benjamin d. young - smelling matter.pdf
DESCRIPTION
Smelling Matterby Benjamin D. YoungAbstract: It is a tacit assumption of contemporary philosophy of perception that any evidence or conclusions drawn from vision will generalize across all perceptual modalities. While the objects of olfaction – smells – are intuitively individuated by reference to the ordinary objects from which they arise, this view is incorrect. This paper argues that smells are neither ordinary three dimensional objects, identified according to the criteria employed to individuate visual or auditory objects, nor vapors (as Plato proposed), nor odors. Rather, smells are the chemical structures of molecular compounds within odor plumes.TRANSCRIPT
1
Smelling Matter Benjamin D. Young∗
Abstract: It is a tacit assumption of contemporary philosophy of perception that any evidence or conclusions drawn from vision will generalize across all perceptual modalities. While the objects of olfaction – smells – are intuitively individuated by reference to the ordinary objects from which they arise, this view is incorrect. This paper argues that smells are neither ordinary three-dimensional objects, identified according to the criteria employed to individuate visual or auditory objects, nor vapors (as Plato proposed), nor odors. Rather, smells are the chemical structures of molecular compounds within odor plumes. "Smell and its objects are much less easy to determine then what we have hitherto discussed; the distinguishing characteristic of smell is less obvious than those of sound or color."
- Aristotle DA, II 9 1. Introduction
What does your current location smell like? Close your eyes for a second. Can
you identify the things around you by smell? While we do not usually notice the odors
enveloping us, we surely do know what our environment looks like without being asked.
There is no doubt that we are primarily visual beings who sense and navigate the world
using vision, yet our behavior is also mediated by olfactory experiences.1 We are
constantly bombarded with smells, which continually shape our perception and
interactions with our surroundings.2 Intuitively, we smell ordinary three-dimensional
objects: the wet dog entering the elevator, the cookies from the bakery down the block,
or the musty rotting leaves outside the window on a warm autumn night. While smells
are commonly and intuitively individuated by reference to the ordinary objects from
which they arise, I will argue that this intuition is incorrect; the objects of olfaction (i.e.
what one smells) are the chemical structures of molecular compounds, which determine
the qualitative character of olfaction (i.e. how things smell).
While Plato, Aristotle and their Medieval Commentators engaged in a debate
regarding the nature and mechanisms of olfaction and how they differ from those of the ∗ I would like to thank Jesse Prinz and Richard Sorabji for their helpful feedback. 1 While human olfactory accuracy is excellent and on par with dogs, we systematically mistrust our olfactory abilities (Sela, and Sobel, 2010). 2 Olfaction guides our food choices (Fallon and Rozin, 1983), dietary preferences (Rozin, 1978; Rozin, et al. 1986; E. Rozin, S. Rozin, and E. Rozin, 1992), our selection of mates (Wilson and Stevenson, 2006) and social acquaintances (Li, et al., 2007) , and is responsible for identification of kin (Russell, 1976; Porter, Cernoch, and McLaughlin, 1983; Porter, et al. 1986).
Smelling Matter 2
other perceptual modalities,3 most of the recent work in the philosophy of mind and
perception takes vision as the primary modality of interest and tacitly assumes that any
evidence or conclusions drawn from its study can be generalized to all the perceptual
modalities. It is time to rectify contemporary philosophyʼs neglect of olfaction by re-
examining what the objects of our olfactory experiences are without simply assuming
that our best theory of vision will generalize to olfaction.
The focus of this article is the nature of smells – the objects that our olfactory
systems perceive in the world. I argue that the objects of olfactory perception are not,
as our common sense conception supposes, ordinary three-dimensional objects: What
we smell are neither ordinary three-dimensional objects, Platonic vapors, nor odors, but
rather the chemical structures of simple molecules (or mixtures of such) within odor
plumes. We literally smell the chemical structure of matter.
2. Olfactory objects are not ordinary Objects
Pre-theoretically our conception of an object is based on that of ordinary three-
dimensional objects. Apples, chairs, garbage trucks, and trees are all such objects; they
have clear boundaries as perceived by sight or touch. We certainly claim to have
olfactory experiences of ordinary objects. Commonly, we individuate smells by
reference to the objects from which they emanate. Our ordinary language usage most
likely derives from tracking olfactory experiences that are of ecological import,4 thus
making it natural to individuate the objects of olfactory experience by reference to the
ordinary objects from which we correlate their emanations. Intuitively we assume that
olfactory objects are ordinary objects, yet the spatiotemporal boundaries and chemical
composition of ordinary objects make them ill-suited to be the perceptual objects of
olfactory experiences.
3 Platoʼs theory of smell is contained within two paragraphs of the Timaeus 66d-67, while Aristotleʼs theory is developed in De Anima and De Sensu. Johansen (2006) provides a detailed assessment of the differences between Aristotleʼs theory in De Anima and De Sensu. Additionally, for an introduction to the Commentators debate over Plato and Aristotleʼs theories of olfaction see Kemp (1997). 4 Although ecological theories of the olfactory object are preferred by contemporary scientists, they are rejected for reasons discussed in Section 3.3.
Smelling Matter 3
The entities that are commonly thought of as having smells do not have the same
boundary conditions as ordinary objects. What we smell is spatially and temporally less
truncated then the object we identify by touch or sight. For example, the smell of
honeysuckle occupies a greater space, for a longer period of time, than the object (the
honeysuckle itself) that we identify tactilely and visually. Or consider the experience of
smelling the autumn leaves outside oneʼs window while sitting in a room. There is no
sense according to the ordinary object conception in which these entities are in the
room, yet the experience of smelling the leaves is certainly veridical and not a
hallucination (Batty, 2007).
Furthermore, it is an ill-advised theoretical starting point to assume that the
objects of olfactory experience is identical to those of visual or tactile experience, since
large portions of the ordinary object are unnecessary for generating an olfactory
experience (in contrast to what prima facie appears to be in the case of visual
experience).5 The average medium-sized object is composed of hundreds of different
molecules, yet only a dozen or so of these constitute its smell. Modern flavor and
fragrances houses are so profitable6 precisely because they are able to identify the
minimum number of chemical components of an ordinary object that are required to
produce an olfactory experience like that of smelling the object itself.7
A further problem for the ordinary object conception arises from cases of
olfactory misrepresentation (Lycan, 1996) where natural odors are mimicked using
synthetic chemical compounds. One can erroneously believe that one is smelling a rose
without there being a rose in the nearby environment, since there are synthetic chemical
compounds that mimic its odor; when smelling such a compound someone might quite
understandably come to both truly believe that it smells like a rose but falsely believe 5 It might be objected that this is unfair on the grounds that the visual experience of, say, a bowling ball does not require the presence of the ball in its entirety but merely a part of it. However, this might be taken to show that the ordinary object conception of visual experience is itself not tenable. Additionally, there is a longstanding debate within the philosophy of perception regarding whether we do in fact visually perceive aspects of an ordinary object that are not within our line of sight, such as the back of a bowling ball (Noë, 2005). 6 Flavor and Fragrance businesses generate multibillion-dollar revenues that on average exceed that of Hollywood and the entertainment industry. 7 Gilbert (2008) has a nice description of how the headspace model of International Flavors and Fragrances (IFF) is utilized in creating flavors and fragrances.
