to “Something is $F$”) is consistent with its not being a denoting term, since the existential quantifier can be interpreted, in such cases, substitutionally. (This is why the simple fact that we can infer “there is something I dreamt of last night” from “I dreamt of a unicorn last night” doesn’t force us to admit the existence of dreamt unicorns.) Concerning the second point, even granting that nominalizations are genuine singular terms, one could still embrace Schiffer’s (2003) view, according to which the entities referred to by these terms are pleonastic entities, that is, entities supervening on our linguistic practices and having no explanatory power concerning the rest of the world.⁶

If this is true, then, although one can infer, in virtue of the syntactic transformation taking us from “$p$” to “nom $(p)$,”⁷

The fictional character J. Gatsby exists in virtue of the fact that Fitzgerald wrote The Great Gatsby/Fitzgerald’s writing The Great Gatsby/Fitzgerald qua author of The Great Gatsby

from

The fictional character J. Gatsby exists because Fitzgerald wrote The Great Gatsby,

one cannot infer from (8)

The fictional character J. Gatsby exists because the fact that Fitzgerald wrote The Great Gatsby exists/Fitzgerald’s writing The Great Gatsby exists/Fitzgerald qua author of The Great Gatsby exists,

since it is not granted that the grammatically singular term involved in the explanans of (8) refers and, provided it does, that it refers to a non-pleonastic entity.

But if the relation of existential dependence must be able to do a serious metaphysical job, such as the job of grounding claims of existential grounding, as is claimed by the advocate of Dep $\rightarrow$ Ground*, what is needed is (10) and not just (8). This can also be seen by taking note of the following fact.

Since (8) is obtained by (9) through the trivial linguistic transformation from a sentence to its nominalization, it is reasonable to claim that (9) is conceptually prior to (8): understanding a nominalization, in fact, involves understanding the sentence nominalized but not the other way around. One can generalize this point by introducing what could be called a “pleonastic” notion of existential dependence, which is implicitly defined by the following schema:

Pleonastic Existential Dependence (PED). $x$ existentially depends on/exists in virtue of ^PL nom $(p)=_{df}\Box$[$x$ exists $\rightarrow(x$ exists because $p)$].

Now, PED cannot be the notion involved in the sentences of the form

[$\langle y$ is $F\rangle$ grounds $\langle x$ exists$\rangle$] because [$x$ existentially depends on nom $(y$ is $F)$]

to which the advocate of Dep $\rightarrow$ Ground* is committed. In fact, if the notion of existential dependence at stake here were the pleonastic one, the sentences of this form would be false for the following reason.

Once it is admitted that definitional links (or even just conceptual priority) holding between a sentence $P$ and another sentence $Q$ (where $P$ is conceptually prior or definitional in respect of $Q$) justify the assertion of an explanation of the form $Q$ because $P$, one should be prepared to accept, when PED is at stake, a sentence like

[$x$ existentially depends on/exists in virtue of ^PL nom $(y$ is $F)$] because [$x$ exists because $y$ is $F$].⁸

But (12) together with (11) entail, given the transitivity of “because,”

[$\langle y$ is $F\rangle$ grounds $\langle x$ exists$\rangle$] because [$x$ exists because $y$ is $F$],

which is equivalent, shifting from “ground” to “because” talk to

[$x$ exists because $y$ is $F$] because [$x$ exists because $y$ is $F$]

which violates the irreflexivity of “because.” So, if the existential dependence at stake in the explanans of (11) were PED, (11) would be false, since it would entail something false, and Dep $\rightarrow$ Ground* should be rejected.

Therefore, what is needed to defend Dep $\rightarrow$ Ground* is not PED but a full-blooded notion of existential dependence that can be characterized as follows.

