# Existence, Really? Tacit Disagreements about ``Existence'' in Disputes about Group Minds and Corporate Agents
**Authors:** Johannes Himmelreich
**Published in:** Synthese 198, 2021
**DOI:** [10.1007/s11229-019-02379-3](https://doi.org/10.1007/s11229-019-02379-3)
*This is the author's manuscript. Please cite the published version.*
---
**Johannes Himmelreich**
> A central dispute in social ontology concerns the existence of group minds and actions. I argue that some authors in this dispute rely on rival views of existence without sufficiently acknowledging this divergence. I proceed in three steps in arguing for this claim. First, I define the phenomenon as an implicit higher-order disagreement by drawing on an analysis of verbal disputes. Second, I distinguish two theories of existence – the theory-commitments view and the truthmaker view – in both their eliminativist and their constructivist variants. Third, I examine individual contributions to the dispute about the existence of group minds and actions to argue that these contributions have an implicit higher-order disagreement. This paper serves two purposes. First, it is a study to apply recent advances in meta-ontology. Second, it contributes to the debate on social ontology by illustrating how meta-ontology matters for social ontology.
## Introduction
Many central questions of social ontology concern existence. Are there group minds? Are there group actions over and above individual actions? In this paper, I argue that some authors who engage in disputes about these questions rely implicitly on rival views of existence.[^1] “Existence” and its cognates can be understood in different ways. When two authors operate with different understandings of “existence” and do not sufficiently acknowledge this divergence, then an important reason for their disagreement is likely to be overlooked – by themselves or other participants to the dispute. I provide textual evidence to suggest that this has happened in individual cases in disputes about the existence of group minds and group actions.[^2]
I proceed in three steps. First, I draw on an analysis of verbal disputes to theorize implicit rival views about “existence.” Second, I distinguish two theories of existence – the theory-commitments view and the truthmaker view – in both their eliminativist and their constructivist variants and observe that the theory-commitments view naturally lends stronger support to the existence of group minds and actions than the truthmaker view. In a third step, I offer textual evidence to suggest that some authors in their explicit conclusions about the existence of group minds and actions implicitly rely on rival views of “existence.”
To motivate the worry that parties to the same dispute implicitly hold rival views of “existence,” consider the following two quotes from authors who disagree about whether there are group, or corporate, agents.
> Given that talk of group agents is not readily translatable into individualistic terms, and given that it supports a distinct way of understanding and relating to the social world, we can think of such entities as autonomous realities. (List and Pettit 2011, p. 6)
>
> \[W\]e have addressed the question of whether genuine corporate agents are required for the truth of corporate agency discourse. When we see what the corporation is by its nature, … and uncover the mechanisms of the expression of corporation agency, we find behind the veil only individual agents. (Ludwig 2017, p. 295)
The authors have an obvious first-order dispute. List and Pettit, the authors of the first quote, contend that corporate agents exist. Ludwig, the author of the second quote, contends that corporate agents do *not* exist. But the authors also have a less-acknowledged higher-order disagreement: The way in which the authors think about existence differs starkly. Whereas List and Pettit emphasize the fact that “talk of group agents is not readily translatable,” Ludwig instead proceeds on semantic considerations contending that only individual agents are “required for the truth of corporate agency discourse.”
These two ways of thinking about existence resemble to two meta-ontological views about existence, namely, the theory-commitments view and the truthmaker view of existence (Cameron 2008a, 2010; Quine 1948, 1969). The authors’ first-order dispute might disappear, at least in part, if the authors agreed on one theory of existence.
Authors in disputes about group minds and corporate agents generally do not clarify their meta-ontology, let alone discuss how their claims about existence may rest on theories of existence. Social ontology seems sometimes short on meta-ontology.[^3] My aim is to illustrate through an examination of individual instances the importance of meta-ontology for social ontology.
## Implicit Higher-order Disagreements
I see higher-order disagreements as a semantic phenomenon of speakers who understand constituent expressions in a sentence differently. Seen in this way, I define higher-order disagreements by drawing on an existing analysis of verbal disputes (Chalmers 2011),[^4] even if higher-order disagreements, unlike verbal disputes, need not be associated with a sense of pointlessness of the dispute at hand. This idea of implicit higher-order disagreements can be stated more precisely as follows, where *S* stands for some statement and *T* for some expression (or term) in *S*.
A dispute over *S* is an implicit higher-order disagreement if and only if
1)
the parties disagree about the meaning of *T*, and
2)
the dispute over *S* arises partly in virtue of this disagreement regarding *T*, and
3)
the parties do not acknowledge or do not seem to be aware of (b).
Three clarifications are in order (cf. Chalmers 2011). First, as mentioned, I do not claim or suggest that disputes in social ontology are pointless. Unlike merely verbal disputes, implicit higher-order disagreements can bear on substantive issues.
