# Hearing Objects and Events **Abstract:** move or interact so that they vibrate and produce sound waves, such as when they roll, collide, or scrape together. It is often claimed that we do not simply hear sounds and infer what event caused them, but hear source events themselves, through hearing sounds. Here I investigate how the idea that we hear source events should be understood, with a focus on how hearing an event relates to hearing the objects involved in that event. I argue that whereas we see events such as rollings and collisions by seeing objects move through space, this cannot be how we hear them, and go on to examine two other possible models. On the first, we hear events but not their participant objects. On the second, to hear an event is to hear the appearance of an object to change. I argue that neither is satisfactory and endorse a third option: to hear a source event is to hear an object as extending through time. ## 1 Hearing the sources of sounds Hallucinations notwithstanding, we have auditory experiences when sound waves hit our eardrums. Sound waves are usually produced when material objects[^1] move or interact so that they vibrate, such as when they are rolled, scraped, struck, or smashed. Such events are referred to as sound sources, or source events. We very often learn what sort of source event is taking place just from what we hear: if you close your eyes and I roll a bowling ball across the floor you will learn that a heavy, hard, round, object is rolling, if I drop it you will learn that that an object with those same properties has collided with something. A majority of contemporary philosophers who work on auditory perception think that our awareness of these events is due to our hearing these events. The aim of this paper is to determine how we should understand such a notion. What is it to hear a source event? The idea that we hear sources at all might strike some as odd. Do we not hear the sounds that these events produce, rather than the events themselves: the rumble of the rolling and the clunk of the collision? If sounds are all we hear then this would mean that our awareness of everyday events is indirect: we hear the rumble and infer the rolling in the same way that we might hear the rumble and infer that the bowling alley is open, or see smoke and infer fire. While this idea has some historical precedent, Berkeley claimed ‘‘in truth and strictness, nothing can be heard but sound’’ (1713/2012, p. 67), it is a very unpopular amongst contemporary philosophers interested in auditory perception.[^2] O’Callaghan says that, sources ‘‘meet causal requirements on auditory perception, and it would be severe to deny they belong among what it seems to us we hear’’ (2011, p. 1), Kulvicki is concerned that if sounds are understood as individuals distinct from material things, we risk putting ‘‘events of interest’’ (that is, source events) at ‘‘one remove’’ from listeners (2008, p. 15), while Leddington argues that a phenomenological intimacy between listener and source event is apparent in auditory experience: it does not seem as though we are hearing signals or signs of source events, but rather the events themselves (2014, pp. 328–329). Many accounts of auditory perception avoid the severity of Berkeley’s view by tethering the hearing of sources to the hearing of sounds. One way of doing this is to say that we hear sounds as properties, either of objects that participate in source events (e.g. Kulvicki 2008) or the source events themselves (Leddington 2014). On ## 2 Seeing and hearing events ## 2.1 What is heard? Although the claim that we hear source events is popular, I shall argue in this section that it is not obvious how it should be understood. This can be brought out through a comparison with vision: while we have a clear idea of what it is to see events, it is implausible to think that we hear events in the same way. If this is correct, we must characterise source hearing in some other way. We can begin by asking what is heard when we hear source events. It is common in contemporary philosophy of perception to think that our perceptual systems supply information about features of the immediate environment. As Matthen puts it, perception is ‘‘outward–oriented’’, in that it ‘‘conveys… a message about the external world’’ (Matthen, 2005, pp. 3–4).[^3] If this is correct, we can ask what types of feature a given perceptual modality is oriented towards. The mainstream view for vision, for example, is that we receive messages about material objects, the medium-sized dry goods which surround us (e.g. Cohen 2004; Nanay 2013, p. 51; ## 2.2 Perceiving movement and the spatial field When our eyes are open we are not presented simply with material objects and their properties, but rather a portion of space with some regions filled by objects, and some empty regions that could potentially be filled by objects, this feature of visual experience I will refer to (following for example, Martin 1992; Richardson 2014) as the spatial field. Taking in a region of space at a time adds to visual experience in various ways. First, it allows for the perception of spatial relations between different objects: you see the bowling ball as in front and to the right of the region of your feet. Second, it allows for the perception of property variation: you can see your bowling shoes gradually change from a bright green at the toe, to a darker shade at the heel.[^6] Third, it allows for the perception of the precise location of objects, in that objects can be seen to extend across, or fill out, certain regions of space: seeing the bowling ball to be located here rather that there is to see it to fill out this ball-shaped region of space rather than that one. It is this final aspect of spatial seeing which allow us to see movement. Watching an object move involves seeing it to extend over different portions of space at different moments: seeing the bowling ball roll down the lane involves seeing it to be first by your feet, then increasingly further away as it travels down the lane.[^7] Audition however, is not field-like in the same way that vision is. It does not present an array of filled and empty regions of space, nor of objects as extending through space. As O’Callaghan puts it: It is counterintuitive to say you can auditorily single out material objects because auditory experiences do not reveal three–dimensional solid bodies with rich internal spatial structure as such. (2011, p. 23) This is not to say that audition provides no spatial information. We often hear source events as located: footsteps behind us, a ball colliding with the ground to the left. We can also hear source events to be at different locations from moment to moment: the rolling is heard to be close by and then increasingly further away as the ball travels down the lane. However, hearing source events as located is not equivalent to hearing clearly bounded, three-dimensional objects. We do not hear the shape of the rolling ball, where its stops and the rest of the world begins. Consider a different case: you lie in bed while a mosquito flies around your head, and from what you hear you learn that it is first on the left and then the right. But this is still quite dissimilar to how we would see the creature move. First, as with the bowling ball case, our experience is not of a three dimensional, mosquito-shaped object extending over different portions of space. Second, the source event in this case, the movement which produces sound waves, is not the mosquito flying from the left to the right per se, but rather the beating of its wings, and we do not hear this event by hearing each wing to fill out different wing shaped regions of space at different moments. In both of these cases, it seems that while the approximate location of ## 3 Not hearing objects O’Callaghan suggests that the fact we do not hear material objects as extended in space is a reason to give up the idea that we hear material objects at all. He says: ‘‘Plausibly, perceiving a particular thing requires being able to differentiate, discriminate, or distinguish it from the surrounding environment’’ (2011, p. 8), and points to similar claims that have been made about vision, such as Siegel saying: ‘‘If S sees o, then S’s visual phenomenology differentiates o from its immediate surroundings’’ (2009, p. 434).