Full-Blooded Existential Dependence (FED). $x$ existentially depends on/exists in virtue of ^FB $z=_{df}\Box(x$ exists $\rightarrow(x$ exists because $z$ exists$))$.⁹

It should be noted that FED is an instance of Df.Dependence since the former can be obtained from the latter by instantiating the predicate variable $F$ with the existence predicate. For this reason, FED is more demanding than the notion defined by Df.Dependence. In fact, whereas according to the latter, what is needed in order that $x$ existentially depends on $y$ is that $x$’s existence is explained by some feature of $y$; according to FED, however, the existence of the dependent entity must be explained by the bare existence of the dependee. So, “$x$ depends on ^FB $y$” entails “$x$ depends on ^DEF.DEP$y$” but not the other way around.¹⁰

4 The Third Way

Let’s take stock. So far, I have argued that an advocate of existential dependence can circumvent Schnieder’s argument against Dep $\rightarrow$ Ground by putting forward another principle according to which claims of existential dependence ground claims of existential grounding, that is, Dep $\rightarrow$ Ground*. Then I explained why, in order to defend this principle, a notion of existential dependence (FED) is needed, which is an instance of Schnieder’s characterization of existential dependence in terms of grounding. This means that the supporter of Dep $\rightarrow$ Ground* must argue in favor of sentences such as

[J. Gatsby exists because Fitzgerald wrote The Great Gatsby] because [J. Gatsby exists because nom (Fitzgerald wrote The Great Gatsby) exists].¹¹

This amounts to arguing for the claim that a sentence like

J. Gatsby exists because Fitzgerald wrote The Great Gatsby.

That is to say, the explanandum of (15), is true in virtue of the fact that the entities referred to by expressions such as “Fitzgerald qua author of The Great Gatsby,” or “the fact that Fitzgerald wrote The Great Gatsby,” explain, by their existence, the existence of the fictional character J. Gatsby.

I’m going now to show how this could be done. The resulting picture will be a syncretistic third way between that of Schnieder/Correia and that of Orilia/Koslicki; on this picture, in fact, on the one hand, the grounding/explanatory characterization of existential dependence put forward by the first pair of authors is accepted, but, on the other hand, the idea of the second pair of authors, according to which existential dependence does the real metaphysical job in structuring reality, is maintained.

I think that what the supporter of Dep $\rightarrow$ Ground* should do to defend (15) is, first of all, to argue in favor of the explanans of it, that is to say,

J. Gatsby exists because nom (Fitzgerald wrote The Great Gatsby) exists

which says that J. Gatsby full-bloodedly existentially depends on the entity referred to by the nominalization of “Fitzgerald wrote The Great Gatsby.” This can be done by arguing for

Fitzgerald wrote The Great Gatsby because nom (Fitzgerald wrote The Great Gatsby) exists.

(18), in fact, together with

J. Gatsby exists because Fitzgerald wrote The Great Gatsby

(which is the explanatory claim endorsed by the advocate of Ground $\rightarrow$ Dep) entails (17).

How could (18), the crucial premise of the argument, be justified? I think that what the advocate of Dep $\rightarrow$ Ground* should do is claim that what it is for something to be $F$ is for the entity referred to by “nom $(y$ is $F)$” to exist. More generally, what the advocate of Dep $\rightarrow$ Ground* should argue for is a conception of reality according to which what it is for an n-tuple of individuals to exemplify an n-ary property is for some entity to exist. This amounts to defending a metaphysical view, according to which the entities referred by the nominalizations in question are not pleonastic entities, but, on the contrary, are bona fide and fundamental constituents of reality. For instance, if one thinks that “nom $(y$ is $F)$” refers to a given fact, one should embrace a Tractarian or Armstrongian metaphysics, according to which the world is a world of facts. According to such a view, what it is for Fitzgerald to write The Great Gatsby is for a given entity (a fact) to exist. Defending this kind of metaphysics is indeed a difficult task but not an incoherent one, and someone could and has actually claimed one worth pursuing.

Once the advocate of Dep $\rightarrow$ Ground* has successfully argued that what it is for $y$ to be $F$ is for some entity to exist, she can reasonably claim that an explanation of the form “$x$ exists because $y$ is $F$” is made true by the fact that $x$ exists in virtue of the entity in question, and that, therefore, the explanation “$x$ exists because nom $(y$ is $F)$ exists” is metaphysically more revealing than the explanation “$x$ exists because $y$ is $F$.” This justifies the assertion of sentences of the form

[$x$ exists because $y$ is $F$] because [$x$ exists because nom $(y$ is $F)$ exists],

of which (15), above, is an instance, and which, given the definition of FED, are equivalent to

[$x$ exists because $y$ is $F$] because [$x$ existentially depends^FB on nom $(y$ is $F)$],

which states that a claim of existential grounding (“$x$ exists because $y$ is $F$”) is grounded in a claim of (full-blooded) existential dependence.