Second, to count as a higher-order disagreement, the disagreement over *T* must *explain* the existence of a dispute over *S*. This explanatory relation can be approximated with a counterfactual variant. Would the dispute over *S* persist if we were to bar the use of *T*? If the dispute over *S* would be reduced or disappear altogether, then the underlying disagreement explains the first-order dispute. Call this the *elimination test*. As an alternative to the elimination test, you can employ the *subscript gambit*. When there are different ways of understanding *T*, disambiguate *T* (e.g. distinguish *T*1 and *T*2). If the dispute over *S* disappears when you disambiguate *T* in this way, then the underlying disagreement explains the dispute (Chalmers 2011, pp. 523–524; 530–532; 547).
Third, a dispute may arise only *in part* because of a meta-linguistic disagreement concerning the meaning of *T*. A first-order dispute might remain even if the higher-order disagreement about “existence” is resolved. To the extent that these further disagreements arise among proponents of the same meta-ontological view, I call these *internal disputes*.
Moreover, disputes about existence face the problem of *vocabulary exhaustion* (Chalmers 2011, p. 543). That is, in the case of “existence,” the elimination test seems infeasible. One cannot easily bar or substitute the use of “existence” or “there is” insofar as these terms are needed to formulate the underlying disagreement. To avoid this problem, I use a version of the subscript gambit instead. I distinguish different views about existence, which we could label “existence1”, “existence2”, and so forth.
## Four Views of Existence
I now present two established rival views of existence as foils and in a stark contrast to one another in order to illustrate the potentially vast ramifications that an implicit higher-order disagreement about “existence” may have. I do not aim to develop each of the views about existence in detail or even defend them.[^5] Each of these two views about existence can be held in either an eliminativist or a constructivist variant, which gives us four views of existence in total.
### The Theory-commitments View
To be, or so goes the slogan of W.V.O. Quine, is to be the value of a bound variable (Quine 1948, p. 32). This quotation represents a distinct approach to ontology. At the heart of this approach is the idea that science is our guide to what exists and that scientific theories generate ontological commitments. The ontological commitments of a theory are uncovered by *regimenting* the theory, that is, by expressing it in a specialized language that makes semantic functions transparent, for example, by expressing it in first-order logic (Sennet and Fisher 2013). A theory is ontologically committed to those things over which the existential quantifier in the theory’s regimented version ranges. The totality of things that exist are the ontological commitments of all of our best scientific theories.
This view of Quine (1948, 1969) has long been seen as something like the orthodoxy in ontology and it has been developed and defended in detail (Egerton 2016; Van Inwagen 2009). Quine’s view is what I call a *theory-commitments view of existence*.
> ***Theory-commitments View*** Some thing exists if and only if it is in the range of the quantifiers in the regimented statements of our best theories.
For the purposes here, because authors in social ontology generally do not make their underlying view of existence explicit, I am less concerned with what this view says but more with how it can be recognized. Before this backdrop, two indicators of this view are noteworthy.
First, the object of interest to a Quinean meta-ontology is theories. This contrasts with the truthmaker view that will take as the object that incurs ontological commitments, primarily, true sentences in ordinary language. Heuristically, if an author concentrates on scientific theories in an ontological investigation, this suggests, to some measure, that she operates before the backdrop of the theory-commitments view.
Second, ontological commitments are arrived at by regimenting theories and paraphrasing away those parts of theories that are not required in order to avoid unnecessary commitments (Quine 1969, p. 96). The terms “regimenting” and “paraphrasing” are Quine’s. Authors might operate with a theory-commitments view of existence but use different terms such as “translating,” things being “dispensable,” or “explanatory superfluous,” which capture the idea behind Quine’s method of paraphrase, albeit in different terms. For the purposes of this paper, I take each of these expressions as an indication that an author is operating with a theory-commitments view of existence.
How much of a theory can be paraphrased away? For Quine (1951, pp. 14–15), a theory has two competing “aspects”: ontology and ideology. The ontology of a theory is what is bound by its quantifiers. The ideology of a theory, by contrast, are its predicates and “what *ideas* are expressible in the language of the theory.” How much of a theory can be paraphrased away hinges on a trade-off between ontological parsimony (“ontology”) and expressivity (“ideology”). One can schematically take the extreme ends of this trade-off to distinguish two variants of the theory-commitments view: an eliminativist and a constructionist variant.[^6]
The *eliminativist variant* of the theory-commitments view evaluates theories by their ontological parsimony (Egerton 2018, p. 41). The “best theory” is the one with the fewest ontological commitments, even if such parsimony comes at the expense of expressivity. The *constructivist variant*, by contrast, evaluates theories by their expressivity. The “best theory” is the one that employs predicates and terms in a way that improves understanding or allows simple or common-sense explanations (cf. Egerton 2019).
The eliminativist variant figures prominently in the argument for the view that combines the claim that good explanations may involve reference to corporate agents or group intentions (*explanatory non-individualism*) with the claim that these things do not actually exist (*ontological individualism*).[^7] This view rests on two key assumptions. First, it rests on a reductionist claim that theories involving corporate agents or group minds can be reduced to or paraphrased into theories involving only individual agents and individual minds. Second, it assumes the eliminativist variant of the theory-commitments view.