[^10] He suggests instead that ‘‘Hearing is notable in that…you do not auditorily experience ordinary material objects as such, while you do auditorily experience their activities’’ (2011, p. 23) and goes on to suggest that an advantage of his own account of source perception is that it allows for ‘‘the evidential parity of vision and audition regarding, respectively, material things and their activities’’ (ibid., p. 26). A similar idea can be found in Casati et al.: The immediate object of auditory perception is just the event that is the sound…in virtue of the perception of it we may recognize the thing source. ‘In virtue of’ means, roughly, that one has indirect awareness of x when one’s awareness of x is caused by one’s awareness of something else (2013, p. 465; see also Leddington 2014, p. 330) ## 4 Hearing objects change ## 4.1 Changing appearances A different approach to source hearing can be found in Kulvicki (2008), who suggests that, like vision, we hear sources via hearing their objects. Unlike the ## 4.2 Change perception and the temporal window We can think of seeing change as having two components. First, seeing change involves seeing an individual to take on a sequence of properties, or, in the case of ## 5 Hearing perdurance ## 5.1 Endurance and perdurance The extent of the problems faced when we try to explicate source hearing are now clear. Objectless accounts are uninformative and not well-motivated, and source hearing cannot be a case of hearing the locations or properties of objects to change. Given such difficulties, we might ask whether it is possible to give a coherent ## 5.2 Temporal shapes and temporal windows Prosser’s reasons for thinking that objects are seen as enduring rather than perduring stem in part from a dubiousness as to how the successive temporal parts of an object can be perceived as unified, as ‘‘somehow’’ belonging to the same composite whole. He imagines that the phenomenology of perdurance perception would be something like the following: there would be representations of time-slices of objects at different positions, perhaps understood as related to one another as parts of the persisting whole (perhaps in a manner analogous to the way in which dots in a grid may be seen as grouped into objects consisting of rows or columns). But there would be no experience of dynamic motion or change (ibid. p. 25) It is unclear to me why Prosser thinks experiencing an object as perduring would consist in perceiving a series of non-dynamic slices, disconnected in the way that dots on a grid would be. On a perdurantist metaphysics, objects extend smoothly and continuously through time, and there is no reason to think our perceptions would not be of continuous objects also.[^21] We saw in Sect. 2 that one way in which the spatial field adds to visual experience is that it allows objects to be seen as extending across a portion of space. However, seeing an object as extended is not to see its spatial parts slice by slice, as it were, but rather as a unified whole. The temporal field, I suggest, allows for objects to be heard as temporally extended in the same way, as composite wholes, made up of unified temporal parts. ## 6 Conclusion Here I have argued that anyone wishing to defend the idea that we hear sound sources is required to give an account of what we hear and how we hear it, examined and criticised two possible explanations, and put forward what I hope to be a plausible and illuminating alternative. We do not hear sources by hearing objects to change their locations or properties, nor by hearing events without objects. Rather, to hear a source event is to hear an object extend through time and as having a certain temporal shape. This account departs to some degree from how temporally extended experience is typically characterised, as it rejects the idea that perception over time must be as of either changing or unchanging individuals. At the same time, if we consider the ideas on which this account is based, we find that it is not an especially radical break from what has gone before. As well as drawing on a well-established and popular position in metaphysics, it retains the idea that the world is perceived through a temporal window which structures what we perceive over short durations. Given that the temporal window is frequently compared to the spatial field, and that the spatial field allows for the perception of spatial extension, we should take seriously the possibility that the temporal window allows for the perception of its temporal counterpart. Acknowledgements I am grateful to have received extremely useful comments on the ideas presented in this paper from my colleagues at the Centre for Philosophy of Time at the University of Milan, members of Bence Nanay’s Action and Perception research group at the University of Antwerp, and my audience at the Joint Session of the Aristotelian Society & the Mind Association in July 2016. Particular thanks to Davide Bordini, Clotilde Calabi, Laura Gow, Dave Ingram, Bence Nanay, Matthew Nudds, Casey O’Callaghan, Maarten Steenhagen, and Giuliano Torrengo. Funding This study was funded by Università degli Studi di Milano (Project Number 15-6-30070002021) ## Footnotes [^1]: Sound waves can of course also be produced by the movement of liquids or gases. Although I will talk only about material objects here, everything I say applies mutatis mutandis in these cases as well. [^2]: In contemporary philosophy of perception, Matthen comes closest to a sounds only view of audition: ‘‘objects are characterizable by the sound they make—there are squeaky violins, baritone singers, alto saxophones—but the inferential path from auditory experience to such dispositional characterizations is quite indirect’’ (2005, p. 285). Martin says: ‘‘One may pick out the source of the sound via picking out the sound itself—we might then understand the demonstrative expression, ‘that dog’ as involving deferred ostension, perhaps as the descriptive phrase, ‘the dog which is actually the source of this sound’’’ (1997, p. 93) but has a less hardline view in his later work, saying that, in some cases at least, source events can be heard (2012, p. 348). this sort of approach, we, as Leddington puts it, hear sources in hearing sounds (ibid., p. 324). Another is to say that sounds are individuals distinct from sources but that we hear sources mediately, through or by hearing sounds. O’Callaghan, for example, proposes that we hear sounds as parts of events (2011), and Nudds that we hear sounds as having been produced by sources of one kind or another (2010, p. 285). My interest here is not in what sounds are, nor in the relation between sound hearing and source hearing, but in source perception itself. All the views just mentioned are united in saying that sources are heard, not inferred, and I want to explore what such a claim amounts to. Questions about source hearing can be decoupled, by and large, from questions about sound hearing: claiming that we hear a source as causing, as part of, or as bearing, a sound, does not explain what it is to hear a source, in the same way that seeing an object as a cause, as having other objects as parts, or as bearing a colour, does not explain what it is to see an object. You and I might agree on how sound hearing relates to source hearing, yet disagree about what sounds are, conversely, you and I might agree on what sounds are, and on how hearing them relates to hearing sources, yet disagree about how the