Someone could observe that there is a more direct way for the advocate of existential dependence to defend Dep $\rightarrow$ Ground*. This way is the semantic strategy, according to which the two sentences “$x$ exists” and “$y$ is $F$” – which are, respectively, the explanandum and the explanans in “$x$ exists because $y$ is $F$” – represent two states of affairs, respectively $\langle x$ exists$\rangle$ and $\langle y$ is $F\rangle$, which stand in turn in a relation of existential dependence (since the former obtains in virtue of the latter). What I would say concerning this semantic strategy is that, although it is actually a viable one, it is not, appearances notwithstanding, a more direct strategy than the one I propose. In fact, what an advocate of the semantic strategy should do is, first, advocate a Tractarian-like metaphysics according to which the world is a world of facts and, second, defend a semantics according to which (true) sentences in natural languages are endowed with meaning in virtue of representing these facts. On the contrary, someone pursuing the strategy I propose can just care about the metaphysical question concerning what in the world, if anything, makes true predicative sentences of the form “$y$ is $F$,” remaining neutral with regard to the semantic question concerning the nature of sentence-meaning.

5 Conclusion: Beyond Ideology

An advocate of the third way I sketched above concedes a lot to the supporter of grounding: in fact, he accepts Df.Dependence, and, therefore, also Ground $\rightarrow$ Dep, which is a consequence of it. He can, however, point out that to acknowledge that claims of existential dependence are better analyzed by notions such as “ground” or “because” is tantamount to granting the notion of grounding ideological priority in respect of the notion of existential dependence. He can further stress that once the ideological match has been conceded to the supporter of grounding, there is still room for the view that existential dependence wears, so to say, the metaphysical trousers in structuring reality. The reason is that, according to the supporter of the third way, existential explanations, explanations of why a given thing exists, are existential not only in their explanandum but, at the most fundamental level, also as far as their explanans is concerned: the existence of things is, at bottom, always metaphysically explained by the bare existence of other things. The metaphysically fundamental truths are existential truths, and the “is so and so” of predication is always grounded in the “is” of existence.

Since this view does not concern the best way of defining existential dependence but the form of the metaphysically fundamental truths, it is not sufficient to counter it by putting forward a definition of existential dependence in terms of grounding. What is instead needed is a defense of the idea that existential explanations may have, also at the more fundamental level, predicative and not existential form as far as their explanans is concerned, and that the metaphysically fundamental truths may not be truths stating that a given thing exists. However, they can be, and most often are, truths to the effect that a given thing is a certain way.

The supporter of grounding, unless he is content with his ideological victory, should meet this challenge.