But as Bratman (2018) points out, this reduction can be constructive. If we instead assume the constructivist variant of the theory-commitments view, then explanatory non-individualism implies ontological non-individualism; in other words, if our best explanations of actual phenomena involve group minds, then there are group minds. Yet, this ontological non-individualism is consistent with the reductionist claim and the availability of a paraphrase (Himmelreich 2017, p. 88).
In sum, an emphasis on theories and the use of expressions such as “dispensability”, “explanatory superfluous,” indicate that authors subscribe to Quine’s theory-commitments view. Within this view, eliminativist or constructivist variants are available. Moreover, there is room for ample *internal disputes* among those who hold the theory-commitments view. What counts as explanatory superfluous or dispensable is not obvious. Are references to collective entities dispensable in our explanations? Authors might disagree over good-making features of theories, such as between a theory’s simplicity and its explanatory power, and how these should be balanced against each another.[^8]
### The Truthmaker View
Truthmaker theory is, very broadly, a comprehensive approach in metaphysics spelling out the general picture of truth as depending on being. The motivating intuition is that a sentence is true because of truthmaker(s) that make the sentence true.[^9] Truthmaker theory is generally taken to entail two main tenets (Armstrong 2004; Cameron 2008b). First, every truth has at least one truthmaker (*truthmaker maximalism*). Second, the existence of a truthmaker necessitates the truth of which it is the truthmaker (*truthmaker necessitarianism*). This general program of truthmaker theory has given rise, more specifically, to a meta-ontological view about existence (Cameron 2008a, 2010; Rettler 2016).
> ***Truthmaker View** (Constructivist Variant)* Some thing exists if and only if it is a truthmaker of at least one actually true sentence.
Two observations are in order. First, the objects that incur ontological commitments on the truthmaker view are statements understood very broadly as all sentences in a language (Rettler 2016).[^10] Second, ontological commitments are incurred through the truthmaking relation. The ontological commitments of a sentence are its truthmakers. Expressions such as “truthmakers,” “is made true,” entities existing “because of,” or properties being ascribed to something “in virtue of,” can all count as indicators of a truthmaker view of existence at work insofar as the truthmaking relation is understood as an explanatory relation (Asay 2018).
Just as the theory-commitments view, also the truthmaker view can be distinguished into a constructivist and an eliminativist variant. The constructivist variant is stated above. Being a truthmaker for a true sentence is necessary and sufficient for existence. The eliminativist variant is more restrictive.
> ***Truthmaker View** (Eliminativist Variant)* Some thing exists if and only if it is a fundamental truthmaker of at least one actually true sentence.
The eliminativist variant introduces a further necessary condition for existence. Only those things exist that are *fundamental* truthmakers of true statements. On this eliminativist variant of the truthmaker view, ontological commitment and ordinary notions of existence come apart. To accommodate this implication, proponents of the truthmaker view distinguish what exists from what *really* exists (Fine 2001). Truthmaker theorists accept that tables exist, in an ordinary sense of “exist,” but they deny that tables *really* or *fundamentally* exist, in the sense that we are ontologically committed to these entities (Rettler 2016). This quest for fundamentality is a point of similarity between the truthmaker theory and neo-Aristotelian theories of grounding (Schaffer 2009).
### Relevance of Meta-ontology
On which meta-ontology an author operates matters. Of course, when two authors disagree about the existence of group minds or corporate actions, it might be because they hold different views of existence or different variants of the same view of existence. But the relevance of meta-ontology can be discerned without examining individual of instances of higher-order disagreements. First, the eliminativist variant of any of the two views of existence yields a smaller ontology than its respective constructivist variant. Second, the eliminativist theory-commitments view is generally taken to yield a larger ontology than the eliminativist truthmaker view (Asay 2018, p. 915; Cameron 2010; Schaffer 2008).[^11] The eliminativist truthmaker view leads to a “radically minimal” ontology such that “what has real being – that there really is – is what makes the true theory of the world true, and this is a proper subset of the things that the theory says there are” (Cameron 2010, p. 250).
For illustration, suppose that all sentences referring to collectives are made true by individuals. When the sentences referring to collectives are part of our best regimented theories, then collectives would exist according to the theory-commitments view. But these entities would not exist according to the truthmaker view because the sentences in the theory are made true by individuals (or other more fundamental entities).
Moreover, also the details of meta-ontological assumptions matter. Two authors can have a higher-order disagreement even if both subscribe to the theory-commitments view. For example, two authors may disagree about whether a paraphrase is available. Likewise, internal disputes may arise among proponents of the truthmaker view. For example, the analysis of the truthmaking relation is a formidable venue for an internal dispute, not unlike arguments over “indispensability” and “explanatory value” among proponents of the theory-commitments view (Asay 2018). Furthermore, the logics of the truthmaking relation – its reflexivity, transitivity, and asymmetry – are contested (Rodriguez-Pereyra 2015). In short, there are ample disputes beyond the different conditions for existence – on both sides of the aisle. Many questions will remain open, even after participants agree on a common ground of existence theories.