Smelling Matter 4
that the source of the smell is a rose. Such a case is neither one of perceptual
misrepresentation (e.g. it doesnʼt smell like a dog) nor one of hallucinatory experience
(there is something he is smelling). Rather, such a case involves a misrepresentation of
the actual source of the smell. Thus, ordinary objects should not be considered the
objects of our olfactory experience because we can have veridical perceptions of
olfactory qualities in the absence of the ordinary objects that produce them. If however
olfactory objects are not ordinary objects then do olfactory perceptions have objects at
all?
3. Criteria of Visual and Auditory Objects
To assuage any worry that olfactory experiences might not be object-directed,
since the object of olfactory perception is not an ordinary three-dimensional object, I
argue in this section that olfactory objects satisfy the criteria typically used to ascertain
the nature of visual and auditory objects of perception, such as having spatiotemporal
boundaries, a mereologically-complex nature across time and presentations, and
separability from against a background. Additionally, this section should allay any
concerns regarding the promiscuity of the notion of object at play within the paper by
demonstrating that the perceptual objects of olfactory experience satisfy the criteria
used to individuate visual and auditory objects, thus the objective status of the olfactory
objects is on par with that of visual and auditory objects.8 While this generates support
for the claim that our olfactory perceptual states are object-directed, I argue that none of
these criteria adequately explain the robust nature of olfactory experiences.
3.1 Spatial Temporal Entities Our common sense conception of objects is shaped by vision and is of a
spatiotemporal entity with clear boundaries. Visual objects are prime examples of
perceptual objects in this sense, since they are either “punctate spatiotemporal clusters”
(Scholl and Pylyshyn 1999, p. 26) or spatiotemporally-bound entities that maintain their
8 Furthermore, section 5.3 outlines the properties of the olfactory perceptual object, thereby solidifying the claim that we smell objective entities with mereological structure on par with the objects of visual and auditory perception.
Smelling Matter 5
features when being moved through an environment (Matthen, 2005). As such, our
olfactory experiences are not object-directed. The lack of strict spatiotemporal
boundaries of smells has lead some philosophers to argue that olfaction is an outlier or
counter-example to this conception of a perceptual object and that olfactory experience
does not have, as a proper part, the location or direction of the olfactory object (Lycan,
2000; Smith, 2002; Matthen, 2005; Batty 2007; Peacocke, 2008). But they are
mistaken; although the olfactory object is more dispersed in time and space than visual
or auditory objects, olfaction satisfies both the spatial and temporal requirement, but in a
less truncated form.
Empirically it has been demonstrated that slight differences in the anatomical
structure of each nostril (Yeshurun et al., 2008) and the resulting binaural airflow (Zhou
and Chen, 2009) enable the localization of smells to within 7-10 degrees of their location
(von Bekesy, 1964). There has also been some research showing that olfactory objects
can be tracked through an environment across time (Porter et al., 2005, 2007).9 Thus,
the majority of philosophers who have claimed that the location or direction of the
olfactory object is not a proper part of olfactory experience are simply mistaken. The
olfactory object has spatial boundaries, but they are diffused across an environment in a
manner quite unlike visual objects.
It would be imprudent to claim that the olfactory object is not a spatial entity
simply because its boundary conditions are not as truncated, especially given the
empirical work that shows that the olfactory object can be demarcated given the
concentration gradient of the olfactory objectʼs odor plume.10 However, the ability to
demarcate olfactory objects requires changing oneʼs spatiotemporal relations relative to
the olfactory object by either sniffing or changing oneʼs position. Matthen (2005) argues 9 Batty (2010a, 2010c) dismisses these experimental results on the grounds that they are atypical of olfactory experience, perhaps due to assuming the correctness of the common sense conception of the phenomenology of olfaction However, without further argument as to why it is more authoritative and how the ʻtypicalityʼ of an olfactory experience is determined Battyʼs theory is highly questionable. 10 A perfumeʼs sillage is an excellent example of the spatial aspect of an olfactory object. In designing a new product, perfume chemists (perfumers) must consider a perfumeʼs sillage - the diffusion rate across space of a perfume. Some scents are designed to announce their wearerʼs presence or to turn the heads of those in a room; colognes are usually designed to be noticeable only within a small radius around the wearer.
Smelling Matter 6
that olfactory experiences are thus not object-directed because although olfactory
experience can, across time (i.e. diachronically), be aided by merely moving around, it
does not at a particular time (i.e. synchronically) present us with a spatiotemporally-
bound entity. This is not the case for visual experience: opening oneʼs eyes, one is
immediately presented with three-dimensional objects in a spatial environment.11 The
olfactory object might have spatial boundaries, Matthen claims, but its temporal aspect
makes it unlike the spatiotemporal bound entities of vision.
Matthenʼs line of argument is predicated on the assumption that olfactory
experiences can be identified and individuated merely by using phenomenal experience
(i.e. how things present themselves to us) as a guide. Lycan (2000) clearly states that
the reason olfactory experiences should not be considered to have a spatial aspect that
the synchronic experience of smells does not seem to contain any spatial information as
part of the experiential content. Even though, over time, one can locate and attend to
the source of an odor, these theorists deny that olfactory experience is inherently
spatial. Using this line of reasoning, Matthen (2005, p. 284) denies that our experiences
of smell can be characterized in the usual object-attribute form and claims that at best
smells have a primitive undifferentiated feature-location structure. Smith (2002) goes so
far as to use this line of reasoning to question whether physical objects are smelt at all.
Following their lead, Batty (2009, 2010a-c) concludes that the olfactory object is best
thought of as a loosely defined, spatiotemporal object (something surrounding me has a
11 The assumption that vision automatically presents us, synchronically, with spatial objects might be challenged because similar temporal processes occur in vision. To see things, oneʼs eyes must be in constant motion either through volitional control or through saccadic and microsaccadic movements. If oneʼs eyes were to stop moving oneʼs visual field would shrink and eventually turn a uniform grey. (This can easily be demonstrated by holding oneʼs eyeballs still in their sockets or by using a ganzfeld.) This presents a prima facie analogy to the role of inhalation in olfaction. If inhalation is excluded as either a part of the olfactory experience or an enabling condition of the olfactory percept, then why not saccadic eye movement? Anticipating this reply, Smith (2002) argues that sniffing is not equivalent to saccadic eye movements. While the latter is not under volitional control) the former always requires it whenlocating and moving towards a smell. However, some forms of saccadic eye movement required for visual perception are under volitional control in an analogous way to sniffing. Thus, Smithʼs claim is most charitably interpreted as one about the awareness of movement such that one is not usually aware of saccadic eye movements, but always aware of sniffing when attempting to locate a smell.
Smelling Matter 7
smell) that cannot be described in subject-predicate form unlike the objects of the other
sensations.
Phenomenologically if synchronic visual and olfactory experiences are
considered together, it might be admitted that the latter has fewer spatial qualities.
However, at times smells are presented as appearing from or at a location.
Phenomenological evidence may however be quite misleading. While I agree that the
phenomenology of experience should be accounted for in generating a theory,
phenomenology should not be the sole determining factor. Rather than denying that
olfactory experience has an inherent spatial aspect, I claim that there is a better
explanation for the apparent lack of spatial phenomenology.
