Abstract
Inter-level mechanistic explanations in the sciences have long been a focus of philosophical interest, but attention has recently turned to the compositional character of these explanations which work by explaining higher level entities, whether processes, individuals or properties, using the lower level entities they take to compose them. However, we still have no theoretical account of the constitution or parthood relations between individuals deployed in such explanations, nor any accounts of multiple constitution. My primary focus in this paper is to outline a positive account of the constitution/part-whole relations between individuals posited in inter-level mechanistic explanations that takes constituents in the sciences to be ‘working parts’. Using this account, I then go on to illuminate the nature, and varieties, of multiple constitution that we find in the sciences and provide a starting theoretical framework for multiple constitution as well.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Throughout the paper, following the common scientific practice, I shall use ‘constituent’ and ‘part’ interchangeably, and hence also use the terms ‘constitution’ and ‘part-whole’ to refer to the same compositional relation. I thus intend my account of constitution as an account of parthood in the ways laid out in later sections. The reader should note that, again following scientific practice, I take ‘composition’ to be a generic term covering a family of relations which hold between a range of different categories of entity. Analytic metaphysicians use ‘composition’ in a different fashion, but I follow the scientific usage going back at least to Newton and stretching down to the present.
See for example the ‘Flat’ accounts of Kim (1993), Shoemaker (2001), Clapp (2001) or Wilson (2009), and my own contrasting ‘Dimensioned’ account (Gillett 2002, 2003 or 2007b). The Flat and Dimensioned accounts differ over whether realized and realizer properties can fail to overlap in powers or processes, require compositional relations between other categories of entity, or can be had by distinct individuals, amongst other issues, but both kinds of account agree in taking realization to be an ontological determination relation between entities in nature.
It bears emphasis that the swathe of recent work on “constitution” and/or parthood in analytic metaphysics is not focused either upon scientific cases or their concepts, thus having a rather different focus from my work here. I examine such work below and I argue that such metaphysical accounts do not fit the scientific notion.
The account of constitution in the sciences I defend below was originally outlined in Gillett (2007a), but in this paper I refine this account in response to our findings about neuroscientific cases. Shoemaker (2007) independently offers a related account, but in this paper I focus on outlining my own view and leave comparison to Shoemaker’s related account to other venues.
The reader should note that I use the term ‘entity’ as a catch-all referring to powers, properties, individuals, processes, etc. I use the term ‘individual’ to refer to individuals. Throughout the paper, I focus on property/relation instances and, unless otherwise noted, I use the terms ‘property’ or ‘properties’ to refer to property/relation instances.
There are other accounts of “production” distinct in nature from “powerful production”. The reader should therefore treat the latter as either an alternative view of “production” or an account of a narrower concept.
It is worth noting a couple of points. First, I do not take all causal processes to be determinative. For, second, note that all productive processes will likely be causal processes, depending upon ones account of causation, but not all causal processes line-up with corresponding productive ones. Thus some causal processes may not be determinative. And there are other differences depending on what account one gives of causation. For instance, it is also worth marking that productive processes are plausibly not transitive unlike causal processes under many accounts of them. Thus if individual X has the power to produce acidity in a certain liquid, and this level of acidity produces erosion in metal, then it does not follow that X produces erosion in metal. X only produces something else that produces such erosion.
I use ‘level’ in the way common in the sciences, to mean a ‘compositional’ level of compositionally related individuals, properties, powers and processes. As Wimsatt (1976, 2007), and also Craver (2007), make clear, such levels are ‘local’ in the sense that they are built around particular compositional relations holding under certain background conditions. The section will show that our examples support these points.
We thus find Machamer et al. (2000) telling us that: “Mechanisms are entities and activities organized such that they are productive of regular changes from start or set-up to finish or termination conditions”.
In conversation, recent writers on mechanistic explanation like Machamer, and also Craver, each endorse this interpretation of their work as embracing properties.
With what modal strength do components X1–Xn, under the right conditions $, non-causally result in Y? I am unsure what methods could be reliably used to answer this kind of question, but we can say that it is at least nomologically necessary that X1–Xn, under $, result in Y. Whether some stronger modality holds remains to be seen.
For a fuller list of the general properties of scientific composition see Gillett (Forthcoming) or Aizawa and Gillett (Unpublished).
Notice that here we are looking at whether the productive, rather than merely causal, processes of the neuron are implemented by the relevant individuals. The neuron engages in all manner of causal processes, but our concern is only on its productive processes.