second relatum should be characterised, about what hearing a source consists in. For these reasons, sounds will feature little in what follows, with the focus instead being on how source hearing should be understood in and of itself. [^3]: This is in contrast to the sense-data theories of perception popular in the first half of the twentieth Century. Matthen 2005, pp. 277–282).4,5 They are what is seen in the first instance, meaning that anything else we see, we see by seeing them and their properties. Disagreements as to what sounds are can be understood as a debate about whether the auditory system is oriented towards one feature of the environment or another: does hearing a sound consist in the auditory system picking out sound waves (Nudds 2010), disturbances of a medium (O’Callaghan 2007) or properties of objects (Kulvicki 2008)? If, as I have suggested, questions about source perception can be considered separately from questions about sound perception we can ask what aspects of the environment are picked out when we hear source events. One might think that in asking this question we have answered it: we are auditorily oriented towards events. Whereas the visual system deals in balls, doors, and vases, the auditory system tracks keeps track of rollings, slammings, and smashings. Things, however, cannot be so straightforward, as source events include material objects: there is no such thing as a rolling (or collision, scraping etc.) simpliciter, something needs to be doing the rolling. Given this complication, is it possible to make sense of the idea that events and not objects are fundamentally what is heard? Can we hear a rolling without hearing that which rolls? We shall see later that there are those who argue that we can. Alternatively, perhaps audition, like vision, picks out material objects, and that ‘hearing a rolling’ is just an elliptical way of saying ‘hearing a ball’. But questions can be asked of this approach too. How does hearing a material object differ from seeing it? How should we characterise the difference between different types of events involving the same object, such a ball rolling and a ball bouncing? In vision the relation between object seeing and event seeing is straightforward. If material things are what we see in the first instance, then we see events by seeing objects: seeing a rolling consists in seeing a ball and seeing it to move, to change its location and orientation. Seeing different types of event consists in seeing the object move in different ways: the ball rolls horizontally across the ground, and falls vertically towards the ground. Might we hear events in the same way? That is, does source perception consist in the auditory system picking out objects and hearing them to change location? We might already be suspicious that this is the correct way to characterise source hearing, but it will be instructive to bring out exactly why. I suggest that we cannot hear events in the same way that we see them because seeing movement depends upon a certain structural feature of vision, the spatial field, which audition does not share. [^4]: Representationalists such as Cohen, Nanay, and Matthen hold that perceivers represent material objects and attribute, or predicate, properties to them, but relationalists also think that material objects are, fundamentally, what we see, in that perception consists in their standing in a certain type of relation to a subject. See, for example, Brewer (2007). [^5]: Less mainstream views of vision can be found in Clark (2000) who argues that we see, first and foremost, places, portions of space, not the things which occupy them. And Matthen (2010) who suggests that we see two types of individual simultaneously: material objects, and the ambient light illuminating those objects. [^6]: Property variation will be returned to later when change perception is discussed. [^7]: We will see later that there is more to seeing movement than this. moving objects is revealed by hearing the location of the source event in which they participate, we do not hear source events by hearing these objects to extend over different regions of space at different moments.8 Finally, notice that, unlike vision, auditory experiences can sometimes lack spatial information entirely, yet our ability to recognise events is not impeded. If a mono recording of a rolling ball is played back on headphones it will still be identifiable as a rolling, yet it will present the listener with no information as to which direction the object is travelling. There is no visual equivalent of such a case.9 If hearing movement cannot be modelled on seeing it, we need a different account of source perception. I shall now examine two suggestions as to what this might be. On the first, we can hear source events but not the objects participating in those events. On the second, hearing sources is conceived of as analogous to seeing the appearances of objects to change. I shall argue that neither approach is satisfactory and go on in Sect. 5 to put forward my own account. [^8]: The degree to which audition provides spatial information is controversial See, for example, Strawson (1959, pp. 65–66) Casati and Dokic (2009). [^9]: At least, no non-pathological equivalent. Patients suffering from simultanagnosia (a symptom of Bálint’s syndrome) can perceive only one object at a time and are unable to locate that object in relation to others. It is possible that if a sufferer of simultanagnosia were able to track an object as it moves (those with this condition also have difficulty fixating on moving objects) they would perceive an object moving, but not see where it was moving from or to (see Rafal 1997, 2003). Thanks to an anonymous referee for drawing my attention to these cases. [^10]: He also cites Strawson (1959), Dretske (1969, p. 20), Martin (2007, p. 706). This type of approach reverses the order of priority that we find in vision: we do not see events, but only objects moving or changing; we do not hear objects, but only events unfolding. While the idea that vision deals with objects and audition with events has a certain neatness, I think objectless accounts can be challenged in three ways: a lack of spatial singling out is not a good reason to give up the idea that we hear material objects, the idea that we hear source events but not their participating objects raises more questions than it answers, and we can ask whether an account of source perception which discards material objects is even desirable. First, while it may be that spatially singling something out is a necessary condition for seeing it, making a similar requirement for audition would seem to preclude the hearing of just about anything. As we have seen, although audition is spatial, it is not field-like in the way that vision is; whatever we hear, we do not hear as filling regions of space. Source events are not auditorily singled out: we might hear the rolling to be over there, rather than over there, but we do not differentiate it from its surroundings or hear it as shaped. If this is correct, why should we think event individuals are a better candidate for what is heard than material objects? Neither, if they are understood as individuals distinct from source events, it is plausible to think that we spatially single out sounds. Consider O’Callaghan’s own proposal as to what sounds are: sounds are the events in which a medium is disturbed or changed or set into motion in a wave–like way by the motions of bodies. Events such as collisions and vibrations of objects cause the sound events. (2010, p. 36) On this account, sounds extend in space: there are portions of the environment in which the air (or some other medium such as liquid) is disturbed by the motions of bodies, and portions where it is not, but we do not hear them to extend over space in this way, we do not hear where they end and the rest of the world begins. It might be objected