# Hearing Spaces **Nick Young** University of Antwerp ## Abstract In this paper I argue that empty space can be heard. This position contrasts with the generally held view that the only things that can be heard are sounds, their properties, echoes, and perhaps sound sources. Specifically, I suggest that when sounds reverberate in enclosed environments we auditorily represent the volume of space surrounding us. Clearly, we can learn the approximate size of an enclosed space through hearing a sound reverberate within it, and so any account that denies that we hear empty space must instead show how beliefs about volumes of space can be derived indirectly from what is heard. That is, if space is not auditorily represented when we hear sounds reverberate, what is? I consider whether hearing reverberation can be thought of as hearing a distinct sound, hearing echoes, or hearing a property of a sound. I argue that experiences of reverberation cannot be reduced to the perception of any of these types and that therefore empty space is represented in auditory perceptual content. In the final section I outline two ways in which space might be represented. **Keywords:** perception; hearing; sounds; space ## 1. Reverberation and Empty Space What do we hear? Three obvious candidates are sounds, properties of sounds, and echoes. We hear the chime of a bell, its timbre and pitch, and—in some cases—its echo a moment later. Slightly more controversially, we might think that we hear sound sources, the vibrating objects that produce sound waves in the air.[^1] We hear the bell. My aim in this paper is to show that we also hear empty spaces. We hear the volume of empty space enclosed by the walls of the church. My argument will be as follows. It is clear that the size of an enclosed space has an effect on auditory experience and that we come to know the approximate size of the room from what we hear. We know, for example, that the bell was struck in a large room rather than a small one, because hearing a chime in a large room sounds different to hearing one in a small room. The most promising way of denying that this is due to empty space being heard is to say that the size of the room affects what we hear in some other manner—such as producing extra sounds or echoes, or affecting the properties of sounds heard within them—and that unconscious inferences are made about the volume of surrounding space from these features. I will argue that it is implausible to think that enclosed spaces affect what we hear in any of these ways and that there is therefore no perceptual basis from which inferences about space can be made. If awareness of space cannot be explained in terms of what we uncontroversially hear, we should consider more than just the obvious candidates. I propose that we learn about a volume of space because we hear that volume of space, and that we might hear spaces in the same way that some philosophers think that we hear sound sources. It is necessary to clarify what I mean by 'hear'. My interest is in the contents of auditory perception, in what individuals and properties are auditorily represented. What perceptual systems represent should not be confused with what causes perceptual experiences: the proximal cause of visual experience is light hitting the retina, but we do not take ourselves to be seeing rays of light. Rather, retinal stimulation is thought to elicit representations of things in the world, such as material objects. For the same reasons, we should not necessarily interpret the claim that we hear sounds as the auditory system representing sound waves. As we shall see in the next section, some philosophers do indeed think that the auditory system represents sound waves themselves; but others argue that it represents other types of individual, such as the disturbance of air around a vibrating object [O'Callaghan 2007], and use the term 'sound' to refer to those. Still others argue that, like some accounts of vision, audition involves the representation of material objects and use the term 'sound' to refer to properties they are heard as possessing [Kulvicki 2008]. Debates over what sounds are, or whether we hear material objects, can both be thought of as disagreements over what individuals are represented by the auditory system when sound waves hit the ears. I claim that sound waves hitting the ear also lead to the representation of volumes of empty space. This position contrasts with that of Nudds, who explicitly denies that empty space can be heard [2009: 88]: > Unlike visual experience, auditory experience does not represent empty places—it does not represent places as unoccupied. … Unlike our visual experience, our auditory experience of space is exhausted by our awareness of spatial relations between sound sources and us, and between sound sources and other sound sources. In a dark room, we can be aware of a bell ringing on our left and someone speaking on our right but have no awareness of whether there are any objects standing, or silent events unfolding, in between them. Nudds argues that this makes audition more similar to touch than to vision. If we touch the rim of a glass with our eyes closed, we can feel where the different parts of the rim are located in relation to each other, but our experience remains neutral as to the layout of the rest of the scene: we would not know if the glass was empty or filled, for example [Martin 1992: 199–200]. However, if two or more objects are in our line of sight we see not only their locations in relation to ourselves and each other, but also the space that separates them—the places that could potentially be filled with an object. I will argue here that our auditory experience of space is not 'exhausted' by our awareness of spatial relations between different sound sources and ourselves. Specifically, I will suggest that, when we hear sounds reverberate, we hear the unfilled space surrounding us. What is reverberation? In an enclosed space, sound waves can reach the ear either directly, emanating straight from the vibrating object, or indirectly, having first been reflected from objects in the vicinity such as the walls, ceiling, and