With what modal strength is it true that, say, constitution and realization relations are always found together? Once more, I am unsure what methods could be reliably used to answer this kind of question, but we can again say that it is at least nomologically necessary that wherever we find individuals related by the constitution relations outlined we will always find realization relations. Whether some stronger modality holds remains to be shown.
What kinds of refinements might be necessary? Should we make the stronger demand that most, rather than many, of the individuals in the team bear productive and/or powerful relations? Or should we weaken the condition to say that some, rather than many, individuals in the team bear such relations? Answering such questions is a task for future work, since it requires a wider survey of examples than I can pursue here.
I have starred this schema to mark that it is not an element of my integrated framework for scientific composition.
We need to have reference to a range of background conditions beyond that under which the individual may now be found because the constituted individual may only ground certain processes in virtue of its present properties under some background condition. So the constituents must be such that under this other background condition they would implement this process of the constituted individual.
Such conceptually non-reductive, and interconnected, accounts are common for families of basic notions. For example, compare the conceptually non-reductive accounts that Shoemaker (1980) gives of properties, powers and causation, or that Plantinga (1979) gives of modal notions. Crucially, such work shows that providing intertwined accounts of basic concepts can obviously be highly illuminating.
Here the account only refers to background condition that the individual is presently found under, since the constituents are only required to realize all the present properties of the individual—hence limiting us to the present background condition.
The account of implementation is too long to include here, but see Gillett (Unpublished-a, Unpublished-b) and Aizawa and Gillett (Unpublished). However, I can note the theory schema for comprising which is more compact and stated as follows:
(Comprising) Powers C1-Cn, had by individuals s1-sn (or individual s*), comprise the power C*, had by individual s* under background condition $, if and only if the productive processes grounded by the triggering and manifestation of powers C1-Cn, under conditions $, would together implement the productive processes grounded by the triggering and manifestation of C*, under $, but not vice versa. (Gillett 2007b, p. 202).
For example, what does our account say about mitochondria that are now widely taken to be vestiges of bacteria that entered into a symbiotic relationship with another bacterium? At this point, the properties and processes of mitochondria are taken to realize and implement the properties and processes of eukaryotic cells, so mitochondria are taken to be constituents under our account. And this accords with scientific practice.
In my (2007a) I independently offered examples using the neutrino to illustrate that constituents are individuals whose properties realize those of the constituted entity. Mellor’s example of the photon, and my case of the neutrino, are each largely causally inert and each thus shows that spatial containment is not sufficient for constituency.
Williams (2008) raises this type of concern about Mellor’s account.
Note that I am not claiming that any organism within a human, or other organism, cannot be a constituent of it. The Dimensioned account of constituents as working parts implies no such thing on its own. We need to look case by case at the relevant details. So, for example, beneficial organisms in our gut, or elsewhere, could be counted as parts of our bodies under my account if they implement processes of our bodies. As I noted above, mitochondria in cells count as parts of cells, and our bodies, for just this reason.
As an aside, note that we can see why the other parts of the car are still parts, for the car enters into many processes—washing its windows for example—and the other parts (such as the wipers) are components of those other processes that the car enters into, rather than the process of moving around.
Whether these relations pass the stringent requirements laid down in philosophical frameworks for interventionist accounts is another, and vexed, question, since writers like Baumgartner (2010) have raised important concerns about the utility of using the interventionist framework in cases involving compositional relations. However, my claim here is simply that we have crude manipulability relations between the relevant entities, rather than ones fulfilling the strict conditions arising from particular philosophical frameworks.
Notice that Craver’s including a condition like (i) in a criterion of implementational relevance is far from ad hoc for two reasons. First, Craver is not seeking an account parthood or implementation. And, second, we have excellent evidence, from our panoply of inter-level mechanistic explanations, that all individuals are constituted and so working scientists now routinely assume that everything in nature is composed. Craver’s condition (i) simply reflects the working assumption that scientists consequently make.
Once again, concerns arise here (Baumgartner (2010)) about whether these conditions can be satisfied under the strict accounts of mutual manipulability embodied in interventionist accounts arise.
Unsurprisingly, using our accounts of compositional notions we can produce more ontologically oriented criteria of implementational relevance as we unpack the various requirements for scientific composition. Remember we are asking when an individual X, which is a part of S, is a component of S’s process of ψ-ing grounded by its property. For example, given the connections between realization and implementation we can offer this simple criterion amongst others:
(Implementational Relevance) An individual X is a component of a process that implements S’s One-Step process of ψ-ing, grounded by S’s property H, under condition $, only if, under $, (i) X is a constituent of S; and (ii) X’s properties P1-Pn are amongst the realizers of S’s property H.