here that sounds, unlike objects, have fuzzy boundaries, as air will be disturbed less and less the further away it is from the vibrating object, but this is beside the point; we do not hear individuals with fuzzy boundaries, as we might see a cloud of smoke to fade at its edges, instead audition presents no information about boundaries at all. Again, this should not be taken as the claim that audition is non-spatial: we hear what we hear to be located, but we do not hear things as located in the same way that we see them to be. In short, a lack of spatial singling out can be understood as a characteristic way in which audition presents whichever features of the world it presents, not as evidence that it presents one type of individual or another. If this is correct then material objects have no less a claim of being heard than anything else. Second, this approach to source hearing offers a muddy answer to the question of what we hear when we hear sources. We have already seen in Sect. 2 the difficulties in making sense of the idea that we hear source events but not the objects participating in those events. How can we hear a rolling without hearing that which rolls? Moreover, unless they are willing to give up outward orientation, defenders of this position are required to say which aspects of the physical environment are auditorily picked out in source hearing if not material objects. It cannot be the air that moving objects disturb, as this would be to hear sounds produced by sources rather than sources themselves, but there appear to be a dearth of other candidates. What other aspect of the physical environment could be picked out by our auditory systems and still be properly considered a source event? Discarding objects also causes problems in accounting for the audible similarities between different source events involving the same object. Consider, vision: the experience of seeing a ball dropped on the ground, and seeing it roll along the ground have something in common. Specifically, obviously, we see the ball in both cases. Consider audition: if I drop the hard, heavy, bowling ball on the ground, and then roll it towards you, the auditory experiences each event elicits have, arguably, something in common. You can tell, just through hearing, that it is the same object, or at least an object with the same properties, taking part in both events. But if the audible similarity between these cases cannot be explained in terms of them sharing objects, how can it? Problems remain even if a slightly weaker version of the position is adopted. Elsewhere in his (2013) O’Callaghan considers the possibility that we in fact do hear objects, but only derivatively, as ‘‘constituents’’, ‘‘protagonists’’, or ‘‘sufferers’’ of source events (p. 23). However, it is again not obvious how such an idea should be understood.11 While it is straightforward to explain how we see events by seeing objects—seeing an event is to see an object move or change—it is much less clear how to make sense of the idea that we hear events first and foremost, and through doing so hear their protagonists. Third, it is questionable whether we would want an account of source hearing which excludes material objects. Recall the reasons mentioned in Sect. 1 for wanting an account of audition which allows for source perception: to not do so would be ‘‘severe’’ (O’Callaghan 2011, p. 1), put interesting events at ‘‘one remove’’ from the listener (Kulvicki 2008, p. 15), and conflict with the ‘‘phenomenological intimacy’’ apparent in auditory perception (Leddington 2014, pp. 328–329). However, excluding material objects from what we hear is, arguably, quite severe, as our interest in source events is not only in the type of event taking place (a rolling, collision etc.), but also in which objects are participating (what is rolling, colliding etc.): we are interested in the smashing because it is the smashing of an expensive vase, or our kitchen window. Neither is it obvious that we are any less auditorily intimate with the objects than with their activities: we learn immediately from what we hear that a rolling is taking place, but we become aware, no less immediately, that it is a heavy, round, object doing the rolling. Excluding objects from what we hear risks throwing out the baby with the bathwater. [^11]: I am also uncertain how such a claim can be reconciled with O’Callaghan’s stronger claims quoted at the beginning of this section. visual model of seeing movement, however, event perception is not a case of perceiving objects to move through space, but is instead analogous to seeing an object’s appearance change as the light illuminating it changes. To see how this works, we first need to look at his account of what sounds are. On Kulvicki’s stable disposition view (2008, 2016) sounds are conceived of as stable properties which objects are heard to bear, much like colours are properties objects are seen to bear.12 An object’s size, shape, and material constitution, will dispose it to vibrate at particular frequencies when it is stimulated, in the same way that features of an object’s surface will dispose it to reflect certain frequencies of light when it is illuminated. Reflectance physicalist theories of colour, such as that defended by Byrne and Hilbert (1997), identify colours with reflectance dispositions and Kulvicki has argued that we should identify sounds with vibratory dispositions: flicking the light on reveals an object’s colour, throwing it on the floor reveals its sound (2008, p. 5). In contrast to objectless accounts, vibratory dispositions provide an explanation of the audible similarities between different source events involving the same object. Because an object will vibrate at more or less the same frequencies, and so be heard to have the same sound property regardless of how it is stimulated, meaning they the same object can be recognised across different source events: you can hear that the object rolling along the ground is the same type of object you heard dropped onto the ground a moment ago, because you hear it to have the same sound in both cases. In and of itself, however, this does not provide a complete account of how we hear source events, as stable sound properties do not account for the audible dissimilarities between source events. While a dropped ball and a rolled ball sound similar in some ways, hearing one is still quite different from hearing the other. Something therefore needs to be added to the stable disposition view to account for these differences. Kulvicki’s solution is to argue that hearing what is happening to an object is analogous to seeing its appearance change as the light illuminating it changes. We can think of different types of illumination as changing the appearance of an object, but not affecting the colour it is perceived to have: the green grass at dusk looks like green grass illuminated oddly. The grass at noon on a sunny day looks like grass illuminated brightly, while on a cloudy day it looks more softly illuminated. These are all aspects of the object’s appearance, and some of them concern the illuminant. Visual experience reveals aspects of the illuminants that allow one to see color in the first place (2008, p. 9, my emphasis) Although he does not put it in these terms, that ‘‘an aspect’’ of the grass’s appearance differs depending on the illumination can be understood as the object being seen to bear two types of property simultaneously: its stable colour property [^12]: In more recent work, Kulvicki argues that ‘‘aspects of the events that cause vibrations, such as intensity, and properties of the spaces in which one hears’’ (2016, p. 91) are heard as well as stable vibratory properties. He does not, however, return to the question of how source hearing should be understood. and its illumination dependent appearance property.13 It is easy to think of cases in which appearance properties are