Plato and Aristotle both correctly recognized that inhalation is necessary for
olfaction, and it is not usually accompanied by a conscious awareness of modulating
breathing. Most of the time, odors seemingly present themselves as if they have just
appeared before the nose or occurred within our nostrils. However, while breathing can
be volitionally controlled this is usually not noticed and is only attended to when sniffing
once an interesting olfactory object catches the attention. Furthermore, recent studies
on the role of sniffing suggest that the act of sniffing itself is part of the olfactory percept
(Sobel, et al., 1999; Kareken et al., 2004; Kepecs et al., 2006; Mainland, et al. 2006;
Koritnik et al., 2008;).
Olfactory experiences, I claim, seem to lack a spatial phenomenology because
we do not commonly attend to our breathing patterns, which are both necessary for
olfactory experience and provide its spatial aspects. An awareness of the spatial
aspects of olfactory experience is given by diachronically attending to an olfactory
object, because attention is directed to breathing patterns and the movement of airflow.
Thus, although the spatiotemporal aspects of synchronic olfactory experience are not
commonly attended to, they are nevertheless present.
The objects of olfactory experience are spatiotemporally-bounded entities.
However, given the distinct spatial and temporal characteristics of smells, the criteria for
perceptual objects, as they are derived from vision, do not adequately capture the
Smelling Matter 8
nature of our olfactory experiences. More accommodating criteria are required, which
account for the variegated spatiotemporal nature of the objects of each modality.
3.2 Figure-Ground Separation The objects of auditory and olfactory perception have spatiotemporal boundaries
that exceed those of vision. Figure-ground separation can be used as a criterion, in
addition to spatial extent, to establish that olfactory objects are spatial. The olfactory
system is constantly bombarded by stimuli, yet we are able to separate and group
together stimuli that belong to one as opposed to another stimulus. The theory of
indispensable attributes (Kubovy, 1988; Kubovy and Van Valkenburg, 2001); – which
claims that essential attributes for each modality allow the perception of more than one
object at a time - is proposed to account for the ability for figure-ground separation and
as a general criterion for perceptual objecthood across modalities. For vision, the
essential attributes are claimed to be space and time, while for audition they are pitch
and time. While Kubovy and Van Valkenburg do not consider the case of olfaction, the
olfactory object satisfies this criterion of objecthood as well.
Chemicals, which could lead to olfactory experiences, constantly surround us.
Nonetheless different smells are separable from one another and detectable in the
environment. For instance, I can smell the roses within the bouquet of lilies and gerbera,
the rosemary on the roasted chicken, or the honeysuckle from the fresh-mown grass in
the meadow. Additional, empirical evidence for olfactory figure-ground separation may
be garnered from the overshadowing effect in odor mixture qualities. When combining
odorants in a mixture, if the constituents smell similar on their own it is often harder to
tease them apart when combined; conversely, if they smell dissimilar on their own it is
often quite easy to distinguish them within a mixture. However, in every variation of
similar and dissimilar pairings of odorants there is “evidence of overshadowing of one
component by another, depending upon the concentration level” (Kay et al., 2005, p.
727). Furthermore, if the concentration level of the overshadowed item is increased it is
possible to switch the overshadowing effect. Indeed, whether one smells an odor a
Smelling Matter 9
against a background of odor b (or vice-versa) can be manipulated merely by altering
the concentration levels of the components of a complex odor mixture
The olfactory object can meet the spatial criterion of a perceptual object using the
phenomenon of figure-ground separation. But the criterion still does not adequately
capture the nature of our olfactory experience. Odors can be recognized, discriminated
between, and identified across multiple contexts, presentations, and changes in their
properties. For example, the smell of a peach can be identified across changes in
intensity and concentration (when it is unripe, ripe, and overripe), and in different
contexts (as a fruity drink, in a baking pie, or in a perfume). Despite these changes, all
the various token olfactory experiences can still be recognized as all under a particular
type i.e. the peach smell.
3.3 Mereologically-Complex Entities
In an attempt to determine the nature of the auditory object, O'Callaghan (2008,
2009) suggests thinking about perceptual objects as entities with mereologically-
complex structures such that they can be recognized across contexts and changes in
properties. This approach to olfaction is seductive because it fits nicely with ecological
views of the olfactory object (Wilson and Stevenson, 2006; Gottfried, 2010). The
methodological assumption of ecological theories is that the olfactory object is to be
identified with the complex set of molecular compounds that enable us to track, locate,
and secure objects that are of value to us in maintaining our homeostatic needs. Such
theories define the olfactory object in terms of the function of olfaction in guiding the
ability to identify ecologically valuable information about the environment.
Wilson and Stevensonʼs (2006) theory is the most exhaustive scientific account
of the olfactory object that is in keeping with the criterion of a perceptual object as a
mereologically-structured entity. Central to this theory is the issue of how certain
chemicals can be recognized as bound together to form the perception of objects
despite the presence of countless other chemical structures in the environment. Given
their overall methodology, Wilson and Stevenson think that the perception of the
olfactory object partially depends upon ʻsyntheticʼ processing, such as memory
Smelling Matter 10
Figure 1. Polysantol – (4-Penten-2-ol, 3,3-dimethyl-5-(2,2,3-trimethyl-3-
cyclopenten-1-yl)) gives off a robust sandalwood smell.
processing that is modulated by the organismʼs current mental state, previous olfactory
states, expectations, and its situation. Thus, they identify the olfactory object with a
complex of molecular compounds, which is responsible for the ability to locate medium-
sized ordinary objects in the environment.
This ecological theory of the olfactory object - as a complex chemical mixture
composed of multiple molecular compounds that derive from an ordinary object - is not
satisfactory because simple molecular compounds can be recognized, identified, and
discriminated across contexts and against the background of other odors. Singular
synthetic molecules provide the simplest counter-example to Wilson and Stevensonʼs
ecological theory. For instance, Firmenichʼs Polysantol (See Figure 1) can be
recognized across presentations and varying levels of concentration.12 Wilson and
Stevensonʼs approach only accounts for complex mixtures. Simple odors composed of a
single molecular compound are experienced in the same manner as complex mixtures,
despite the fact that they are not mereologically-complex in the required ecological
sense.
Wilson and Stevensonʼs theory of the olfactory object is also not adequate
because it only generates a theory of how the perception of olfactory objects is
produced. While the theory provides an answer to how smells are recognized, identified,
and discriminated from one another, it provides no answer to the question of what
constitutes the olfactory quality of olfactory objects.
Olfactory experience is certainly object-directed, as it satisfies the
aforementioned spatiotemporal criteria used to ascertain the nature of the object of 12 One might debate whether the current formulation of Montaleʼs Aoud Pur Oriental contains this Polysantol or Firmenichʼs Oud Synthetic 10760 E (e.g. because they might be hard to discriminate between), but not, say whether it contains Givaudanʼs Black Agar Givco 215.
Smelling Matter 11
perception in vision and audition. The olfactory object has been shown to be a
spatiotemporal entity that can be separated from against a background of competing
stimuli, and maintain a mereological structure such that it can be recognized and
identified across changes in its location and properties. However, none of the criteria
discussed are sufficient to capture the nature of the olfactory object, which is less
spatiotemporally truncated than the objects of vision and audition, which at the same
time, is nevertheless mereologically-complex (even at the size of a simple molecular
compound).