## Implicit Higher-order Disagreements in Social Ontology
Be it about views of existence, their variants, or about some internal dispute – assumptions about meta-ontology matter for first-order disputes over what exists. I argue that such higher-order disagreements have remained implicit in individual instances in disputes about group minds and corporate actions. Some authors rest their arguments on the theory-commitments view, others on the truthmaker view of existence without sufficient acknowledgement of this divergence. Moreover, we can see something like a correlation between implicit assumptions about meta-ontology and explicit conclusions about what exists.
I classify arguments along two dimensions. The first dimension concerns the existence of corporate actions or group minds. This is the dimension about the first-order dispute. I call *realism* the view that group minds (or actions) exist and *eliminativism* the view that group minds (or actions) do not exist. Because the same author(s) may put forth more than one argument for a given conclusion, I classify not only by authors but also by their arguments. The second dimension is that of meta-ontology. This is the dimension of the higher-order disagreement. Is an author assuming the theory-commitments view or the truthmaker view of existence, and in which variant? Insofar as authors do not state their meta-ontological assumptions explicitly, I classify them by the indicators proposed above and by trying to plausibly reconstruct the argument with either theory of existence.[^12]
I first look at authors defending realism, then I will look at some authors defending eliminativism about group minds and corporate actions. Specifically, I will concentrate on the arguments of List and Pettit (List 2016; 2011; Pettit 2003, 2009) in defending a realist position and on the arguments of Ludwig (2007a, 2007b, 2011, 2014) and Rupert (2005, 2011, 2014) for an eliminativist position.
### Realism: List and Pettit
I start by revisiting the quote from List and Pettit (2011) from the introduction.
> Given that talk of group agents is not readily translatable into individualistic terms, and given that it supports a distinct way of understanding and relating to the social world, we can think of such entities as autonomous realities. (List and Pettit 2011, p. 6)
List and Pettit (2011, p. 6) make two points for realism about group agents. The first point concerns that sentences referring to collectives are not “readily translatable.” This point, since it is about translation and not truthmaking or explanation, relates plausibly to the Quinean idea of paraphrase. When something is not translatable, it is not dispensable. The second point concerns the “distinct way of understanding” that is opened up by accepting realism of group agents. This point relates to the explanatory value of a theory. Part of the reason for why theories about group agency are not translatable seems to be that too much of a theory’s expressivity would be lost.[^13]
Pettit (2003, 2009) defends realism about group minds for similar reasons.
> \[A\]scribing reality to group agents \[…\] means that there is good reason to seek explanations at a level where group agents are treated as agents in their own right without always exploring the nuts and bolts of individual contribution. (Pettit 2009, p. 89)
Pettit argues that there is a stable and rational pattern of a group’s behavior. Postulating group minds is thereby the best explanation of at least some collective phenomena. Again, ontology is taken to be a matter of theories and explanations. Roth (2014) reconstructs Pettit’s argument in a way that is clearly suggestive of the theory-commitments view:
> If the group-theoretic concepts find their home as *indispensable elements of the explanatory or predictive theory T*, then we have reason to believe the group-mind thesis \[that there are such group minds\]. (Roth 2014, p. 139 my emphasis)
The meta-ontological assumption – that existence is a matter of best theories – is not stated explicitly.[^14] Recently, List (2016, p. 298) has made this meta-ontological assumption explicit in stating that he assumes the “Quinean definition of ontological commitment” in an argument for group agency. List also draws on the assumption that “\[o\]ur best social-scientific theories … attribute belief-desire agency of the functionalist kind to (some of) the collectives involved.” The expression of “best … theories” is clearly in line with the theory-commitment view of existence.
Other proponents of a realism of group agency are Theiner et al. (2010) who argue for a realism about group cognitive states. In what set off a subsequent dispute with Ludwig (2015), Theiner et al. argue that well-established models explaining individual cognitive behavior can be applied to collective systems. They suggest that to settle the question of whether groups have minds “\[w\]e should … be asking whether specific cognitive models that work at the level of individuals also work at the level of groups.” (2010, p. 379) They come to the conclusion that “the cognitive capacities of groups are amenable to the same type of mechanistic explanation as the cognitive capacity of individuals (or their brains).” (2010, p. 391) In other words, Theiner et al. (2010) argue that group minds exist because the same models of cognitive science that are used to describe individual cognition can be used to investigate collective cognition. Notably, by taking models as theories, they depart from the Quinean orthodoxy of taking theories to be sets of sentences in first order logic. Other than that, however, their approach is still decidedly in the spirit of the theory-commitments view.