floor. Direct waves reach the ears first, followed shortly afterwards by the reflected ones. As the object continues to vibrate, a mixture of direct and indirect waves hit the ears. When it stops, and the direct waves cease, reflected waves will continue to arrive for a few brief moments until they are absorbed by the surrounding surfaces. Waves are reflected differently, depending on the properties of the room in which they are enclosed. One variable is the size of the room: the delay between the arrival of the direct sound waves and the first reflected ones—which are referred to as 'early reflections'—is determined by how far away the walls are from the listener and the sound source. Sound waves will have further to travel if the listener and source are located at the centre of a large room than of a small one. Another is the properties of the enclosing surfaces themselves: materials such as stone or concrete are more reflective than wood or plaster, leading to more reflections, and a longer lasting reverberant field; bumpier surfaces cause waves to be reflected more diffusely than flat ones. All of these conditions can affect the character of auditory experience: hearing someone clap their hands in a stone cathedral is quite different from hearing them clap their hands in a tiled bathroom.[^2] Reflected sound waves are commonly thought of in terms of hearing echoes. However, I suggest that there is an interesting and important distinction to be made between hearing a sound reverberate and hearing a sound echo. If reflected sound waves reach the ears a significant amount of time after the direct ones,[^3] then a subject will have the impression of an initial sound being followed by a distinct, qualitatively identical (or almost identical), second sound, an extra chime or handclap following the first. This is an echo experience. When reflected sound waves reach the ear more quickly,[^4] and bounce off surfaces multiple times so as to blend and overlap with one another, this also has an impact on auditory experience: things sound different, but we do not have an impression of a second distinct sound. This is an experience of reverberation. A sound can be recognised from hearing only its echo, but not from only hearing its reverberation. These phenomenological differences can be seen in two examples. In the opening two bars of Be My Baby by The Ronettes, the drums can be heard to reverberate conspicuously: the snare drum played on the fourth beat of bar in particular has a distinct 'splashing' sound.[^5] This can be compared to Mystery Train by Elvis Presley, in which the guitar is played through a 'Slapback Delay' effect that artificially produces brief but distinct echoes. Indeed, the distinction between reverberation and echoes can be found in the setup of recording studios: rather than having a single effects unit that simulates reflected sound waves, audio engineers add artificial reverberation to a sound using one type of processor and add artificial echoes with another. The differences and similarities between echoes and reverberation will be discussed further in section 3, but it is interesting to note that, while contemporary theories of sound and audition have considered the ontology and perception of echoes [Nudds 2001: 221–2; O'Callaghan 2007: chs 8, 9], very little has been written about reverberation despite the fact that in everyday life experiences of distinct echoes are rare, whereas reverberation is commonplace.[^6] Every enclosed space[^7] allows for the reflection of sound waves, and thus every sound that we hear in an enclosed space will reverberate to some degree. Although it is most apparent in large rooms, such as churches or halls, even reflections generated by smaller rooms will have an impact on the character of auditory experience. Sounds without reverberation sound 'dry' and unnatural: a sound engineer would never record an electric guitar straight into a mixing desk without also adding some artificial reverberation; a film producer who needs to redub the dialogue of a scene in post-production must process speech recorded in a studio, so that it sounds like it was uttered in the same space where the scene was originally shot. Before considering whether spaces are heard, it is important to note that reverberation clearly provides awareness of spaces. Consider the difference between hearing fingers being clicked in a cathedral and hearing them being clicked in a garden shed. The character of each experience differs, depending on the size of the room, and a straightforward way of describing this difference is to say that the former sounds like fingers being clicked in a much larger volume of space than the latter. This is a marked dissimilarity with touch: the experience you have when you place your fingers on the rim of the glass will provide no indication of whether you are doing so in a large space or a small one. However, this is not to say that auditory awareness of space is the same as that of vision, as we cannot tell from hearing the exact locations of spaces. We do not, for example, learn that there are three metres of space between the fingers being clicked and the closest wall. Rather, we learn that in our general vicinity there is a certain volume of space that is unoccupied. This lack of specificity should not count against the possibility that we hear space: there are clear cases of our hearing sounds as coming from indeterminate locations, such as the rumble of a plane's engines filling the cabin. In such cases there is no reason to doubt that the sound itself is heard, and so there is no reason to think that an indeterminate location should count against the claim that we can hear empty space. However, becoming aware of a volume of empty space through hearing is insufficient to show that we hear empty space. It could be that Nudds's claims about audition and space are incorrect and that empty space can in fact be auditorily represented. On the other hand, it could be that experience of reverberation provides only an indirect and non-perceptual awareness of the space around us. It is clear that not every perceptual experience that contributes towards the forming