4. Olfactory objects: Platonic vapors and odors
One suggestion that repeatedly appears in different guises in the literature is that
the olfactory object is the vapor, effluvia, or chemical gas given off by an ordinary object.
Platonic Vapor theories maintain that olfactory objects are detached proper parts of the
ordinary object itself (e.g. one literally smells the blood of an Englishman). Odor theories
claim that olfactory objects are detached subsets of the ordinary object, which are not
themselves indicative of it. According to odor theories, the objects of our olfactory
experience are chemicals given off by the ordinary object, which are diffused in odor
plumes. Although one does not perceive the entire ordinary object, nor the form of the
object, some aspects of the object are responsible for our experience of smell.
4.1. Platonic Vapors
Platoʼs theory of smell consists of little more than two paragraphs within the
Timaeus (66d-67) in which he claims that the vapors of objects are smelt as they
transform from one elemental shape to another, although they cannot be identified with
these elements themselves. Ordinary objects, asserts Plato, give off particles as a
substance condenses or becomes gaseous. Ordinary objects are perceived as they
transform between elemental kinds such that the smells of objects are the aerated forms
of the objects proper. According to this view, the olfactory object (what is smelled) is not
the elemental object itself, since our nostril cannot accommodate entities of such size,
but rather the (Platonic) form of the object.
Smelling Matter 12
Plato, interpreted as holding the view that olfactory objects are detached proper
parts (or the forms) of the entire ordinary source objects, does not explicitly endorse the
ordinary object view. However, it nonetheless suffers from some of the same problems
detailed above. Olfactory objects are not ordinary medium-sized objects because only a
small chemical subgroup of its material makeup and certainly not the form of the entire
object is required for olfactory experience. Moreover, synthetic chemicals can be
manufactured that mimic the smell of natural objects, suggesting that not even the form
of ordinary objects is necessary for creating the experience of the same olfactory quality
(its smell).
4.2. Odors
Contemporary odor theories are an outgrowth of the vapor view, although they
pay no homage to their theoretical ancestors. Odor theories maintain that the olfactory
object is an odor, which is identified with a gaseous emanation of an ordinary object. For
instance, a rose exudes diffused parts of itself, which is what we smell. Most
philosophers who discuss olfaction assume that smells are odors (Lycan, 1996, 2000;
Tye, 2000, 2002; Smith, 2002; Matthen, 2005; Batty, 2007, 2009, 2010a, 2010b).
Odors, thus identified, are certainly an interesting perceptual object, since their
spatiotemporal boundaries are less truncated than those of the other modalities. What
makes odor theories enticing is that they can account for olfactory misrepresentation,
since the olfactory quality of an odor does not depend upon the ordinary object itself.
Additionally, since only diffused subparts of the object (and not the entire ordinary object
itself) are required to elicit a smell, odor theories can prima facie provide the resources
to explain how olfactory experience can be veridical even when the ordinary object is
not present in the immediate surroundings.
The odor theory is partially correct but does not provide an explanation of what it
is about the odor that makes it smell the way it does. Since the experience of a smell
changes across the spatially diffused odor and can be spotty across presentations
within the odorʼs gradient, odor theory needs to provide an explanation of how the
Smelling Matter 13
olfactory quality of an odor is identified and individuated. What is responsible for an
odor smelling as it does? That is, what determines the olfactory quality of an odor?
The most obvious reply would be to individuate the qualitative character of odors
by reference to the natural source from which they emanate (Tye, 2000, 2002).
However this would be to return to the ordinary object view that was already rejected in
Section 2. Olfactory qualities cannot be individuated by reference to ordinary objects
because not all their aspects are required to generate the relevant smells. Moreover it is
possible to have veridical olfactory experience of an olfactory quality in the absence of
any ordinary objects (e.g. a synthetic rose smell still smells just as sweet). Even if it
were possible to account for these phenomena by appealing to the typical source of an
odor, the odor theory would still not be able to generate an account of what determines
the quality of an olfactory object. Why is it that a particular synthetic rose smell has
almost exactly the same olfactory quality as a Gardenia? A way of specifying the
olfactory quality of an odor independent of its ordinary object source is needed.
5. Olfactory objects as chemical (material) objects
If olfaction is a perceptual modality that is sensitive to chemicals then prima facie
olfactory objects should be specified in light of chemistry. Since chemistry studies
matter, i.e. anything with mass and volume, the criteria of a material object ought to be
an object that has a chemical structure such that it has mass and occupies space. For
the case of human beings olfactory objects, can be specified by noting the size of
molecules (not larger than twenty chemical groups and no smaller than three) that are
biologically detectable.13 The general requirements for a material object to be classed
as an odorant is that it should be volatile, hydrophobic and have a molecular weight less
than ~300 daltons (Ohloff, 1986).14
13 Specifying the olfactory object in this manner rules out ordinary objects. While they might satisfy the criterion of having a chemical structure, mass and volume, their chemical size clearly falls well beyond our bounds of perception. 14 In humans, the cut-off point for a chemical structureʼs size having a quality of smell is not gradual. This was demonstrated by Wrobel and Wannagat (1982), who showed, using benzenoid musk, that replacing one of itsʼ carbon atoms with a larger silicon atom caused the entire molecule to become odorless.
Smelling Matter 14
In this section, I argue that the objects of olfaction are the chemical structures of
molecular compounds within an odor cloud (i.e. a diffusion of molecules in space and
time). This conception of olfactory objects combines the virtues of the ecological and
odor theories with the added advantages that it can accommodate simple molecular
compounds having a smell and provide an explanation of olfactory qualities (how things
smell).
5.1 Olfactory objects are the chemical structures of molecules
Olfactory objects are neither molecules nor the elements that compose them. I
argue, rather that olfactory objects are three-dimensional chemical structures of
molecules (see Figure 3). The structure of a molecule is neither an abstract property nor
a specification of the molecular compound, but rather itʼs three-dimensional structure in
space and time. While it is natural to think of molecular structure as two-dimensional in
a similar fashion to Figure 2, it is better to think of it as it is represented in Figure 3.
Empirical evidence supporting the identification of olfactory objects with the molecular
structures of elemental compounds can be derived from the leading scientific theories of
primary olfactory transduction, our physiological olfactory capacities, psychological
olfactory abilities, and animal models of olfaction.15
15 Since the olfactory system is ontogenetically ancient, a great deal can be inferred about human olfactory experiences from animal models. Aside from the fact that it is a generally accepted practice throughout the sciences, Aristotle (DA II 7 419a33 – b1, DAII 9, De Sensu 5) also supports the methodology of thinking of olfaction as lying on a continuum with other
Figure 2: Two-dimensional representation of Bisphenol A (4,4'-
(propan-2-ylidene)diphenol)
Figure 3: Three-dimensional representation of Bisphenol A (4,4'-
(propan-2-ylidene)diphenol)
Smelling Matter 15
It is uncontroversial within the scientific community that the cause and quality of
olfaction is the structure of the molecules that come into contact with the olfactory
epithelium. However, there is less consensus regarding what it is about that the
structure of molecules that determines its quality (i.e. smell). The leading scientific
theories (which both agree in principle with the identification of olfactory objects with
molecular structures) are that the olfactory quality of molecules is determined by either
their shape or the vibrational patterns of the functional groups within the molecule itself.