How does this kind of account relate to Craver’s mutual manipulability view? In fact, Craver says something closely related, since he tells us that “in the constitutive relation [i.e. when we have implementational relevance], a token instance of the property of ϕ is, in part, constituted by an instance of the property of ψ…” (Craver 2007, p. 153) where the relation between properties posited here, though again termed “constitution”, looks like the realization relation in the account just outlined. So Craver may be close to accepting this type of account already. However, the twin tendencies to downplay the role of properties in recent work, as well as the failure to draw out all the compositional relations at play, have apparently precluded his providing such an account explicitly.
See Koslicki (2008) for a summary of related criticisms of standard mereology based upon its neglect of properties and relations.
For example, Dimensioned accounts of constitution and realization have now been used to engage the issue of what conception scientists have of the individuals that are background conditions for compositional relations and how they differ, if at all, from the entities treated as constituents (Gillett 2007a). Such accounts have also illuminated the nature of multiple realization in the sciences (Gillett 2003; Aizawa 2007; Aizawa and Gillett 2009a, b, Unpublished); and allowed us to understand, and then assess, the ‘nothing but’ claims of so-called ‘scientific reductionists’ and their ‘emergentist’ opponents (Gillett 2007b, Forthcoming), amongst other issues.
As the new area of neuroproteomics advances (Bayes and Grant 2009), we may eventually be able to pinpoint which new proteins are synthesized during synaptogenesis and become constituents at this point, and other points of development. Neuroproteomics is focused on cataloging the proteins we find in the neuron, and various areas of it like its pre-synaptic and post-synaptic structures, at various points in its development. Neuroproteomics may thus eventually document at what point in its developmental cycle we have which new proteins as constituents in the neuron.
The following schemata adapts that given for multiple realization in Aizawa and Gillett (2009a, b). However, this earlier account of multiple realization plausibly needs to be altered to allow for corresponding Qualitative, Quantitative and Mixed forms of multiple realization with the composition of property instances mirroring the Qualitative, Quantitative and Mixed forms of multiple constitution we have now illuminated.
Some readers may be concerned that adding (iv), and the notion of a ‘level’, leaves us with what many regard as the vague notion of a ‘level’. However, there are reasonably clear scientific notions of a ‘level of entities’, under some condition, as entities that do, or can, participate with each other in the same chains of production, i.e. causal mechanisms, under those conditions and/or which participate in processes that together implement other processes. Furthermore, elsewhere I have outlined a precise account of such a ‘level’, so we have both informal and more precise reasons to reject such concerns (Gillett Unpublished-a).
References
Aizawa, K. (2007). The biochemistry of memory consolidation: A model system for the philosophy of mind. Synthese, 155, 65–98.
Aizawa, K., & Gillett, C. (2009a). Levels, individual variation and massive multiple realization in neurobiology. In J. Bickle (Ed.), Oxford handbook of philosophy and neuroscience (pp. 539–582). Oxford: Oxford University Press.
Aizawa, K., & Gillett, C. (2009b). The (multiple) realization of psychological and other properties in the sciences. Mind and Language, 24, 181–208.
Aizawa, K., & Gillett, C. (Unpublished). Building nature, building neuroscience: Composition, explanation and methodology.
Alberts, B., Johnson, A., Lewis, J., & Raff, M. (2007). The molecular biology of the cell (7th ed.). New York: Garland.
Baker, M., & Goetz, S. (Eds.). (2010). The soul hypothesis. New York: Continuum.
Baumgartner, M. (2010). Interventionism and epiphenomenalism. Canadian Journal of Philosophy, 40, 359–384.
Bayes, A., & Grant, S. (2009). Neuroproteomics: Understanding the molecular organization and complexity of the brain. Nature Reviews Neuroscience, 10, 635–646.
Clapp, L. (2001). Disjunctive properties: Multiple realizations. Journal of Philosophy, XCVIII, 111–136.
Craver, C. (2007). Explaining the brain. New York: Oxford University Press.
Cummins, R. (1983). The nature of psychological explanation. Cambridge, MA: MIT Press.
Fodor, J. (1968). Psychological explanation. New York: Random House.
Gillett, C. (2002). The dimensions of realization: A critique of the standard view. Analysis, 62, 316–323.
Gillett, C. (2003). The metaphysics of realization, multiple realizability and the special sciences. Journal of Philosophy, 100, 591–603.