of a belief about something involves perceptual representation of that thing: seeing a plume of smoke leads to thoughts about fire, but we do not see fire by seeing smoke; nor do we hear it through hearing someone shout 'fire!' Instead, the beliefs we form in these cases seem better described in terms of making what are often unconscious inferences from individuals that are represented in perceptual content to individuals that are not. Unconscious inference provides a way in which Nudds's claim might be reconciled with awareness of space through reverberation: volumes of space are not heard themselves, but beliefs about them are deduced from something we do hear. That is, the size of the space, and therefore the length of delay between direct and indirect sound waves, must affect auditory content in some other way, along with inferences about the size of the room made from these effects. I have suggested that there are three types of thing that are uncontroversially audible: sounds, echoes, and properties of sounds. It could be, therefore, that reflected sound waves lead to the representation of a second sound, or an echo, or that they affect which properties are attributed to a sound, and that one of these is the perceptual basis for inferences about volumes of space. In the following three sections, I will argue that experiences of reverberation cannot be reduced to the representation of any of these three things, and, lacking other plausible candidates, we should therefore allow that empty space is represented in auditory content. In the final section, I will suggest two ways in which empty space might be represented. ## 2. Reverberation as a Distinct Sound At first, it might seem that there is an obvious way to support Nudds's position: we perceive the sound of the hand clap, followed very shortly after by the sound of the reverberation of the hand clap. We have seen that experiences of reverberation are brought about by reflected sound waves reaching the ear immediately after direct sound waves, and one possibility is that the perception of one sound is caused by the direct sound waves and the perception of another by the reflected ones. We saw also that the distance of the walls from the perceiver affects how long it takes the second set of sound waves to arrive. Therefore, beliefs about how large a room is could be (unconsciously) calculated from the length of the delay between the first sound and the second, similarly to judging how high a ball bounces from the length of delay between the sound of its first impact with the ground and the sound of its second. However, this approach faces a difficulty: if reflected sound waves lead to a second sound being represented, then, by all contemporary accounts of what sounds are, this would be a case of misperception. As mentioned in the previous section, there are differing views as to the sort of individuals represented in audition and so different views as to what sounds are. However, all accounts have one thing in common: sounds depend on vibrating objects for their existence. Sounds are sometimes identified as a type of event: Casati and Dokic say that sounds are the vibrations of objects [2009], whereas O'Callaghan argues that they are disturbances of the air surrounding the object caused by its vibrating [2007]. Nudds, on the other hand, identifies sounds as 'structures or patterns of frequency components … instantiated by sound waves, that would normally be interpreted by the auditory system as having been produced by a single source event' [2010: 290], and defines source events as when material objects 'are struck, tapped, scraped, broken, or otherwise caused to vibrate' [2009: 70, my emphasis]. Alternatively, on Kulvicki's 'stable disposition view' [2008], material objects themselves are represented by the auditory system, and sounds identified with objects' vibratory dispositions. If stimulated, an object vibrates in a characteristic manner, at characteristic frequencies, depending on its material composition. The stable disposition view holds that this tendency is the object's sound. On this view, when an object vibrates or disturbs the air around it, this is not so much an instantiation of an individual sound as the permanent sound of the object being temporarily revealed, similarly to how illumination can reveal objects' permanent colour properties. Reverberation experiences, however, are elicited by reflected sound waves hitting the ears, meaning that on any of these accounts there is in fact no second sound to hear. A wall reflecting sound waves does not itself vibrate and so does not disturb the air surrounding it: it therefore cannot be thought of as generating a new sound, on either event view.[^8] Similarly, if we adopt Nudds's view, a group of reflected sound waves cannot be considered a sound in its own right, because they have no corresponding vibratory source event. Finally, while the enclosing surfaces might be thought of as revealing something of their material composition when sound waves bounce off them—as mentioned, the way that surfaces reflect sound can reveal qualities of the material from which they are made, such as degree of reflectiveness, or of bumpiness—they cannot be thought of as revealing their stable vibratory dispositions and so cannot, on Kulvicki's view, be thought of as disclosing their sounds. If reflected sound waves do elicit a representation of a sound, it seems that we must classify this as an illusion: we hear a sound when there is not one. The problem with this is that experiences of sounds reverberating in enclosed spaces are extremely common, and theories of perception that postulate widespread illusions are generally considered unacceptable. To avoid this, it would need to be shown that the hearing of phantom reverberatory sounds is somehow a tolerable illusion. One way to do this would be to reduce experience of reverberation to experience of echoes: this will be the topic of the next section. Here, I shall focus on another possibility in which the illusion is explained as the result of differences between the auditory environment in which we now live and the one in which our species evolved. The function of our auditory system is to track material objects in our environment, and, when stimulated, a given object will tend to vibrate simultaneously at multiple, harmonically-related, frequencies. Our auditory system has therefore evolved to group together harmonically related frequency components of sound waves.