In sum, we have examples of two sets of authors who argue for realism about group minds or corporate agency on the assumption of the theory-commitments view of existence. All authors stress expressivity and explanatory power of theories or the indispensability of concepts such as collective intentions or group agency. This emphasis on theoretical virtues suggests that these authors might see themselves in an argument with proponents of the theory-commitments view of a more eliminativist variant and not necessarily with proponents of a truthmaker theory of existence. But also such internal disputes over, for example what makes a “best theory” and, relatedly, to what extent ontological commitments can be paraphrased away, can lead to implicit higher-order disagreements. Moreover, beyond the possibility of such internal disputes, there is also evidence for an implicit higher-order disagreement between rival views of existence.
### Eliminativism: Ludwig and Rupert
Let me again start by revisiting the quote from the introduction. Ludwig (2017) argues against group agency and, among others, targets the view of List and Pettit (2011).
> \[W\]e have addressed the question of whether genuine corporate agents are required for the truth of corporate agency discourse. When we see what the corporation is by its nature, …we find behind the veil only individual agents. (Ludwig 2017, p. 295)
The argument is, in short, that there are no corporate agents because they are not “required for the truth of corporate agency discourse.” Ludwig implicitly assumes the eliminativist variant of the truthmaker view of existence. His argument also draws on the assumption that ordinary discourse incurs ontological commitments through its fundamental truthmakers.
In earlier works, Ludwig (2007a, 2007b, 2011, 2014) similarly argues for an eliminativist thesis by investigating sentences about collective actions[^15] “from the standpoint of semantics,” such as “we insulted the host.”
> \[“We insulted the host”\] is ambiguous between a distributive and a collective reading. On the distributive reading, it is made true, for example, by my insulting the host before dinner and your insulting him afterwards. On the collective reading*,* it would be made true, for example, by our deliberately talking in our host’s presence as if he were not there. Here we do it together. …
>
> We do something together, then, when we (and only we) are all agents … via our various individual actions, of a single event. … \[W\]e can see that plural action sentences do not commit us to group agents per se. (Ludwig 2011, pp. 47–48)
Ludwig’s argument is that there are no group agents because, even on a collective reading, the sentence “we insulted the host” is made true by actions of individuals. Ludwig’s exclusive attention to sentences in ordinary discourse and in particular the use of the locution “is made true” suggests that his argument rests on the truthmaker view of existence.
Similarly, Ludwig (2014) explicitly derives conclusions about the existence of group agents from truth conditions about collective action sentences.[^16] Elsewhere he describes his goal as “to give the truth conditions for the collective reading of plural sentences.” (2007b, n. 17) Of course, truth conditions are not truthmakers, but Ludwig uses the terms interchangeably. Moreover, the way in which he uses the terms suggests that his topic is in fact truthmakers rather than truth conditions because he examines what makes token action sentences are true.
Similarly, Ludwig (2015) responds to Theiner et al. (2010) that their arguments “so obviously fail” and are “question begging” and seeks to register a “terminological drift or sliding” in their arguments. My diagnosis of this situation is that Theiner et al. operate with the theory-commitments view, whereas Ludwig assumes a truthmaker view of existence.
At the same time, Ludwig might offer more than one line of argument (2007a, 2007b, 2011). In addition to the argument based on the truthmaker view of existence reconstructed above, specifically Ludwig (2007b) has the resources to provide an argument based on the theory-commitments view(Himmelreich 2017).
Consider another author arguing for an eliminativist position. Rupert (2014) examines statements such as “Microsoft intends to …” that seem committed to group minds because they reference Microsoft’s intention.[^17]
> \[I\]f the statements in question are true, but their truth-makers involve nothing more than the instantiation of cognitive properties by individuals \[…\], then there are no group cognitive states in the relevant sense. (Rupert 2014, p. 106)
Rupert’s argument rests on the assumption that the statement “Microsoft intends to…” is made true by things other than the collective entity. Therefore, or so his argument goes, the statement does not in fact generate an ontological commitment to a group mind. This argument seems best reconstructed as drawing on the truthmaker view of existence, which Rupert even invokes with “their truth-makers.”
A similar argument can be found in Rupert (2005).
> A theory of mental representation seems to apply to group states *only in virtue of* the contents of individuals’ mental representations. (Rupert 2005, p. 183 my emphasis)
>
> \[A\]lthough canonical representations issued by Gilbert’s court \[i.e. a putative group mind\] may have causal efficacy, they have it in virtue of individuals’ states. (Rupert 2005, p. 179)
This argument also can be reconstructed as based on the truthmaker view. With the truthmaker view in the background, the assumption that collective mental representations are true only in virtue of individual representations immediately leads to the eliminativist conclusion about group minds.[^18] Rupert uses the locution “in virtue of,” instead of “are made true by.” But, as observed above, this locution relates to the idea of truthmaking (Asay 2018).