As it is customary, angle brackets have here the same role as the clause “the fact that,” that is to say, the role of forming names of facts starting from sentences. So $\langle$$p$$\rangle$ means the same as “the fact that $p$.” Square brackets will instead be used in some of the examples below as means of syntactic disambiguation.↩︎
I borrow this example from Schnieder (2020) just replacing his preferred fictional characters (Ahab, Gandalf) with mine.↩︎
There is a wide consensus that grounding is an explanatory notion and that claims of grounding may be expressed both in relational and non-relational form (cf. Correia 2010; Correia and Schnieder 2012; Fine 2012a; Berker 2018), leaving open the questions concerning which of these two forms is metaphysically more revealing (what Schnieder 2020 calls the “grammatical question”), and whether grounding reduces to metaphysical explanation or is a relation “backing” this kind of explanation (cf. Maurin 2019; Glazier 2020). Here and in many of the examples in the rest of the article, I express claims of grounding in non-relational form using the connective “because,” although nothing I will say in the course of the paper depends essentially on this choice. My preference for the “because” form has two reasons. Firstly, using this form will make many of the examples less cumbersome; secondly, since many arguments that will be discussed below are concerned with the explanatory dimension of grounding, it seems to me that to express the examples involved in them in the “because” form, which is the form in which explanations are usually expressed, will make it easier to appreciate and assess them.↩︎
That grounding explanations should be systematic has been stressed by deRosset (2013).↩︎
Notice that (7) is an instance of “$\exists z(\exists F(\langle y$ is $F\rangle$ grounds $\langle x$ exists$\rangle)$ because $x$ existentially depends on $z)$” which is the non-relational version of the consequent of the conditional open formula contained in Dep $\rightarrow$ Ground*.↩︎
That pleonastic entities lack explanatory power concerning the rest of the world is a consequence of Schiffer’s view that admitting their existence does not affect in any way our prior picture of the world. Schiffer tries to make this idea more precise by saying that if the concept of an $F$ is a pleonastic concept then, for any theory $T$, the theory $T$¹ obtained by adding to $T$^{$\sim$$F$} (the theory that results from restricting each quantifier of $T$ to things that aren’t $F$). This concept, together with the corresponding linguistic transformations (which he calls “something-from-nothing transformations”), is a conservative extension of $T$^{$\sim$$F$}, that is, for any sentence $S$ expressible in the language of $T$^{$\sim$$F$}, if $S$ is a theorem of $T$¹, it is also a theorem of $T$^{$\sim$$F$} (Schiffer 2003, 57).↩︎
From now on I will use “nom ($p$)” as a schematic expression whose instances are the nominalizations of (the instances of) “$p$.”↩︎
In the explanans of (12), I use a non-necessitated version of the definition of PED. This is, in this context, harmless and renders (12) less cumbersome.↩︎
A definition of existential dependence along these lines was put forward for the first time by Lowe (1998, 145). Both Schnieder (2006b) and Correia (2005) view Df.Dependence as an improvement on a definition of this kind. A definition of truth-making along the lines of FED can be found in Schnieder (2006a) and in Caputo (2007).↩︎
Concerning the relation between the extensions of FED and PED, it should be noticed what follows: on the one hand, there are entities which full-bloodedly existentially depend on others without doing it pleonastically. For instance, the singleton of Socrates full bloodedly existentially depends on Socrates, since the singleton exists because Socrates exists, but doesn’t depend pleonastically on him, since the nominalizations of “Socrates exists” are expressions such as “the fact that Socrates exists” or “Socrates’ existence” which do not refer, if they refer at all, to Socrates. On the other hand, as I stressed before, the fact that an entity pleonastically existentially depends on another does not guarantee that the former full-bloodedly existentially depends on the latter, since it is not granted either that the term “nom ($p$)” refers or that, provided it does, it refers to a non-pleonastic entity.↩︎
Notice that (15) is just (7) with “because” substituted for “grounds.”↩︎

References

Berker, Selim. 2018. “The Unity of Grounding.” Mind 127(506): 729–777, doi:10.1093/mind/fzw069.

Cameron, Ross P. 2008. “Turtles all the Way Down: Regress, Priority and Fundamentality.” The Philosophical Quarterly 58(230): 1–14, doi:10.1111/j.1467-9213.2007.509.x.

Caputo, Stefano. 2007. “Truth-making: What it Is Not and What it Could Be.” in Metaphysics and Truthmakers, edited by Jean-Maurice Monnoyer, pp. 275–312. Philosophische Analyse / Philosophical Analysis n. 18. Heusenstamm b. Frankfurt: Ontos Verlag, doi:10.1515/9783110326918.275.

Correia, Fabrice. 2002. “Existential Dependence and Cognate Notions.” Th{\`e}se de doctorat, Genève: Département de philosophie, Université de Genève. Book publication: Correia (2005).

—. 2005. Existential Dependence and Cognate Notions. Analytica: Investigations in Logic, Ontology, and the Philosophy of Language. München: Philosophia Verlag. Book publication of Correia (2002), doi:10.2307/j.ctv2nrzj0h.

—. 2010. “Grounding and Truth-Functions.” Logique et Analyse 53(211): 251–279.

