We are proud to present the winning essay of our 2015 competition.
Congratulations Jonny Lee!
Towards a New Understanding of the Senses: Does Sensory Pluralism offer a satisfactory solution to the problem of individuating the senses?
The problem of individuating the senses is the problem of how we should distinguish and type the senses. Philosophers have traditionally responded to the problem by attempting to establish privileged individuation criteria. I will suggest that this approach is committed to “sensory monism” (SM), that there exists a definitive taxonomy of the senses. This essay will examine the prospects of an alternative approach called sensory pluralism (SP). Sensory pluralism, as I take it, claims that there are no independent facts about how the senses are taxonomized. Instead, how we should individuate the senses depends inextricably on our explanatory practices. I will begin by outlining the problem, the traditional response, and the difficulty that response faces given developments in perception research. Following this, I will outline the pluralist picture before finally assessing two objections to this alternative.
2. The Problem of Individuating the Senses
The traditional response to the question of how to individuate the senses can be characterised by the attempt to establish a privileged “sparse” criterion (MacPherson, 2011: 129). “Sparse” here means a commitment to the existence of a small number of limited and distinct senses. Each criterion attempts to establish what it is to be a sense modality, or type of sense, by that modality’s individuation properties. To borrow from MacPherson (2011), the major contending criteria include:
- Representational Criterion: A sense modality is such because of its representational properties, the kinds of things it represents.
- Phenomenal Character Criterion: A sense modality is such because of its experiential character.
- Sense Organ Criterion: A sense modality is such because of the activity, composition or structure of a particular organ or sensory system.
- Proximal Stimulus Criterion: A sense modality is such because of the kinds of proximal stimuli that affect the relevant sensory organ e.g. electromagnetic waves or odor molecules.
MacPherson characterises these criteria, as I have above, as “sparse”. But I take it that they are committed to a further assumption: that how the senses taxonomize is a fact independent of our explanatory practices, and therefore that there exists a definitive account of how we ought to individuate the senses. Following Fulkerson (2014), I will call views that are committed to a definitive answer to the question of individuating the senses “monist”. It is important to note that the distinctions between sparse/dense and monist/pluralist are different. One can believe, for example, that there are very many senses but that there is nonetheless a definitive taxonomy, and therefore be some form of dense monist. However, much of the motivation for monism is, I suggest, derived from the apparent existence of just a few distinct and definitive senses. Likewise, as I will suggest below, the ever increasing number and complexity of the senses identified in perception research puts pressure on the possibility of explaining what sort of definitive account of the senses would be possible. I will touch on this again below.
As MacPherson (2011) points out, empirical observations create havoc with the neat taxonomy implied by sparse criteria. I suggest there have been two relevant trends in perception research in recent years. The first trend is the discovery that in various contexts our “folk five senses” operate with greater complexity and variety than our relatively small “folk taxonomy” suggests. The second trend is research into multisensory integration, that is, the integration of different sensory streams to create new sensory information. Perception science, it seems, has both proliferated and integrated the senses.
An example of the first trend is found in olfaction research. Researchers typically consider two senses of smell: retronasal and orthonasal. Orthonasal olfaction is the result of odorant stimulation of the nostrils. Retronasal olfaction is the stimulation by odorants internally to the mouth, most notably contributing to flavour perception. Though the two olfactions have much in common (both involve stimulating the olfactory epithelium), they are functionally distinct enough for researchers to usefully distinguish the two (e.g. Landis et al, 2005; Small et al, 2005).
A classic example demonstrating the second trend is the “ventriloquist effect”. This effect involves the misidentification of an auditory source due to visual cues. It is the ventriloquist effect that is responsible for us hearing voices emanate from actors’ mouths when watching films, despite the different and often distant audio source. This effect can be explained by a model of optimal combination of visual and auditory information (Alais & Burr, 2004). In short, visual information is allowed to dominate due to bimodal integration; the creation of a unique sensory experience.