## Conclusion
One way to make progress in philosophy is to identify unrecognized underlying disagreements (Chalmers 2011). In recent years, the literature on group minds and corporate agents has made such progress. For example, it is now widely recognized that whether groups have minds depends on whether some version of functionalism is true.[^19] Moreover, it seems well recognized that whether groups have minds depends on whether having a mind requires phenomenal consciousness. Finally, whether groups are agents may depend on whether agency requires bodily movements (Himmelreich 2018). These issues – functionalism, phenomenal consciousness and embodiment as requirements for mind and agency – are underlying disagreements that impact first-order disputes about group minds and corporate agents.
In this paper, I argued for a further underlying disagreement that has, as of yet, not been recognized. I distinguished different views about existence and illustrated how they matter: Which view is chosen impacts how readily certain first-order conclusions can be reached. Truthmaker theory naturally lends stronger support to eliminativism about group minds. Authors considered here who favor a truthmaker view of existence reject the existence of group minds, whereas authors who assume a theory-commitments view of existence tend to accept the existence of group minds. Implicit assumptions about theories of existence seem to correlate with explicit conclusions about what exists. Yet, the authors generally do not distinguish rival views of existence and their relevance for first-order conclusions. Their dispute seems to rest on a higher-order disagreement that has remained implicit.
Beyond the disagreement between rival views of “existence,” proponents of the different theories of existence may also have internal disputes over questions such as whether group-theoretic concepts are dispensable or to what extent group-theoretic concepts increase the explanatory value of theories. This again illustrates and adds to my conclusion: Meta-ontology matters for social ontology. Assumptions about existence, to which some authors in social ontology so far only appeal tacitly, need to be brought out in the open.
**Acknowledgements** In writing this paper I benefitted from an early conversation with Daniel Nolan, discussions with Jesse Saloom, Sebastian Köhler and Ryan Cox, comments from two anonymous referees for this journal, and discussions at the European Congress of Analytic Philosophy (ECAP9).
This is a post-peer-review, pre-copyedit version of an article published in *Synthese*. The final authenticated version is available online at:
## References
Armstrong, D. M. (2004). *Truth and Truthmakers*. Cambridge: Cambridge University Press.
Asay, J. (2018). We don’t need no explanation. *Philosophical Studies*, *175*(4), 903–921. doi:10.1007/s11098-017-0898-1
Bratman, M. E. (2018). Review of From Plural to Institutional Agency: Collective Action II. *Notre Dame Philosophical Reviews*. https://ndpr.nd.edu/news/from-plural-to-institutional-agency-collective-action-ii/. Accessed 4 March 2019
Cameron, R. P. (2008a). Truthmakers and ontological commitment: or how to deal with complex objects and mathematical ontology without getting into trouble. *Philosophical Studies*, *140*(1), 1–18. doi:10.1007/s11098-008-9223-3
Cameron, R. P. (2008b). Truthmakers, Realism and Ontology. *Royal Institute of Philosophy Supplements*, *62*, 107–128. doi:10.1017/S135824610800060X
Cameron, R. P. (2010). How to have a radically minimal ontology. *Philosophical Studies*, *151*(2), 249–264.
Chalmers, D. J. (2011). Verbal Disputes. *Philosophical Review*, *120*(4), 515–566. doi:10.1215/00318108-1334478
Egerton, K. (2016). Getting off the Inwagen: A Critique of Quinean Metaontology. *Journal for the History of Analytical Philosophy*, *4*(6). doi:10.15173/jhap.v4i6.2962
Egerton, K. (2018). Found Guilty by Association: In Defence of the Quinean Criterion. *Ratio*, *31*(1), 37–56. doi:10.1111/rati.12150
Egerton, K. (2019). It Takes More than Moore to Answer Existence-Questions. *Erkenntnis*. doi:10.1007/s10670-019-00107-4
Epstein, B. (2007). Ontological individualism reconsidered. *Synthese*, *166*(1), 187–213. doi:10.1007/s11229-007-9272-8
Fine, K. (2001). The Question of Realism. *Philosopher’s Imprint*, *1*(2). http://hdl.handle.net/2027/spo.3521354.0001.002. Accessed 27 February 2018
Himmelreich, J. (2015). From Individual to Collective Intentionality: New Essays, edited by Sara Rachel Chant, Frank Hindriks and Gerhard Preyer. Oxford University Press, 2014, 225 pages. *Economics and Philosophy*, *31*(3), 479–486. doi:10.1017/S0266267115000218
Himmelreich, J. (2017). The Paraphrase Argument Against Collective Actions. *Australasian Journal of Philosophy*, *95*(1), 81–95. doi:10.1080/00048402.2016.1153685
Himmelreich, J. (2018). Agency and Embodiment: Groups, Human–Machine Interactions, and Virtual Realities. *Ratio*, *31*(2), 197–213. doi:10.1111/rati.12158
List, C. (2016). What is it like to be a group agent? *Noûs*, *52*(2), 295–319.
List, C., & Pettit, P. (2011). *Group Agency: The Possibility, Design, and Status of Corporate Agents*. Oxford: Oxford University Press.