Since the issue of primary sensory transduction is still under debate, there is as yet no
empirical reason to prefer either of the theories and none of the arguments offered will
depend on them.
5.2 Evidence for smelling the structure of matter 5.2.1 Physiological evidence
Evidence that the olfactory object is the molecular structure of matter can be
derived from studies on stimuli transduction by the olfactory system. While the Odotope
theory (Mori and Shepherd, 1994; Shepherd, 2005) and the biological spectroscope
theory (Turin, 1996, 2002, 2006; Turin and Yoshii, 2002) share the view that molecular
structures are olfactory objects and that a distributed activity pattern across multiple
olfactory receptor neurons (ORNs) determines the qualitative character of the olfactory
object, they disagree about the exact structural properties, and mechanisms of stimuli
transduction, responsible for it.
5.2.2 Psychological evidence for smelling matter
With the exception of perfume chemists16, most people would be surprised to
learn that they can detect, discriminate, and perhaps even identify the molecular
organisms. Animal models do not provide any further evidence for the claim that olfactory objects are the structures of molecular compounds, but only reinforce it (for example, both rodents and aphids can discriminate between enantiomers) For the sake of brevity, evidence from animal models in relation to human olfactory capacities is noted, where relevant, in the footnotes. 16 Burr (2007) gives a wonderful example of a perfume chemist, who can name the chemical structure of most of the common smells encountered when on a plane journey, from the lemon smell of the hot towels to the chemical odorants used in the bathroom disinfectants.
Smelling Matter 16
Figure 4: R-carvone and S-carvone
structure of the things they smell. However, the majority of people can discriminate
miniscule differences between and within the molecular structure of odorants. The
greatest source of evidence for this claim is the existence of enantiomers. Enantiomers
are molecular compounds whose structure and functional groups are identical, but
whose chiral properties (i.e. handedness) differ such that the molecules cannot be
symmetrically superimposed on top of each other. While most enantiomers have the
same smell (though both may not always be detectable by an individual), some do not.17
For example, R-carvone smells minty, while the S-carvone smells like caraway (Figure
4) (Boelens and van Gemert, 1993). Thus, the symmetry of the functional groups and
the orientation of a molecular compound are responsible for a different quality of smell.
Research on olfactory sensory conditioning also demonstrates, using classical
conditioning, that enantiomers can be discriminated despite subjects reporting that they
smell the same (Li et al., 2008). The results suggest that while optical isomers are
supraliminally indistinguishable, the two kinds of enantiomers can be discriminated
unconsciously, which can be interpreted as evidence that there are detectable
differences between the two types of smells.18
17 Since anosmia of a given enantiomer varies across populations, most flavor and fragrance products contain both enantiomers to ensure that the entire population is sensitive to the odor. The most exhaustive list of the smells of enantiomers is maintained by John Leffingwell. According to calculations of Leffingwellʼs listing in Turin (2006), 64% of enantiomers smell the same, 17% smell different and the remaining 19% are currently unknown. 18 Enantiomers have different behavioral effects across species. For instance Aphidius ervi is attracted to one enantiomer of nepetalactone (7S), but not the other (7R). When the two are combined the effect is an overall reduction in attraction (Glinwood et al., 1999). Further studies of enantiomers conducted on rats without any training, demonstrate that they are able to discriminate between R- and S- caravone, which as noted smells different to humans. With similar results to Li et al. (2008), Wilson and Stevenson (2006) have shown that, with some training, rats can learn to discriminate between limonene (R) and (S). These results further
Smelling Matter 17
In addition to being able to discriminate between the different versions of some
enantiomers we are also able to distinguish between aldehydes, compounds that differ
in one carbon group (Iamura et al., 1992). Our ability to discriminate between
aldehydes19 nicely demonstrates our sensitivity to changes in functional groups. The
case of aldehydes is intriguing because the smell of the compound changes as carbon
groups are added (Turin, 1996, 2006). The aldehydes from C-8 to C-12 all display an
interesting shift in smell as each carbon group is added: those with an even number of
carbon groups smell fruity, almost orange-like, while those with an odd number have a
floral, waxy odor (Arctander, 1994).
Further evidence that we are able to psychologically detect the structure of
molecular compounds is provided by the fact that we can detect the presence of
functional groups (Klopping, 1971). Functional groups are atomic groups within a
molecule that account for its chemical properties and structure. Interchanging functional
groups within molecules creates a predictable change in the qualitative character of
odorants (Turin and Yoshi, 2003). Thus, the identification of the olfactory object with the
structure of a molecular compound fits with these results, since the structure of a
moleculeʼs functional groups determines its characteristic smell in a predictable manner.
5.3 Olfactory properties
Having argued for the identification of the olfactory object with the chemical
structure of molecular compounds, what remains is specifying their properties. In
addition to the spatiotemporal properties of the olfactory object, which are best
determined empirically for each molecule or mixture according to the laws of chemistry,
I claim that olfactory objects have intensity, concentration, valence, and hedonic value.20
strengthen the claims that the object of olfaction is the molecular structure of chemical compounds and that the olfactory object is identical across species. 19 The reader may in fact have smelt Aldehydes, which are the primary synthetic ingredients of Chanel No. 5 and is what makes it so distinctive. 20 Experimental evidence that olfactory objects have these properties can be derived from the study by Li et al. (2008). While subjectsʼ were trained to discriminate between enantiomers that smell the same, their ratings of the valence, intensity, or familiarity of these structures were not affected. Since differences in the enantiomerʼs structure did not lead to a change in the
Smelling Matter 18
Intensity is simply the strength of the quality of smell. Concentration is best
thought of as the density of molecules of a particular odorant within its spatiotemporal
boundary.21 Thus, one might have a smell that is not intense, but highly concentrated.
For example, the intensity of Coumarin22 is quite weak, but walk through the menʼs
fragrance department of Macyʼs and the concentration level of these particles may
become unbearable.
The nauseating smell of the fragrance section of any department store also
provides an appropriate example of the property of valence23 (i.e. pleasantness) of an
odor. Pleasantness might better be thought of as a property of the olfactory object that
can vary across time and situations: steak always smells like steak, but if one has just
consumed thirty-two ounces of prime rib it might be the most repulsive thing one has
ever smelled.24 Finally, the hedonic value of olfactory objects should be distinguished
from, their valence. Some people may find the smell of skunks disgusting though the qualitative character of the olfactory object, it can be inferred that its properties remained the same. 21 Further evidence that concentration is a property of olfactory objects derives from the finding that as the level of concentration of an odorant increases in a given spatiotemporal location, a new olfactory percept is formed (McNamara et al., 2007), in the same way that changing the saturation of a given hue results in a different color. 22 Coumarin smells of golden, green, fresh, cut grass on a long, sunny, summer afternoon. 23 Yeshurun and Sobel (2010) have recently produced a rather exhaustive review of the literature on olfactory detection, discrimination, and identification. They claim that it is the molecular structure and pleasantness of an odor that determines the olfactory object, on the grounds that, based on the quality of smell, we are good at the first two and bad at the third, ,. Although this theory is certainly novel and interesting, their criteria of object-hood is unclear, especially given their claim that the olfactory object is a combination of external molecules together with a subjective judgment of pleasantness based on oneʼs homeostatic state. However, it might be possible to reject their theory on more substantial grounds by generating a more plausible explanation of the data cited in support of their theory. Yeshirun and Sobelʼs argument assumes that if we are bad at identification tasks, but good at labeling pleasantness, then the olfactory object must be determined by hedonic value. However, that we are poor at odor identification may better be explained by appeal to the linguistic deployment of concepts such that we are bad at naming and identifying olfactory objects. This makes the matter a performance issue rather than one involving the nature of the olfactory object itself. It is possible that the content of olfactory experiences cannot be fully captured using linguistic concepts alone, if for example the representational format of olfactory experience occurs in a combinatorial, but not classically-compositional, system. 24 There have been some cross-cultural studies demonstrating that judgments of valence are not innate, since they vary across cultures. For example, the smell of fish in certain island nations (e.g. Japan) is judged as neutral, while in landlocked nations (e.g. Bhutan) they are considered most unpleasant (see Gilbert, 2008, for a review of these studies).