So is olfaction one sense or two? Is the audio-visual experience in the ventriloquist effect the combination of two discrete senses or the creation of a single new one? The problem is that research reveals these senses to be in some ways like one and in some ways like two. As Fulkerson says,
Once we allow that the sensory modalities interact, and do so pervasively at multiple levels of sensory processing, with effects at all levels of our psychology… then it becomes difficult to make sense of what, exactly, these individual senses might be. (2014: Online, Section 1)
I suggest the sparse commitments of the criteria surveyed above struggle to account for research revealing the myriad complexity of sensory operations.
In light of the advances made by perception research, maintaining any sparse criteria seems intolerable. The logical space we are left with contains two options: both of which involve dropping sparse criteria. The first option drops any sparse criteria but maintains SM and attempts to construct another definitive criterion, yet one consistent with trends in perception science. Again rejecting sparseness does not entail rejecting monism. One can maintain that there are very many but definitive senses. Something like this seems close to McPherson’s own position. The second option abandons the sparse criteria and with it any form of SM, instead adopting a form of pluralism. For brevity I cannot assess the first option, and it is not the purpose of this essay to rule out every possible theory of individuation, merely to defend the plausibility of SP. For now I will simply note that proponents of the first view face the challenge of providing a privileged criterion which is both consistent with the complexities revealed by perception research, but which does not collapse into pluralism (observe that MacPherson’s positive account is at least consistent with SP.)
3. Painting a Pluralist Picture
One way of interpreting perception research is as signifying the impossibility of a definitive taxonomy. Instead we could adopt SP. Pluralism is a family of approaches that claim there is more than one kind, answer or principle concerning a particular explanandum. In this context pluralism pushes the view that how we ought to individuate the senses is relative to our contingent and varying explanatory practices and interests. To borrow from MacPherson, “there is no need to choose between the standard criteria that have been proposed as ways of individuating the senses” (2011: 123). Furthermore (and this is where MacPherson may disagree), we should not attempt to replace the standard sparse criteria with any privileged criteria.
SP rejects the premise that there is an essential and independent fact about how the senses taxonomize, and therefore that there is any essential criteria to be discovered. Rather SP allows for a plurality of criteria. The answer to whether olfaction is one sense or two depends on the context in which the use of olfaction is embedded e.g. whether we are interested in studying cross-olfactory neurons that transduce odorants into neural signals, (Rospers, 1998); or in contrast, the processing of odor compounds through the nostrils or mouth (Heilman & Hummel, 2004).
SP should not be mistaken for scepticism. For the sensory-pluralist the point is not that we cannot know what the senses really are, but rather that,
…as a matter of fact, senses really are lots of things, and what counts as explanatory in our theorizing about sensory interactions depends on how we’re carving the systems up and what we are trying to explain. (Fulkerson, 2014: Online, Section 8).
SP turns the problem of individuation on its head by doubting its validity: what is the utility, the pluralist asks, in this quest for definitive criteria? Responding to the disagreements concerning individuation Ward concludes that, “for most scientists who study the senses this question is not seriously debated because the question itself is regarded as ambiguous or unanswerable” (2009: 31).
SP thus offers a solution to the problem of individuation in the form of dissolution, by rejecting the assumption that there is any definitive answer to be found.
4. Objections to Sensory Pluralism
There are two ways that one might object to SP. Both deny that we must reject privileged criteria. The first strategy claims that we already possess a successful privileged criterion, and therefore SP is false. The second strategy admits that we currently lack an account of a privileged criterion, but argues that such a criterion must be possible as pluralism trivialises or is otherwise explanatory impotent, and therefore cannot be correct.
One way to resist the pull of SP is to affirm that a privileged individuating criterion exists. A common problem facing any such strategy is how to avoid question begging; that is, how to avoid the unprincipled assumption of a particular form of individuation.