Gillett, C. (2007a). Hyper-extending the mind? Setting boundaries in the special sciences. Philosophical Topics, 35, 161–187.
Gillett, C. (2007b). Understanding the new reductionism: The metaphysics of science and compositional reduction. The Journal of Philosophy, 104, 193–216.
Gillett, C. (2007c). A mechanist manifesto for the philosophy of mind: A third way for functionalists. Journal of Philosophical Research, 32, 21–42.
Gillett, C. (2010). Moving beyond the subset model of realization: The problem of qualitative distinctness in the metaphysics of science. Synthese, 177, 165–192.
Gillett, C. (2013). Understanding the sciences through the fog of ‘functionalism(s)’. In P. Hunneman (Ed.), Functions: Selection and mechanisms (pp. 159–181). Dordrecht: Kluwer.
Gillett, C. (Forthcoming). Reduction and emergence in science and philosophy. Cambridge: Cambridge University Press.
Gillett, C. (Unpublished-a). Making sense of levels in the sciences: Composing powers, properties, parts and processes.
Gillett, C. (Unpublished-b). Implementing mechanism: Getting ontologically serious about mechanisms.
Gillett, C., & Loewer, B. (Eds.). (2001). Physcialism and its discontents. Cambridge: Cambridge University Press.
Hasker, W. (1999). The emergent self. Ithaca: Cornell University Press.
Hull, D. (1974). Philosophy of biological science. New York: Prentice Hall.
Karp, G. (2008). Cell and molecular biology (5th ed.). Hoboken, NJ: Wiley.
Kim, J. (1993). Supervenience and mind. Cambridge: Cambridge University Press.
Kim, J. (1998). Mind in a physical world. Cambridge, MA: MIT Press.
Koslicki, K. (2008). The structure of objects. New York: Oxford University Press.
Levitan, B. (2001). The neuron: Cell and molecular biology. New York: Oxford University Press.
Lowe, J. (2009). Dualism. In McLaughlin, Beckermann & Walter (Eds.) (2009), pp. 66–84.
Machamer, P., Darden, L., & Craver, C. (2000). Thinking about mechanisms. Philosophy of Science, 67, 1–25.
McLaughlin, B., Beckermann, A., & Walter, S. (2009). The Oxford handbook of philosophy of mind. New York: Oxford University Press.
Mellor, H. (2008). Micro-composition. Royal Institute of Philosophy Supplements, 83, 81–106.
Melnyk, A. (Unpublished). From new mechanistic explanation to good old-fashioned reduction in one easy step. Paper presented to the Society for the Metaphysics of Science group session at the 2010 Central APA Conference.
Piccinini, G. (2004). Functionalism, computationalism, and mental states. Studies in the History and Philosophy of Science, 35, 811–833.
Plantinga, A. (1979). The nature of necessity. New York: Oxford University Press.
Plantinga, A. (2006). Against materialism. Faith and Philosophy, 23, 3–32.
Shoemaker, S. (1980). Causality and properties. In Van. Inwagen (Ed.), Time and cause. Dordrecht: Reidel.
Shoemaker, S. (2001). Realization and mental causation. In C. Gillett & B. Loewer (Eds.) (2001).
Shoemaker, S. (2007). Physical realization. Oxford: Oxford University Press.
Simons, P. (1987). Parts: A study in ontology. Oxford: Clarendon Press.
Williams, R. (2008). Working parts: A reply to Mellor. Royal Institute of Philosophy Supplements, 83, 81–106.
Wilson, J. (2009). Determination, realization, and mental causation. Philosophical Studies, 145, 149–169.
Wimsatt, W. (1976). Reductionism, levels of organization and the mind-body problem. In G. Globus, G. Maxwell, & I. Savodnik (Eds.), Consciousness and the brain. New York: Plenum Press.
Wimsatt, W. (2007). Re-engineering philosophy for limited beings. Cambridge, MA: Harvard University Press.
Acknowledgments
Thanks to Ken Aizawa, Matt Babb, Carl Craver, Andrew Melnyk, Jackie Sullivan and two anonymous referees for helpful comments on an earlier version of the paper. Thanks also to the audience at the 2011 Delaware Workshop on Neuroscience where the paper was presented. Special thank to Ken Aizawa and especially Fred Adams for organizing a highly productive workshop.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Gillett, C. Constitution, and Multiple Constitution, in the Sciences: Using the Neuron to Construct a Starting Framework. Minds & Machines 23, 309–337 (2013). https://doi.org/10.1007/s11023-013-9311-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11023-013-9311-9