Smelling Matter 19
experience of so doing nevertheless quite pleasant.25 While the properties of valence
and hedonic value are certainly a matter of learning and enculturation, intensity and
concentration seem to be neurophysiologically-driven.
6.1 Objection 1: But arenʼt smells subjective?
It is often claimed that smells and how they smell are subjective. An extremely
charitable interpretation of this claim is that the hedonic value, or valence of an olfactory
object is subjective. However, attributing this level of sophistication to the average
person, regarding olfactory properties, is implausible. Moreover the claim of subjectivity
is typically made about the very objects of olfactory experience rather than how they
smell (its quality). Indeed, it is more probable that our intuition about the subjectivity of
smells can be traced back to sensational theories of the olfactory object.
Sensational views of olfaction (Locke, 1690; Reid, 1764; Peacocke, 1983, 2003)
claim that smells are subjective objects that require a perceiver or mind to exist. Smells
are merely sensations within the mind. According to this view, the smell of a rose does
not objectively inhere within the rose itself and the only time a rose smells like anything
is when it is being smelled. As argued in Section 5, the olfactory object (what we smell)
is the chemical structure of molecular compounds, which determines its quality (what it
smells like). Since chemical structures are certainly mind-independent, external objects,
whose qualities are accessible in an objective manner across subjects, sensational
views are simply wrong.26
The olfactory object is not a subjective entity. The quality of an olfactory object
can be identified and individuated using its chemical structure. What something smells
25 While an extremely intense odor might cause a shift in valence, Wilson and Stevenson (2006) are careful to point out that intensity is a property separable from hedonic value. 26 The same reply can be used against projectionist views of olfaction. Perkins (1983) argues for an indirect-realist, projectionist account of smell such that the smell of an object does not inhere within the object itself and is not due to the causal powers of the object to cause in us a certain sensation. Rather, he claims, when smelling something, there is a disposition to judge that the ordinary object itself is attributed with the relevant olfactory property. So we are only indirectly (i.e. not directly) aware of the chemical structure that is responsible for our sensation of smell. The object of smell is an internal sensation, which is later projected upon the object that caused it. However, as argued, olfactory properties inhere in, and are intrinsic to, the chemical structure of odors.
Smelling Matter 20
like is not in the mind of the beholder, rather we can individuate the subjective qualities
of an olfactory experience in an objective and publicly assessable manner using
chemical structures. Metaphysical and epistemological issues about the kinds of
material entities that are to be considered olfactory objects need to be kept separate.
Metaphysically, everything composed of matter that has a molecular structure has a
smell. How we know about the smell of matter is an epistemic issue best left open for
empirical discovery. Relative to human beings, the material objects that are olfactory
objects can be specified by finding the size of molecules that are biologically detectable,
and discerning the conditions necessary for these chemical structures to be perceived
(See Section 5). However, our biological limitations should not lead us to exclude the
chemical structures outside of this range from having a smell, since there might be
beings with different biological systems who can smell what we cannot. (Similarly, it
should not be assumed that the limits of the color spectrum begin and end with our own;
there might be some species who perceive colors in the ultraviolet or infrared
electromagnetic frequencies.)
6.2. Objection 2: If we smell the chemical structure of matter, do we see photons too?
It might be objected that if the theory of olfactory objects presented here is
extended to vision, it creates the absurd conclusion that what we see are photons. The
visual system is sensitive to the wavelengths of photons at the first stage of stimuli
transduction within the retina, since these are the environmental stimuli received by the
rods and cones of the eye. Thus, the stimuli received by the visual system are just
photons of given energy states. However, it should not be assumed, without argument,
that the proffered theory of olfactory objects can be generalized as a theory of the
objects of the other perceptual modalities. The objects of visual experience are best
treated in the way outlined in Section 3.1, that is, in terms of spatiotemporal entities that
have clear boundaries.
Although the visual system is stimulated by photons, they are not what it is
primarily sensitive to. The primitives of the visual system are not infinitesimal dots, but
Smelling Matter 21
lines and edges, whose encoding begins to occur at the ganglion level of
photoreceptors (one step up from the rods and cones). We see three-dimensional
objects when we open our eyes. We see structured, spatiotemporal objects with clear
boundaries, such as coffee mugs, tables, and this article in front of you. The visual
system is wonderfully organized to be sensitive to the lines, edges, and their
orientations, that compose an ordinary object. The visual system is sensitive to photons
with differing energy levels in a way that is dependent upon the surfaces they have been
reflected by. However, these entities are neither detectable nor can they be
discriminated at either the physiological or psychological level. What is visually
perceived are the spatiotemporal properties of bound, ordinary objects. Moreover, none
of the arguments adduced for the view that the olfactory object is not an ordinary object
apply to the objects of visual experience.
The olfactory system is sensitive to the chemical structures of molecular
compounds and these are what are perceived as smells. In olfaction, the objects that
olfactory receptors are sensitive to are the very same entities that can be physiologically
and psychologically detected and discriminated. The olfactory quality of olfactory objects
is constituted by the three-dimensional chemical structures of the molecular
compounds, as derived from the composition and placements of its functional groups.
This is a very strong reason against the extension of the theory of olfactory objects to
the case of vision.
6.3. Objection 3: We donʼt report smelling chemical structures. Our olfactory experiences seem to present themselves to us as subjective
sensations within our own consciousness, rather than - as argued here - experiences of
external, structured molecular compounds. While having an olfactory experience, most
people do not report being aware of the posits of chemistry. Although flavor and
fragrance chemists have olfactory experiences of the chemical structures of molecular
compounds, for the rest of us, chemistry is something to be endured in high school. The
simplest reply to this objection is that since smells are the objects of olfaction, and
smells are the chemical structures of molecular compounds, the mode of presentation of
Smelling Matter 22
our experience of smells need not be as of such entities having this or that molecular
structures.27
The objection that this is not how the awareness of smells presents itself tacitly
assumes that the olfactory object must be specified by appeal to the awareness of
experiential content. Such awareness is typically claimed to occur against the
background of deployed concepts. To be aware of a smell (i.e. the chemical structure of
a molecular compound) as the very compound it is would require the use of a vast
repertoire of concepts from chemistry, which most of us do not possess or deploy on a
regular basis. This lack of conceptual sophistication does not prevent us from having the
same type of olfactory experience. After all, someone might understandably claim to
detect a sweet, flowery putrid undertone in Chanel No. 5, yet at the same deny that their
experience is of a large concentration of Aldehydes. While our awareness of the
olfactory experience does not commonly present us with the posits of chemistry, this
does negate that our olfactory experiences are of the three-dimensional structure of
molecular compounds, rather it is more indicative of our conceptual repertoire. Perhaps
this partially explains the temptation of individuating smells using ordinary objects or
their effluvia (odor theories) as we commonly use these entities to talk about smells, but
this way of talking is a short hand that we utilize as an imperfect crutch to compensate
for our feeble chemical vocabulary.