Perhaps the most intuitive version of this objection takes as its starting premise that it is the folk psychological senses, namely the “folk five” that we are referring to when posing the problem of individuation. From this perspective we already know that we have five senses, and we are searching for a criterion that explains this fact. Proponents acknowledge that perception science is related to the folk five, and may well explain them, but claim that the senses are not the kinds of things that are amenable to revision. According to these philosophers, to discover that there are two kinds of olfactory processing is to reveal something of scientific interest, but it is “changing the subject” to further claim we have discovered two senses of smell (Nudds, 2004: 5). I take it that something like this view is expressed by Nudds when he says,
There have been authors who attempt to give a ‘scientific’ account of the senses; but they do nothing to show that they haven’t simply changed the subject. Whatever they are giving an account of, it’s not the senses as we commonly understand them. (ibid)
This paves the way for a view like Nudds’ own “social kinds theory” which aims to explicate our folk taxonomy in terms of the senses’ role in folk explanations (Nudds, 2011). Though I cannot detail Nudds’ own positive account here, what matters for our purposes is the privileging of the folk five senses. It is worth noting however that Nudds’ positive account is itself perfectly compatible with some form of SP, however under SP it is part of just one of many legitimate ways of individuating the senses- I will touch on this further below.
I have two responses to this objection. The first is to doubt the obviousness that our folk psychological individuations are as coherent and clear-cut as we might first think, and thus a shaky foundation on which to build a definitive taxonomy of the senses. I will note two complications for the idea that it is “obvious” that we have five senses (Nudds, 2004: 1). Firstly, there is substantial evidence of cultural variation in how people individuate the senses (see Howes, 1991). For example, the Hausa of Nigeria at least linguistically distinguish between only two senses (“Gani”, referring to vision, “Ji”, referring to hearing, tasting, smelling, touching, feeling and knowing) (Lillard, 1998). Secondly, it is not clear that the folk five taxonomy is actually so conspicuous and invariable. For example, people are typically willing to accept the possibility of extra-human senses within the animal kingdom e.g. ultrasound (as found in bats) or electroreception (as found in sharks). We also seem willing to accept cases of fictional senses, such as “Spidey sense” or “X-ray vision”. Following MacPherson (2011), I suggest that our conventional answer to “how many senses are there?” might have to do more with our cultural history, the content of our education, and the answers expected given the context of the question, and less to do with an intuitive conviction that there are definitively five senses. These observations cast doubt on the plausibility of the folk psychological senses functioning as a foundation for a successful monism.
Now this response does not by itself count as an argument for SP and it may be possible to construct a version of monism that can explain people’s acceptance of animal and fictional senses as genuine senses. But this response does call into question one particularly convincing version of a privileged taxonomy. If the strength of the folk five senses is not sufficient to ground monism then the proponent will need to look elsewhere to motivate their position.
My second response suspects this objection of drawing arbitrary boundaries. Why treat our folk five senses as the real subject of the individuation problem when we also possess a scientific language of the senses? The claim then is that the privileged criteria objection to SP draws an arbitrary line around a particular account of the senses then declares this to be the definitive account. But though such an account may be an account, it requires further argumentation to establish that it is the account. Why not, for example, claim that the “folk five” are a truth in so far as it concerns folk psychology (if indeed this is true), but that it does not provide a satisfactory framework as it concerns mature scientific investigation into the complex relationships involved in multimodal processing? The SP position legitimises our folk individuations, whatever they are, while at the same time allowing for scientific development to inform other understandings of the senses. Notably this response could be flipped for the parallel objection that science provides us with a definitive individuation of the senses, and a privileged criterion exists that reflects that individuation. As with folk psychology, I am sceptical that science does provide an invariable individuation (scientific individuation depends on levels of abstraction, subject matter etc.), but even if it did, further argument would have to be made to establish that it was this individuation that incurs privilege.
If SP allows for myriad possible ways of individuating the senses to be equally legitimate, does SP trivialise the individuation problem? After all, we started with what seemed like a substantial philosophical question, “how should we individuate the senses?” One might feel that we have substituted that debate with a dismissal, that accepting the permissibility of SP is a form of admitting defeat. This objection refuses to surrender the apparently intuitive appeal of some privileged answer to how we ought to individuate the senses, for to adopt its denial is to adopt an absurd position.