seen to change, while the colour of the object is seen to remain stable: Change the spectral composition, brightness, orientation, or spatial extent of illumination and you notice these events… These events do not seem to affect the colors of things, but they crucially involve perception of things’ colors. Spotlights move quickly across the room. Phenomenologically, perception of their motion is tied to the revealing of things’ shapes and colors as the lights move across the otherwise dark space. We see many events by seeing things’ colors…things are similar with sounds. (ibid., p. 12) If illumination changes are used as a model for source hearing, the picture would be as follows. When an object vibrates in our vicinity, we hear it as having two types of property: a stable sound property and a changing appearance property. Which appearance properties an object is heard to change through, depends upon the source event in which it is participating: a quickly dimming light will lead to an object taking on a different sequence of appearances than a flickering one, a collision will lead to the ball taking on a different sequence of appearances than a rolling. As the grass is still seen as being green regardless of how its appearance changes, so too is the ball’s sound heard across various changes in its appearance. Unlike objectless source hearing, then, the stable disposition view, coupled with appearance properties provides an explanation as to how source events can be heard to be similar or different. Moreover, while on this account event hearing depends on object hearing, it does not suffer the difficulties we encounter when we try to model hearing sources on seeing movement, because it does not require that audition have a spatial field. There is no reason to think that perceiving an object to have a stable sound property or changing appearance requires seeing it as bounded, or to extend over a portion of space. Such an account of object hearing might be compared to touch. If you lay your palms flat on the table, you feel the table. The fact that you do not feel it’s edges, and so are not able to pick it out from its surroundings, does not impede on you feeling the object itself. What is more if the properties of the object change while you are touching it, for example if it started to heat up, you would be able to detect these changes. On Kulvicki’s account, hearing objects can be thought of as similarly non-specific. Despite these advantages, I will now argue that this approach ultimately fails. Hearing sources is not analogous to seeing changing appearance properties because hearing sources cannot be a case of hearing objects change at all. [^13]: This dual property account of colour constancy is somewhat similar to accounts proposed by Schellenberg (2008) and Shoemaker (2006). motion, a sequence of locations. Consider the two visual examples we have been working with so far: seeing the dimming event involves seeing the floor take on a series of different appearance properties as the light dims; seeing the rolling involves seeing the ball to occupy a series of different locations. It is easy to think of similar cases for other types of property: seeing a chameleon change colour is to see it as bearing one colour at one moment and another the next, seeing a balloon deflate involves seeing it to have a succession of different shapes. This I take to be uncontroversial and intuitive: for change to be seen, it is necessary for something to look one way and then another. As has been widely remarked upon, however, it is insufficient: a light can dim, or a chameleon change colour, before your very eyes, but at a rate so slowly as to be imperceptible. In these cases, we still see individuals to bear sequences of properties—stare at the chameleon long enough and you will notice that it has a different colour than it did a moment ago—but you do not see this change, as your visual experience is as of an unchanging, individual. Often, the difference between perceptible and imperceptible change is explained with the idea that perceptual experience is structured by a temporal field, or temporal window (see, for example, Phillips 2011, p. 363; 2013, p. 5; Richardson 2014, p. 494; Dainton 2008, p. 634).14 This is the idea that the world is not experienced instant by instant, but rather at intervals of about a second and a half or less.15 It can be thought of as the temporal counterpart to the visual spatial field: … the things we perceive are perceived as filling, occupying, or having some location within, an interval of time, just as the objects we see are generally seen as filling, occupying, or having a location within a region of space (Soteriou 2011, p. 195) Recall that one way the spatial field adds to visual experience is that it allows for the perception of property variation. This feature suggests a way in which we might differentiate cases of perceptible and imperceptible change. Imagine standing before a painting, consisting entirely of a large block of red getting gradually darker along its horizontal axis: from a bright ‘fire engine’ shade on the far left of the canvas to a darker ‘merlot’ shade on the far right. Standing a sufficient distance away, so that the whole canvas falls within your visual field, will allow you to take in the colour gradient in one go: you see each shade running into the next, and thereby see the variation in colour. Stand too close however, and you will experience a uniform shade of red. Although the colour of the paint within your field of view does vary, the shades of red at the far sides of your visual field of view are so similar as to be indistinguishable: the variation is not seen. [^14]: This idea has a long history and many variations. It can be traced back to William James’ notion of the specious present: ‘‘The practically cognized present is no knife-edge, but a saddle-back, with a certain breadth of its own on which we sit perched’’ (1890, Vol. I, p. 609). [^15]: Although James suggested that the specious present could be as long as a minute, most contemporary versions of this idea estimate it to be much shorter. For example, Dainton (2000, p. 171) estimates it to be about half a second or less, Lockwood ‘‘a second or a second and a half’ (2005, p. 381), and Phillips supposes it to be approximately 300 ms (2011, p. 817). The difference between perceptible and imperceptible change can be thought of as equivalent to that between perceptible and imperceptible variation. That is, if an object’s colour, appearance, or location is changing quickly enough so that it is discriminably different within a single window then the different shades are experienced together [they are ‘‘diachronically co-conscious’’ to use a phrase of Dainton’s (2016, p. 135)] and the change is experienced, equivalent to variation being experienced when standing at a suitable distance from the canvas; if the object changes more slowly, so that it does not have a discriminably different colour within a single temporal window then experience presents an unchanging object, equivalent to variation not being experienced when standing too close to the canvas.16 Visual experience, then, comes in two varieties. When discriminably different properties are unified, there is an experience of a changing individual; when discriminably different properties are not unified (either because the object is not changing, or changing very slowly) we have an experience of a unchanging individual. Notice also that the same colours, appearances, or location properties, are perceived in cases of changing or unchanging objects: any property an individual is seen to change from, to, or through, is one that it could potentially have been seen to possess indefinitely. We might see the chameleon change from red, through various shades of orange, to yellow, but, the creature could decide to settle on any one of these colours for longer, or change so