7. Conclusion
Our intuitions and everyday linguistic practices often serve as a theoretical
starting point in philosophy, yet they are not always accurate guides. We certainly do
talk about smells and claim that our olfactory experiences are object-directed. In
everyday life, smells are individuated using our intuitive grasp of ordinary objects,
thereby making the emanations of ordinary three-dimensional objects, picked out
visually, our most natural conception of the olfactory object. However, our pre-
theoretical intuitions and linguistic practices are themselves theory-laden. In this article I
27 If the claim that experiences of olfactory objects have intentional content is accepted then it is generally agreed that we cannot quantify into intentional attributions, such as reports about beliefs or perceptual experience (See Frege, 1892/1997; Quine, 1953, 1956; Davidson, 1969).
Smelling Matter 23
have argued that our folk conception of smells is incorrect, since it tacitly assumes an
inappropriate conception of the olfactory object that is borrowed from the study of vision.
While we are doubtlessly primarily visual creatures, the workings of the visual system,
and what we know about it, should not be the sole arbiter on what the proper objects of
perception are.
The best theory of visual objects does not adequately capture the nature of
olfactory experience. The perceptual object of olfaction is not an ordinary object and the
criteria typically used to ascertain and individuate visual and auditory objects do not fully
account for the apparent properties of olfactory objects. Our olfactory experiences are
of objects that are both spatiotemporally less truncated than those of vision and
audition, and - at the same time - are caused by such minuscule entities as simple
synthetic molecular compounds. These aspects of the olfactory object generate
sufficient reason enough to reject ecological theories of olfaction, yet at the same time
provide evidence that olfactory experiences are of external objects in our environment.
Contemporary Odor theories are also inadequate because they either cannot account
for the qualitative nature of olfactory objects (what objects smell like) or are empirically
untenable.
We smell the chemical structure of molecular compounds within odor plumes.
This theory of the olfactory object was supported by appeal to empirical evidence
regarding our physiological and psychological abilities to detect, distinguish, and
recognize the chemical structure of molecular compounds. So conceived, an
explanation of what determines the olfactory qualities of such objects was provided,
namely that the smell of each olfactory object is constituted by the structure of the
chemical compound as determined by its functional groups. We smell types of matter
and not subjective entities; the way such things smell are objective facts about the
nature of the material fabric of reality and not about some (private) subjective realm of
experience. Finally, olfactory objects are neither similar to those found in vision nor what
would be pre-theoretically expected (based on our linguistic practices and our first-
person phenomenological reports of olfactory experience). We smell the chemical
structure of matter and it is this that smells.
Smelling Matter 24
References:
Arctander, S., 1994. Perfume and Flavor Chemicals. Carol Stream, IL: Allured Publishing.
Aristotle, 1984a. De Anima. In The Complete Works of Aristotle, vol. 1, edited by Jonathan Barnes. Princeton, NJ: Princeton University Press, 641-692.
Aristotle. 1984b. De Sensu. In The Complete Works of Aristotle, vol. 1, edited by Jonathan Barnes. Princeton, NJ: Princeton University Press, 693-713
Batty, C. E., 2007. Lessions in Smelling: Essays on Olfactory Perception, Linguistics and Philosophy, the MIT Press, Cambridge, MA.
———. 2009. ʻWhat's that Smell?ʼ, Southern Journal of Philosophy 27, 321-348. ———. 2010a. ʻWhat the Nose Doesn't Knowʼ. Journal of Consciousness Studies 17, 10-17. ———. 2010b. ʻScents and Sensibiliaʼ. American Philosophical Quarterly 47, 103-118. ———. 2010c. ʻA Representational Account of Olfactory Experienceʼ. Canadian Journal of
Philosophy Boelens, M. H., and van Gemert ,L. J.ʼ 1993. ʻSensory properties of optical isomersʼ. Perfumer
and Flavorist 18: Nov-Dec. Burr, C., 2007. The Perfect Scent. Holt, Henry & Company, Inc. Fallon, A. E., Rozin, P., 1983. ʻThe psychological bases of food rejections by humansʼ. Ecol
Food Nutr 13: 15-26. Gibson, J. J., 1966. The Senses Considered as Perceptual Systems. New York: Houghton
Mifflin Company. Gilbert, A. N., 2008. What the Nose Knows: The Science of Scent in Everyday Life. New York,
NY, Crown Publishers. Glinwood, R.T., Du, Y.J., Powell, W., 1999. ʻResponses to aphid sex pheromones by the pea
aphid parasitoids Aphidius ervia and Aphidius eadyiʼ. Entomol Exp Appl 92: 227-232. Gottfried, J. A., 2010. ‘Central mechanisms of odour object perception’. Nature Reviews
Neuroscience 11, 628–641. Iamura, K., Mataga, N. and Mori, K., 1992. ʻCoding of odor molecules by mitral/tufted cells in
rabbit olfactory bulb. I. Aliphatic compoundsʼ. J. Neurophysiol., 68, 1986–2002. Johansen, T. K., 1996. ʻAristotle on the Sense of Smellʼ. Phronesis XLI/I: 1-20. ———. 2006. ʻWhat's New in the De Sensu? The place of De Sensu in Aristotleʼs Psychologyʼ.
In Common to body and soul: Philosophical approaches to explaining living behaviour in antiquity (edited by R.King). Berlin: Walter De Gruyter.
Kareken, D. A., Sabri, M., Radnovich, A. J., Claus, E., Foresman, B., Hector, D. and Hutchins, G. D., 2004. ʻOlfactory System activation from sniffing: effects in piriform and orbitofrontal cortexʼ. NeuroImage 22: 456-465.
Kepecs, A., Uchida, N. and Mainen, Z. F., 2006. ʻThe Sniff as a Unit of Olfactory Processingʼ. Chemical Senses 31: 167-179.
Kemp, S., 1997. ʻA Medieval Controversy about Odorʼ. Journal of the History of the Behavioral Sciences 33 (3): 211-219.
Klopping H.L., 1971. ʻOlfactory theories and the odors of small moleculesʼ. J Agric Food Chem 19: 999-1004
Koritnik, B., Azam, S., Andrew, C. M., Leigh, P. N. and Williams, S. C.R., 2008. ʻImaging the brain during sniffing: A pilot fMRI studyʼ. Pulmonary Pharmacology & Therapeutics: 1-5.
Kubovy, M., 1988. ʻShould we resist the seductiveness of the
Smelling Matter 25
space:time::vision:auditionanalogy?ʼ. Journal of Experimental Psychology: Human Perception and Performance,14(2): 318-20.