Relatedly, SP appears to miss something important about what we originally set out to accomplish. Once all the folk psychological and scientific facts are “on the table”, it seems something remains unanswered. One possible version of the argument claims that whilst it is true that, say, a branch of scientific inquiry individuates olfaction differently from folk psychology, the philosophical question remains as to how the senses really type, how we really ought to individuate. Perhaps SP at best highlights the need for toleration of different taxonomies depending on our scientific inquiry, but it does not answer the further philosophical question of how the senses really are, how they are metaphysically, or perhaps how the senses individuate along “nature’s joints” (for discussion on pluralism and natural kinds see Dupré, 1995).
My response to this objection is similar to the first: that it pushes an intuition but offers no real argument. That there is any essential metaphysical taxonomy waiting to be individuated is just the premise SP denies. This rejection is not an argument for SP, but neither is affirming its appeal an argument against it. To claim that SP must be wrong because it does not respect certain intuitions concerning a definitive taxonomy is to beg the question in favour of monism. The opponent must do more to persuade the pluralist that the absence of any definitive criteria would be metaphysically unpalatable or otherwise incoherent. Furthermore, if we accept that SP does, in principle, account for the trends in perception science, then the burden of proof shifts to the opponent of SP to provide a more convincing alternative.
Scientific investigation has drawn a picture of the senses as a complex milieu of interacting processes. This shift suggests the abandonment of any sparse individuation of the senses. I have argued that SP, though perhaps not the only viable option, offers one plausible alternative. The two objections examined, both of which claim that SP does not offer a plausible theory of individuating the senses, fail to provide sufficient argumentation. SP respects the pragmatic differences present in everyday and scientific practices without asking us to give up talk of the sense modalities. In this way SP offers a satisfactory solution to the problem of individuating the senses, namely by dissolving it.
Alais, D., & Burr, D. (2004). The ventriloquist effect results from near-optimal bimodal integration. Current biology, 14(3), 257-262.
Dupré, J. (1995). The disorder of things: Metaphysical foundations of the disunity of science. Harvard University Press.
Fulkerson, M. (2014). Rethinking the senses and their interactions: the case for sensory pluralism. Frontiers in psychology, 5. Retrieved 1st April 2014, from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4261717/
Heilmann, S., & Hummel, T. (2004). A new method for comparing orthonasal and retronasal olfaction. Behavioral neuroscience, 118(2), 412.
Howes, D. (Ed.). (1991). The varieties of sensory experience: A sourcebook in the anthropology of the senses. Toronto: University of Toronto Press.
Landis, B. N., Frasnelli, J., Reden, J., Lacroix, J. S., & Hummel, T. (2005). Differences between orthonasal and retronasal olfactory functions in patients with loss of the sense of smell. Archives of Otolaryngology–Head & Neck Surgery, 131(11), 977-981.
Lillard, A. (1998). Ethnopsychologies: cultural variations in theories of mind.Psychological bulletin, 123(1), 3.
Macpherson, F. (2011). Taxonomising the senses. Philosophical Studies, 153(1), 123-142.
Nudds, M. (2011). The senses as psychological kinds. The senses: Classic and contemporary philosophical perspectives, 311-340.
Nudds, M. (2004). The Significance of the Senses. Proceedings of the Aristotelian Society (Hardback) (Vol. 104, No. 1, pp. 31-51). Blackwell Publishing Ltd.
Rospars, J. (1998). Dendritic integration in olfactory sensory neurons: a steady-state analysis of how the neuron structure and neuron environment influence the coding of odor intensity. J Comput Neurosci 5 (3): 243–266.
Small, D. M., Gerber, J. C., Mak, Y. E., & Hummel, T. (2005). Differential neural responses evoked by orthonasal versus retronasal odorant perception in humans. Neuron, 47(4), 593-605.
Ward, J. (2009). The frog who croaked blue: Synesthesia and the mixing of the senses. New York: Routledge.