slowly that it looks as if it has a stable colour when it is in fact changing; similarly, a floor’s appearance can be seen to change as the light illuminating it dims, but if the dimming is halted halfway, the floor will take on a steady appearance that it would have had only fleetingly. Crucially, however, there is only one variety of source perception. Whereas we can see a bowling ball to be either moving or at rest, changing or unchanging, we only ever hear a bowling ball as being rolled, dropped, or moving in some other way, and never simply a bowling ball, going nowhere and doing nothing. If this is correct, then it is implausible to think that we hear events by hearing unified sequences of potentially stable properties. First, we can note that inactive objects never cause auditory experiences. Whereas material objects can reflect light onto the retina, and thereby elicit visual experiences, regardless of whether they moving or stationary, changing or unchanging, this is not true of audition. Unmoving, unstimulated, objects do not vibrate so as to produce sound waves, and so do not produce auditory experiences in listeners. The more important point, however, is phenomenological: we never have auditory experiences of entirely unchanging objects, even in unusual or contrived circumstances. Notice, first of all, that there is no auditory analogue of slow, imperceptible, change. Regardless of how slowly an object moves, we never have an auditory experience of that object not moving: if the ball rolls quickly enough so as to vibrate and produce sound waves of an amplitude high enough for our ears to detect, then a rolling is heard; if it rolls too slowly, creating sound waves of too low an amplitude, [^16]: Using the temporal field as a means of differentiating perceptible and imperceptible change is based on Hoerl (2013, p. 18) and Phillips (2011, p. 11). we hear nothing. Neither does it seem possible to elicit an experience as of an unchanging object artificially. Consider the differences in how video and audio recordings decompose. If a video were made of the changing chameleon, rolling ball, or dimming spotlight and then opened on some video editing software, it would be possible to look at the individual frames which make up the recording (for an analogue recording it would be possible to look at the individual frames of the film itself). Each frame of each recording would depict an individual with a certain stable, unchanging property: the chameleon a particular colour, the ball at a definite location, the floor under a certain intensity of illumination.17 The order of the frames could be reversed to create a video of a chameleon changing from yellow to red, a ball traveling back up the lane, the floor’s appearance becoming brighter rather than darker. Similarly, if you took a pair of scissors to the painting, individual segments of certain colours could be cut out and rearranged to form, for example, a colour gradient going in the other direction: the painting can be decomposed spatially in the same way that the video of a dimming event can be decomposed temporally. These possibilities show how plausible it is to think that we see events by seeing property sequences, as they allow us to isolate and manipulate the constituent properties of that sequence. Audio recordings, however, cannot be taken apart in the same way. If a recording of a rolling event is sliced into snippets, then, regardless of how fine or coarse the slices are, we never find a snippet presenting simply a ball with an unchanging appearance. Very fine slices would present nothing identifiable as a rolling event, only a ‘click’ would be heard (Gefland 2009, pp. 170–171). Listening to coarser and coarser snippets you will eventually come across one which would sound like a momentary rolling event, but there would be no ‘sweet spot’ in between the coarse and the fine that elicits an experience of an unmoving ball. Similarly, reversing an audio recording of a rolling ball and playing it back does not elicit an experience as of a ball rolling in the other direction but of an unidentifiable woosh, just as reversed recordings of drums or guitars on old psychedelic rock records sound nothing like drums or guitars. These examples show quite how different hearing events must be from seeing events. If source hearing were a matter of hearing an object to change, as bearing a unified sequence of properties, we would be able to find cases in which we auditorily experience objects the properties of which are not changing. Given that we do not, we still lack a proper account of what it is to hear the movements and interactions of objects. [^17]: Certain types of video playback can be thought of as something like the inverse of imperceptible change outlined at the beginning of this section: rapidly changing—rather than slow moving– stimuli can give rise to visual percepts of unchanging objects. The images on old cathode ray tube televisions, for example, are generated by electron beams illuminating different coloured pixels on the screen in rapid succession. Although most of the time this process will generate images of things moving or changing— people talking, balls rolling etc.—it is easy to think of cases in which the successive illumination of pixels generates images of entirely unchanging objects—people staring into the distance, balls at rest. Images generated in this way, however, still decompose in something like the same way as video files or frames of a film: any one of the pixels which make up an image could be presented to the viewer for a much longer time, giving rise to the perception of a single, unchanging, dot on the screen. I am grateful for an anonymous referee for drawing my attention to this type of case. In the next section I shall put forward a new way of thinking about source hearing, but before moving on, two clarificatory remarks are in order. First, denying that we ever have auditory experiences of unchanging objects is not equivalent to denying that we can have experiences of unchanging sounds.18 Recall that sounds are sometimes conceived of as individuals that objects produce, rather than properties that objects bear. If we accept this, along with the common sense idea that sounds themselves bear properties such as timbre, pitch, and loudness, it is easy to think of cases in which a source event produces a sound with stable properties. We might think that a rolling is just such an event, as the pitch, timbre, and loudness of the sound it produces are fairly stable as the ball rolls down the lane. Another example might be a the sound produced by the bowing of a violin string: the note produced would sound more or less the same at three seconds as it would at six. However, cases such as these are not counterexamples to the argument I have made in this section, as they are not examples of our hearing stable objects. The ball and the bow are still heard as doing something, rolling or being bowed, even if we hear the stable sounds that these activities produce as well. Second, I do not deny that we can learn of the presence and location of unmoving objects from what we hear.19 Imagine the auditory experience elicited by a ball slowing to a halt as it rolls from your right to your left. From what you hear you might well judge that there is now an unmoving ball on your left, but this is not an auditory perception of an unmoving ball. Your experience of a ball slowing down is not followed by an experience of a ball at rest; you hear a ball rolling ever more slowly, and then…nothing. Consider also how you might make a mistake in cases like this. Unbeknownst to you I might pick up and move the ball just as it stops, or it might roll onto a plush carpet, continuing its journey but no longer producing sound waves. In either instance, you might think that there is an unmoving ball in a certain location when in fact there is not. The point here is