Ludwig, K. (2007a). Foundations of Social Reality in Collective Intentional Behavior. In S. L. Tsohatzidis (Ed.), *Intentional Acts and Institutional Facts: Essays on John Searle’s Social Ontology*. Springer.
Ludwig, K. (2007b). Collective Intentional Behavior from the Standpoint of Semantics. *Noûs*, *41*(3), 355–393. doi:10.1111/j.1468-0068.2007.00652.x
Ludwig, K. (2011). Adverbs of Action and Logical Form. In T. O’Connor & C. Sandis (Eds.), *A Companion to the Philosophy of Action* (pp. 40–9). John Wiley & Sons.
Ludwig, K. (2014). The Ontology of Collective Action. In S. R. Chant, F. Hindriks, & G. Preyer (Eds.), *From individual to collective intentionality: new essays*. Oxford: Oxford University Press.
Ludwig, K. (2015). Is Distributed Cognition Group Level Cognition? *Journal of Social Ontology*, *1*(2), 189–224. doi:10.1515/jso-2015-0001
Ludwig, K. (2017). Do corporations have minds of their own? *Philosophical Psychology*, *30*(3), 265–297. doi:10.1080/09515089.2017.1295636
Pettit, P. (2003). Groups with minds of their own. In F. F. Schmitt (Ed.), *Socializing metaphysics: the nature of social reality*. Rowman & Littlefield.
Pettit, P. (2009). The reality of group agents. In C. Mantzavinos (Ed.), *Philosophy of the Social Sciences: Philosophical Theory and Scientific Practice* (pp. 67–91). Cambridge: Cambridge University Press.
Quine, W. V. O. (1948). On What There Is. *The Review of Metaphysics*, *2*(5), 21–38.
Quine, W. V. O. (1951). Ontology and ideology. *Philosophical Studies*, *2*(1), 11–15. doi:10.1007/BF02198233
Quine, W. V. O. (1969). Existence and Quantification. In *Ontological Relativity and Other Essays* (pp. 91–113). Columbia University Press.
Rettler, B. (2016). The General Truthmaker View of ontological commitment. *Philosophical Studies*, *173*(5), 1405–1425. doi:10.1007/s11098-015-0526-x
Rodriguez-Pereyra, G. (2015). Grounding is not a Strict Order. *Journal of the American Philosophical Association*, *1*(3), 517–534. doi:10.1017/apa.2014.22
Roth, A. S. (2014). Indispensability, the Discursive Dilemma, and Groups with Minds of Their Own. In S. R. Chant, F. Hindriks, & G. Preyer (Eds.), *From Individual to Collective Intentionality* (pp. 136–162). Oxford: Oxford University Press. Accessed 27 May 2015
Rupert, R. D. (2005). Minding One’s Cognitive Systems: When Does a Group of Minds Constitute a Single Cognitive Unit? *Episteme*, *1*(03), 177–188. doi:10.3366/epi.2004.1.3.177
Rupert, R. D. (2011). Empirical Arguments for Group Minds: A Critical Appraisal. *Philosophy Compass*, *6*(9), 630–639. doi:10.1111/j.1747-9991.2011.00420.x
Rupert, R. D. (2014). Against Group Cognitive States. In S. R. Chant, F. Hindriks, & G. Preyer (Eds.), *From Individual to Collective Intentionality* (pp. 97–111). Oxford University Press. http://www.oxfordscholarship.com/view/10.1093/acprof:oso/9780199936502.001.0001/acprof-9780199936502-chapter-5. Accessed 27 May 2015
Schaffer, J. (2008). Truthmaker commitments. *Philosophical Studies*, *141*(1), 7–19. doi:10.1007/s11098-008-9260-y
Schaffer, J. (2009). On What Grounds What. In D. J. Chalmers, D. Manley, & R. Wasserman (Eds.), *Metametaphysics: New Essays on the Foundations of Ontology* (pp. 347–383). Oxford: Oxford University Press.
Sennet, A., & Fisher, T. (2013). Quine on Paraphrase and Regimentation. In G. Harman & E. Lepore (Eds.), *A Companion to W.V.O. Quine* (pp. 89–113). Oxford, UK: John Wiley & Sons, Ltd. doi:10.1002/9781118607992.ch5
Sylvan, K. L. (2012). How to be a redundant realist. *Episteme*, *9*(03), 271–282. doi:10.1017/epi.2012.16
Theiner, G., Allen, C., & Goldstone, R. L. (2010). Recognizing group cognition. *Cognitive Systems Research*, *11*(4), 378–395. doi:10.1016/j.cogsys.2010.07.002
Van Inwagen, P. (2009). Being, Existence, and Ontological Commitment. In D. J. Chalmers, D. Manley, & R. Wasserman (Eds.), *Metametaphysics: New Essays on the Foundations of Ontology* (pp. 472–506). Oxford: Oxford University Press.
## Appendix
The corpus relevant to this article can be identified using the following search parameters in the advanced fuzzy search of philpapers.org. My argument does not depend on generalizations about this corpus, nor does my argument permit strong generalizations about this corpus. I give the search parameters in full to encourage investigations that allow such generalizations.