Correia, Fabrice and Schnieder, Benjamin Sebastian. 2012. “Grounding: an Opinionated Introduction.” in Metaphysical Grounding. Understanding the Structure of Reality, edited by Fabrice Correia and Benjamin Sebastian Schnieder, pp. 1–36. Cambridge: Cambridge University Press, doi:10.1017/CBO9781139149136.001.

deRosset, Louis. 2013. “No Free Lunch.” in Varieties of Dependence: Ontological Dependence, Supervenience, and Response-Dependence, edited by Miguel Hoeltje, Benjamin Sebastian Schnieder, and Alexander Steinberg, pp. 243–270. Basic Philosophical Concepts. München: Philosophia Verlag, doi:10.2307/j.ctv2nrzhj9.

Fine, Kit. 1982. “Acts, Events and Things.” in Proceedings of the 6th International Wittgenstein Symposium: Language and Ontology, edited by Werner Leinfellner, Eric Russert Kraemer, and Jeffrey C. Schank, pp. 97–105. Schriftenreihe der Österreichischen Ludwig Wittgenstein Gesellschaft n. 8. Wien: Hölder-Pichler-Tempsky.

—. 1995. “Ontological Dependence.” Proceedings of the Aristotelian Society 95: 269–290, doi:10.1093/aristotelian/95.1.269.

—. 2001. “The Question of Realism.” Philosophers’ Imprint 1(2).

—. 2012a. “Guide to Ground.” in Metaphysical Grounding. Understanding the Structure of Reality, edited by Fabrice Correia and Benjamin Sebastian Schnieder, pp. 37–80. Cambridge: Cambridge University Press, doi:10.1017/CBO9781139149136.002.

—. 2012b. “The Pure Logic of Ground.” The Review of Symbolic Logic 5(1): 1–25, doi:10.1017/S1755020311000086.

Glazier, Martin. 2020. “Explanation.” in The Routledge Handbook of Metaphysical Grounding, edited by Michael J. Raven, pp. 121–132. Routledge Handbooks in Philosophy. Milton, Abingdon: Routledge, doi:10.4324/9781351258845.

Koslicki, Kathrin. 2013. “Ontological Dependence: An Opiniated Survey.” in Varieties of Dependence: Ontological Dependence, Supervenience, and Response-Dependence, edited by Miguel Hoeltje, Benjamin Sebastian Schnieder, and Alexander Steinberg, pp. 31–64. Basic Philosophical Concepts. München: Philosophia Verlag, doi:10.2307/j.ctv2nrzhj9.

Lowe, Edward Jonathan. 1998. The Possibility of Metaphysics: Substance, Identity, and Time. Oxford: Oxford University Press, doi:10.1093/0199244995.001.0001.

Maurin, Anna-Sofia. 2019. “Grounding and Metaphysical Explanation: It’s Complicated.” Philosophical Studies 176(6): 1573–1594, doi:10.1007/s11098-018-1080-0.

Orilia, Francesco. 2009. “Bradley’s Regress and Ungrounded Dependence Chains: A Reply to Cameron [cf. Cameron (2008)].” Dialectica 63(3): 333–341, doi:10.1111/j.1746-8361.2009.01189.x.

Schiffer, Stephen. 2003. The Things We Mean. Oxford: Oxford University Press, doi:10.1093/0199257760.001.0001.

Schnieder, Benjamin Sebastian. 2006a. “Truth-Making Without Truth-Makers.” Synthese 152(1): 21–46, doi:10.1007/s11229-004-7905-8.

—. 2006b. “A Certain Kind of Trinity: Dependence, Substance, Explanation.” Philosophical Studies 129(2): 393–419, doi:10.1007/s11098-005-4636-8.

—. 2020. “Grounding and Dependence.” Synthese 197(1): 95–124, doi:10.1007/s11229-017-1378-z.

How Existential Dependence Can Ground Existential Grounding

1 Schnieder’s Point against Dep \(\rightarrow\) Ground

2 The Ontological Challenge

3 Too Easy?

4 The Third Way

5 Conclusion: Beyond Ideology

References