not that mistakes can be made, but that they are much better described as misjudgements on your part, rather than the undergoing any type of perceptual illusion. You are not mishearing the ball to be at one location when it is not, as you might mis-see it if I arranged some clever trick with mirrors. Rather, you are drawing an erroneous conclusion about the state of the world at a given moment from what you perceived a moment before. [^18]: This type of challenge was first put to me by Casey O’Callaghan. [^19]: I am grateful to an anonymous referee for pressing me on this point. account of source hearing after all. Perhaps Berkeley’s severe, ‘sounds only’ view is the only option? In this final section I will suggest how this can be avoided by putting forward a new model for source hearing, one which retains the idea that we hear material objects, provides an account of the audible similarities and differences between source events, and explains why we never hear unmoving objects. In short, I will argue that to hear an event is to hear an object as perduring, as extending through time. This proposal draws on a debate in metaphysics as to the nature of persistence. Lewis distinguished two ways in which individuals might persist across time: An object endures if it persists by strictly retaining its identity at each time, with no temporal parts, such that it exists in its entirety at each moment at which it exists; whereas an object perdures if it persists by having different parts at different times. (See Lewis 1986: §4.2.) We ordinarily think of the things around us as enduring. The bowling ball is complete at any moment of its existence, and if it is black at 9.10am and repainted red at 9.11 a.m., then we would say that the whole ball is black at 9.10 a.m. and, providing the repainting is complete, the whole ball is red at 9.11.a.m. A bowling match, on the other hand, is incomplete at any single moment: it does not exist its entirety at either at 9.10 or 9.11. Instead, we can think of it as having temporal parts. The game has a beginning, a middle, and an end; if we say ‘‘the whole bowling match’’ we are referring to something which stretches over an extended period of time. Some philosophers, including Lewis, have argued that material objects are in fact more like bowling matches than we might think, in that they perdure rather than endure through time. If this is correct, then the bowling ball is actually incomplete at any moment, and has different coloured temporal parts located at different points in time.20 I shall not debate the merits of metaphysical endurantism and perdurantism here. What is important is that each view provides a potential model as to how objects in the world are perceived. For example, Prosser, drawing on Kant (1929/1781-7: pp. 216–217), suggests that, regardless of what is true metaphysically, visual perceptions of change present material objects as enduring and not perduring: in order for there to be a representation of change…there must also be a representation of something that retains its strict identity through the change – the very same thing is first F, then not F…whatever the truth may be about the metaphysics of persistence, I do not think that this adequately captures change as we experience it. Change is not experienced as an F temporal part succeeded by a non-F temporal part, with it somehow understood that both parts belong to the same composite whole; this does not correctly capture the phenomenology. (2012, p. 14) [^20]: Although Endurantism and Perdurantism can be thought of as the two main options for characterising persistence, there are various other possibilities, such as stage theory (Hawley 2001; Sider 2001). I have argued that source hearing cannot be a case of change perception. We do not hear events by hearing objects to be F and then not F, because if we did we would expect to find cases in which we hear an object to be simply F. In perdurantism, however, we have a potential alternative. I suggest that, whatever the truth may be about the metaphysics of persistence, auditory experience presents individuals as perduring, as composite wholes made up of temporal parts. Hearing a source event is to hear a material object extend across a duration of time, and the type of event in which it participates determines which temporal parts it is heard to have. What is more, perdurance perception, like change perception, can be understood as a function of the temporal field. [^21]: Prosser’s overall aim in this paper is to give an account of the phenomenology of temporal passage, the experience of time as flowing, which can be reconciled with metaphysical accounts on which time does not flow: ‘‘…a key factor in time seeming to pass is that change is experienced as dynamic, and change is experienced as dynamic because the experience involves the representation of something enduring through the change. It is this notion of a single entity passing ‘through’ a change that captures at least a very important element of the experience of temporal passage’’ (2012, p. 16). One cost of rejecting the idea that perceptual experience always presents objects as enduring is that it would appear Prosser’s account of experienced passage must be given up as well. Given that there are other accounts available as to why passage is experienced which do not rely on objects being experienced as enduring (e.g. Paul 2010; Skow 2015; Torrengo 2017), as well as doubts that there even is a phenomenology of passage (Hoerl 2014), I think that this is a price worth paying. Whether all of an object’s temporal parts are unified at once depends on the length of the event. An object’s size, or the viewer’s distance from it, determines whether or not its spatial parts fit within a single visual field: at a suitable distance, the whole of a painting can be taken in at once, but if the picture is very large, or the viewer very close, only a portion of it will be taken in at once and a turn of the head, or a step backwards will be needed to see its other parts. Similarly, shorter events will fit within the temporal field, but longer ones will not: a bowling ball colliding with the floor happens very quickly, and so all of its temporal parts will be diachronically co-conscious within the same window; a rolling event, on the other hand, might last much longer and so be heard to fill and continue on from a temporal window, with some but not all of its parts diachronically co-conscious at a given moment.22 Like the appearance change model, perdurance perception provides an account of the audible similarities and differences between source events. We saw in the previous section that the difference between hearing a ball being rolled and hearing it being struck can be explained in terms of the object’s appearance being heard to change differently in each case. On the account I propose here, hearing an object to participate in different types of source event is to hear it to have different temporal parts, or, in other words, to have a different temporal shape, in each case. Consider two differently shaped objects made from the same material, say a bowling ball and a skittle. From looking at them it is obvious that they are made from the same, hard, shiny, substance, but it is also obvious from the way in which each extends through space that one is a ball and one is a skittle. We can think of the difference between hearing a ball rolling and hearing it colliding as similar: we hear the same sort of hard, round, object in each case, but hear it to extend through time differently in each case. Conversely, just as we can recognise the same type of object made from different material,—a wooden, rather than plastic skittle, for example—from their similar shapes, we can recognise the same thing happening to different objects—a rolling ping pong ball, a rolling bottle—by the similarities in temporal shape. Unlike appearance