**Must appear**: (group\* plural\* collecti\* corporati\*) (action\* agen\* behavio\* mind\* intentional\* representatio\* cogniti\*)
**Excellent indicators**: "group agency" "group mind" "group minds" "group agent" "group cognitive" "collective mind"
**Good indicators**: "collective action" "groups with minds" "collective intentional" "collective intentionality" "collective representation" "group cognition"
**Irrelevance**: Stit Anscomb\* Dewey Husserl Heidegger Rousseau Ricoeur Theolog\* Thomist\* consciou\* phenomen\* coloni\* duties history epistemi\* religio\* Lowith Scheler Stein Obligations Responsibility Aesthetics "group rights" "moral agency" photography african awareness team\* memory introduction feasibility children legislat\* plan\*
**Year**: 2003-2018
professional authors only
published only
**Minimal relevance:** 5
[^1]: I here give the argument for a hypothesis I entertained elsewhere (Himmelreich 2015, p. 484).
[^2]: I understand “group minds” as the claim that a group has intentional states. I understand “group actions” as the claim that a group is an agent of an action.
[^3]: This claim refers to a corpus of articles and books in philosophy with titles such as “group agency” or “group mind,” published between 2003 and 2018, omitting works on historical figures, normative or formal aspects. See the appendix for details.
[^4]: I amend the definition of Chalmers (2011, p. 522) in three ways to make it more amenable for the purposes here. First, I do not distinguish between broad and narrow disputes. Second, I define “verbal dispute” as encompassing partly verbal disputes. Third, I add condition (c).
[^5]: In particular, the truthmaker view can be incorporated in the larger project of grounding and neo-Aristotelian metaphysics (Schaffer 2009). Although this latter project might provide a more nuanced discussion of the truthmaker position, for expositional reasons I relate my investigation to the earlier statement of the view.
[^6]: I take these terms from Bratman (2018).
[^7]: I avoid the term “methodological individualism” because methodological individualism “is commonly divided into two different claims,” an ontological and an explanatory one (Epstein 2007, p. 188).
[^8]: Since the totality of things that exist are the ontological commitments of all of our best scientific theories, an author’s ontological commitments depend on what they take to be the best theories.
[^9]: Sometimes truthmaker theory is formulated in terms of propositions instead of sentences. The difference is otherwise important but should not matter for the purposes here.
[^10]: Cameron (2008a, 2010) also occasionally talks of truthmakers of “theories.” This is consistent because Cameron takes theories to be sets of sentences.
[^11]: I suspect this subset claim holds only if one assumes that our best regimented theories are committed (in the sense of the theory-commitments view) to more than their fundamental truthmakers.
[^12]: Some authors views may differ from my classification, often trivially so because my classification is very coarse-grained, whereas authors’ views will be more refined.
[^13]: Although the quote above reads “talk of group agents” List and Pettit are concerned with group agency “both in common and in scientific discourse” (2011, p. 4).
[^14]: Yet, my reading that their argument rests on the theory-commitments view can be supported also by the fact that this view is considered the default (Schaffer 2009, p. 347).
[^15]: From this investigation of collective action sentences, Ludwig infers a thesis about collective intentionality. Ludwig argues that no collective agency implies no collective intentionality: “If there are no collective agents, there are no collective intenders.” (2007b, p. 365) This is rather surprising because Ludwig seems to get the implication backwards. Although intentionality is necessary for agency, contrary to what Ludwig assumes, agency is not necessary for intentionality.
[^16]: Ludwig (2014, p. 125) writes that “groups per se are not primitive agents of any events … so far as the truth conditions of collective action sentences go.”
[^17]: Rupert (2014) provides no references as to whom he is arguing against. It is plausible to take him as a party in the dispute over group minds together with List and Pettit, because he argues for the negation of the claim that they defend.
[^18]: Yet, this argument could instead be read as being based on the theory-commitments view similar to arguments that Rupert makes elsewhere (2005, 2011, 2014). In this case, the dispute would be an internal one between Rupert and authors such as List and Pettit over the constructivist vs. the eliminativist variant of the theory-commitments view.
[^19]: This dependence only concerns some arguments for group minds. Sylvan (2012, p. 271) observes that List and Pettit (2011) hold “functionalist background views.” Ludwig (2015, p. 204) remarks that “\[i\]f functionalism is not an adequate theory of cognition, then the bar for showing that groups have genuine cognitive properties becomes higher.” He remains somewhat skeptical and writes in Ludwig (2017, n. 8): “No one has in fact made a good case that corporations … have a mind on the basis of functionalism. No one has produced an adequate functionalist theory…. What arguments there are consist largely in hand-waving and affirming the consequent (e.g., List & Pettit, 2011) or begging the question under the guise of application of a parity principle (e.g., Theiner, Allen, & Goldstone, 2010) … .” On functionalism and agency see also List (2016).