Kubovy, M. and Valkenburg, D. V., 2001. ʻAuditory and visual objectsʼ. Cognition 80: 97- 126.
Li, W., Moallem, I.;,Paller, K. A. and Gottfried, J. A., 2007. ʻSubliminal Smells Can Guide Social Preferencesʼ. Psychological Science 18 (12): 1044-1050.
Li, W., Howard, J.D., Parrish, T.B. and Gottfried, J. A., 2008. ʻAversive Learning Enhances Perceptual and Cortical Discrimination of Indiscriminable Odor Cuesʼ. Science 319: 1842-1845.
Locke, J., 1959/1690. An Essay Concerning Human Understanding. A. C. Fraser (ed.). New York, NY: Dover Publications.
Lycan, W. G., 1996. Consciousness and Experience. Cambridge MA: MIT Press. ———. 2000. ʻThe Slighting of Smellʼ. In Of minds and molecules; new philosophical
perspectives on chemistry. Oxford; New York: Oxford University Press. Mainland, J., and Sobel, N., 2006. ʻThe Sniff is Part of the Olfactory Perceptʼ. Chemical Senses
31: 181-196. Matthen, M., 2007. Seeing, Doing, and Knowing. Oxford: Oxford University Press. McNamara, A. M., Magidson, P.D., Linster, C., 2007. ʻBinary Mixture Perception is affected by
Concentration of Odor Componentsʼ. Behavioral Neuroscience 121 (5): 1132-1136. Mori, K., and Shepherd, G. M., 1994. ʻEmerging principles of molecular signal processing by
mitral/tufted cells in the olfactory bulbʼ. Cell Biology 5: 65-74. Noë, A., 2005. Action in Perception. Cambridge MA: MIT Press. O'Callaghan, C., 2007. Sounds. Oxford: Oxford University Press. ———. 2008. ʻObject Perception: Vision and Auditionʼ. Blackwell Philosophy Compass: 1-28. Ohloff, G., 1986. ʻChemistry of odor stimuliʼ. Experientia 42: 271-279. Peacocke, C., 1983. Sense and Content. Oxford: Claredon Press. ———. 1986. ʻAnalogue contentʼ. Proceedings of the Aristotelian Society 60,1-17. ———. 1989. ʻPerceptual contentʼ. In J. Almog, J. Perry, and H. Wettstein (Eds.),
Themes from Kaplan. New York: Oxford University Press. ———. 2008. ʻSensational properties: Theses to accept and theses to rejectʼ. Revue
Internationale de Philosophie 62: 7-24. Perkins, M., 1983. Sensing the World. Indianapolis, Indiana: Hackett Publishing. Plato, 1997. Complete Works. J.M. Cooper and D.S. Hutchinson (eds.). Indianapolis, and
Cambridge: Hackett Publishing. Porter, J., Anand, T., Johnson, B.N., Khan, R.M. and Sobel, N., 2005. ʻBrain Mechanisms for
Extracting Spatial Information from Smellʼ. Neuron 47: 581-592. Porter, J., Craven, B., Khan, R.M., Chang, S.J., Kang, I., Judkewitz, B., Volpe, J.,
Settles, G., and Sobel, N., 2007. ʻMechanisms of scent-tracking in humansʼ. Nature Neuroscience 10 (1): 27-29.
Porter, R.H., Balog, R.D., Cernoch, J.M. and Franchi, C., 1986. ʻRecognition of kin through characteristic body odorsʼ. Chem Senses 11: 389-395.
Porter, R.H., Cernoch, J.M. and McLaughlin, F.J., 1983. ʻMaternal recognition of neonates through olfactory cuesʼ. Physiol Behav 30: 151-154.
Porter R.H. and Winberg J., 1999. ʻUnique salience of maternal breast odors for newborn infantsʼ. Neuroscience and Biobehavioral Reviews , 23, 439–449
Smelling Matter 26
Reid, T., 1997/1764. An Inquiry into the Human Mind. D. R. Brookes, (ed.). Edinburgh, UK: Edinburgh University Press.
Rozin, E., Rozin, S. and Rozin, E., 1992. ʻEthnic Cuisineʼ. In Flavor principle cook book. E. Rozin (ed.). East Rutherford, NJ: Penguin Group.
Rozin, P., 1978. ʻThe use of characteristic flavourings in human culinary practiseʼ. In Flavour: Its chemical behavioural and commercial aspects, 101-127. Boulder, CO: Westview Press.
Rozin, P., Hammer, L., Oster, H. and Marmora, V., 1986. ʻThe childʼs conception of foodʼ. Appetite 7: 141-151.
Russell, M. J., 1976. ʻHuman olfactory communicationʼ. Nature 260: 520-522. Scholl, B. and Pylyshyn, Z., 1999. ʻTracking multiple items through occlusion: Clues to
visual objecthoodʼ. Cognitive Psychology 38: 259-90. Sela, L. and Sobel, N., 2010. ʻHuman Olfaction: a constant state of change-blindnessʼ. Exp
Brain Res:1-17. Shepherd, G. M., 2005. ʻOutline of a Theory of Olfactory Processing and its Relevance to
Humansʼ. Chemical Senses 39 (S1): i3-i5. Smith, A. D., 2002. The Problem of Perception. Cambridge, MA.: Harvard University Press. Sobel, N., Khan, R. M., Saltman, A., Sullivan, E.V. and Gabrieli, J. D. 1999. ʻThe world smells
different to each nostrilʼ. Nature 402:35. Turin, L., 1996. ʻA Spectroscopic Mechanism for Primary Olfactory Receptionʼ. Chemical
Senses 21(6): 773-791. ———. 2002. ʻA Method for the calculation of odor character from molecular structureʼ. Journal
of theoreticla biology 216: 367-385. ———. 2006. The secret of scent: adventures in perfume and the science of smell. New York,
NY: HarperCollins. Turin, L. and Yoshii, F., 2002. ʻStructure–odor relations: a modern perspectiveʼ. In: Handbook
of Olfaction and Gustation (edited by R. Doty). New York, NY: Marcel Dekker. Tye, M., 2000, Consciousness, Color, and Content. Cambridge, MA: The MIT Press. ———. 2002. ʻRepresentationalism and the Transparency of Experienceʼ. Nous 36(1): 137-
151. von Bekesy, G., 1964. ʻOlfactory analogue to directional hearingʼ. Journal of Applied
Physiology 19(3): 369-373. Wilson, D. A. and Stevenson, R. J., 2006. Learning to Smell. Baltimore, MD: The
Johns Hopkins University Press. Wrobel, D. and Wannagat, U., 1982. ʻSila substituted perfumes 4 Sila derivatives of some
musk scentsʼ. Journal of Organomertallic Chemistry 225: 203-210. Yeshurun, Y., Dudi, Y., and Sobel, N., 2008. ʻWorking Memory Across Nostrilsʼ. Behavioral
Neuroscience 122(5): 1031-1037. Zelano, C. Montag, J., Johnson, B., Khan, R. and Sobel, N., 2007. ʻDissociated
Representations of Irritation and Valence in Human Primary Olfactory Cortexʼ. Journal of Neurophysiology 97: 1969-1976.
Zhou, W. and Chen, D., 2009. ʻBinaural Rivalry between the Nostrils and in the Cortexʼ. Biology 19(18): 1561-1565.