change, however, temporal shape provides a satisfying account of the differences between vision and audition that we encountered at the end of the previous section. First of all, it explains why a video can be broken down into individual frames depicting an unchanging object, but an audio recording cannot. While taking a pair of scissors to the red painting will result in scraps of canvas with different shades of red, cutting up a drawing of a circle will not produce small scraps of circularity because the scraps no longer stand in the necessary spatial relations to be seen as part of a circle. Similarly, a snippet of audio will not be heard as part of a rolling unless it is heard to stand in appropriate temporal relations to [^22]: Parsons (2000, 2004) proposes an alternative to the standard four dimensionalist model. He argues that one can accept objects are extended in time, without committing to the idea that they have temporal parts. Key to this is the idea that as well as possessing ‘‘qualities’’, such as being red or being warm, objects can also possess spatial or temporal distributional properties, such as being polka-dotted, or being warm at one time and cold at another. On this account, objects endure, and yet are temporally extended. It is possible that the ideas I put forward in this section could be worked out in terms of temporally distributional properties, rather than temporal shape, although I do not see that very much hangs on this difference. Parsons suggests that shape is a ‘‘degenerate case’’ of a distributional property (2004, p. 4). other snippets. We can also say that the reason why we do not hear objects unless they are participating in some event or other is similar to the reason we never see unshaped objects. We have seen that temporally extended visual perception is taken to be similar to seeing properties vary across space: as the spatial field unifies properties synchronically, the temporal field unifies them diachronically. This allows for two sorts of experience in each case: if discriminably different properties are unified then we have experiences of change or variation, if properties are indiscriminable, or if there is only one property to be seen, we have experiences of unchanging or invariant properties. Unification of spatial parts, on the other hand does not have this either/or structure: it is not possible to see an object without seeing it as filling space in some way, as being spherical, cubular, pointy etc. If, as I have argued, source hearing is analogous to shape perception, we would expect it too to a monomorphic structure. It is not possible to hear an object without hearing it as extending through time in some way, as colliding, rolling or having some other type of temporal shape. ## References Berkeley, G. (2012). Three dialogues between Hylas and Philonous. Peterborough: Broadview Press. Brewer, B. (2007). Perception and its objects. *Philosophical Studies*, 132(1), 87–97. Byrne, A., & Hilbert, D. (1997). Colors and reflectances. Behavioral and Brain Sciences, 26(1), 3–21. Casati, R., & Dokic, J. (2009). Some varieties of spatial hearing. In M. Nudds & C. O’Callaghan (Eds.), Sounds and perception: New philosophical essays (pp. 97–110). Oxford: Oxford University Press. Clark, A. (2000). A theory of sentience. Oxford: Oxford University Press. Cohen, J. (2004). Objects, places, and perception. Philosophical Psychology, 17(4), 471–495. Dainton, B. (2000). Stream of consciousness: Unity and continuity in conscious experience. Routledge: Taylor & Francis. Dainton, B. (2008). The experience of time and change. *Philosophy Compass*, 3(4), 619–638. Dainton, B. (2016). Time and space (2nd ed.). New York: Routledge. Dretske, F. I. (1969). Seeing and knowing. Chicago, IL: The University of Chicago Press. Hawley, K. (2001). How things persist. Oxford: Oxford University Press. Hoerl, C. (2013). A succession of feelings, in and of itself, is not a feeling of succession. Mind, 122(486), 373–417. Hoerl, C. (2014). Do we (seem to) perceive passage? Philosophical Explorations, 17(2), 188–202. James, W. (1890). The principles of psychology (Vol. 2). London: Macmillan. Kant, I. (1929/1781–7) Critique of pure reason. Translated by Norman Kemp Smith. London: Macmillan. Kulvicki, J. (2008). The nature of noise. Philosophers’ Imprint, 8(11), 1–16. Kulvicki, J. (2016). Auditory perspectives. In B. Nanay (Ed.), Current controversies in philosophy of perception. New York: Routledge. Leddington, J. (2014). What we hear. In Brown, R. (Ed.), Consciousness inside and out: Phenomenology, neuroscience, and the nature of experience (pp. 321–334). Netherlands: Springer. Lockwood, M. (2005). The labyrinth of time: Introducing the universe. Oxford: Oxford University Press. Martin, M. (1992). Sight and touch. In T. Crane (Ed.), The contents of experience. New York: Cambridge University Press. Martin, M. G. F. (1997). The shallows of the mind. *Proceedings of the Aristotelian Society* Supplementary Volumes, 71, 75–98. Martin, M. G. F. (2007). Perception. In F. Jackson & M. Smith (Eds.), The Oxford handbook of contemporary philosophy, chapter 24 (pp. 701–738). New York: Oxford University Press. Martin, M. G. F. (2012). Sounds and images. *The British Journal of Aesthetics*., 52(4), 331–351. Matthen, M. P. (2005). Seeing, doing, and knowing: A philosophical theory of sense perception. Chicago: Oxford University Press. Matthen, M. (2010). How things look (and what things look that way). In B. Nanay (Ed.), Perceiving the world (pp. 226–253). New York: Oxford University Press. Nanay, B. (2013). Between perception and action. New York: Oxford University Press. Nudds, M. (2010). What sounds are. In D. Zimmerman (Ed.), Oxford studies in metaphysics 5 (pp. 279–302). Oxford: Oxford University Press. O’Callaghan, C. (2007). Sounds: A philosophical theory. Oxford: Oxford University Press. O’Callaghan, C. (2011): Hearing properties, effects, or parts? In Proceedings of the aristotelian society (Vol. 111, No. Part III, 375–405). Parsons, J. (2000). Must a four-dimensionalist believe in temporal parts? *The Monist*, 83(3), 399–418. Parsons, J. (2004). Distributional properties. Lewisian themes, 173–80. Paul, L. A. (2010). Temporal experience. The *Journal of Philosophy*, 107(7), 333–359. Phillips, I. (2011). Indiscriminability and experience of change. The *Philosophical Quarterly*, 61(245), 808–827. Phillips, I. (2013). Hearing and hallucinating silence. In F. Macpherson & D. Platchias (Eds.), *Hallucination* (pp. 333–360). Chicago: MIT Press. Prosser, S. (2012). Why does time seem to pass? *Philosophy and Phenomenological Research*, 85, 92–116. Rafal, R. (1997). Balint syndrome. In T. Feinberg & M. Farah (Eds.), Behavioral neurology and neuropsychology (pp. 337–356). New York, NY: McGraw-Hill. Rafal, R. (2003). Balint’s syndrome: a disorder of visual cognition. In M. D’Esposito (Ed.), Neurological foundations of cognitive neuroscience (pp. 27–40). Cambridge, MA: MIT Press. Richardson, L. (2014). Space, time and Molyneux’s question. *Ratio*, 27(4), 483–505. Schellenberg, S. (2008). The situation–dependency of perception. The *Journal of Philosophy*, 105, 55–84. Shoemaker, S. (2006). On the way things appear. In T. S. Gendler & J. Hawthorne (Eds.), Perceptual experience (pp. 461–480). Oxford: Oxford University Press. Sider, T. (2001). *Four-dimensionalism*: An ontology of persistence and time. Oxford: Oxford University Press on Demand. Siegel, S. (2009). The visual experience of causation. The *Philosophical Quarterly*, 59(236), 519–540. Skow, B. (2015). Objective becoming. Oxford: Oxford University Press. Soteriou, M. (2011). The perception of absence, space, and time. In J. Roessler, H. Lerman, & N. Eilan (Eds.), Perception, causation, and objectivity. Oxford: Oxford University Press. Strawson, P. F. (1959). *Individuals*. New York: Routledge. Torrengo, G. (2017). Feeling the passing of time. *Journal of Philosophy*., 114, 165–188.