[^9] However, a modern listener will be exposed to groups of sound waves that have been produced by various non-ecological means. Nudds gives the examples of listening to multiple sounds emanating from a single speaker, or of a single sound emanating from multiple speakers. These cases can be thought of as the auditory system making an error: treating 'components that were in fact produced by distinct sources as having been produced by a single source, or … components that were produced by the same source as having been produced by distinct sources' [2010: 289]. Hearing reverberation could be a similar type of misinterpretation: due to an out-of-date auditory system, components produced by one source are heard as coming from two. Two things count against this approach. First, unlike recorded sounds, reverberation can be found in the natural world as well as in man-made environments—for example, in valleys or caves—meaning that it would not have been entirely alien to our evolutionary ancestors, certainly not to the degree that recorded or electronic sounds were. Second, experiences of reverberation provide us with correct information about the world, and this makes it difficult to see how they can be considered illusions. Hearing two speakers as a single source, or one speaker as multiple sources, misinforms us about the layout of the world. In contrast, hearing reverberation allows for accurate and potentially useful knowledge about our environment. Given this informativeness, alongside the pervasiveness of experiences of reverberation, theories of auditory perception that reduce reverberation to this type of illusion are implausible. ## 3. Reverberation as Echoes A different approach is to say that hearing reverberation is hearing a sound echoing, and that beliefs about the size of a space are determined by the size of the delay between a sound and its echoes. One reason that reducing experiences of reverberation to experiences of echoes might be thought appealing is that, despite the phenomenological differences, the physical processes that underlie both are very similar. But how should the perception of echoes be understood? Experiences of echoes are often compared to experiences of the reflections of objects in mirrors [Casati and Dokic 2009; O'Callaghan 2007: ch. 9]. O'Callaghan writes [ibid: 128–9]: > Mirrors facilitate our seeing the very objects and events that occur in front of them, albeit with distortion of place and perhaps qualities, as with an aged or funhouse mirror. Likewise, reflecting surfaces allow us to hear the very sounds that occur in front of them, albeit with distortion of place and time. But just as there is no new object that you see when you look in a mirror, there is no new sound that you hear at a reflective surface. Thought of in this way, mirror reflections are a form of illusion. Seeing an object reflected in a mirror is—in some way—as if we are seeing a second object when there is not one. O'Callaghan argues that this type of illusion is tolerable because it is 'explicable' and 'predictable', given what we know about light, reflective surfaces, and the visual system [ibid.: 138]. Echoes, he suggests, can be understood in the same way: given what we know about the behaviour of sound waves and the auditory system, we hear there to be a distinct sound, coming from a distinct source, when in fact there is not one. In section 1, we saw that experiences of echoes are much less widespread than experiences of reverberation, so if echo experiences are illusions the need to make them tolerable is less pressing than in the case of reverberation. Nevertheless, this account of echoes as tolerable illusions is more convincing than the evolutionary strategy for reverberation outlined in the previous section: objecting to the idea that echoes are acceptable illusions would require either showing that echoes are significantly dissimilar to reflections in mirrors, or providing an account of experience that shows that mirror reflections are non-illusory. Neither seems easy to do. Moreover, the mirror analogy also suggests how echo experiences are sufficiently misinformative to count as illusions, but how at the same time they provide correct information about the layout of our immediate environment. Seeing an object and its reflection can be thought of as having the non-veridical perception that there are two objects at two locations when in fact there is only one, and echoes can be thought of as supplying a similar type of misinformation: to hear the sound of a firework explode in front of you, before hearing its echo behind you a moment later [ibid.: 110], is to have an illusory experience as of two separate sound sources, at two separate locations, when in fact there is only one. Despite the illusory nature of the experience, however, hearing an echo could be used to locate a reflective surface in quite a straightforward manner: through hearing the echo of a firework at a brick wall, we can learn approximately where that wall is located in relation to ourselves, just as seeing a reflection reveals the location of the mirror. Experiences of echoes, then, are both sufficiently explicable to count as tolerable and sufficiently misleading to count as illusory, while still allowing for knowledge of our environment. An account of indirect awareness of space through reverberation that reduces experiences of reverberation to experiences of echoes might retain these advantages. However, it is not clear that reducing reverberation experiences to echo experiences can explain the manner in which we become aware of empty space through audition. Specifically, if experiences of reverberation were experiences of echoes we would expect them to provide awareness of the locations of the surrounding reflective surfaces, just as they do in the firework example; but this is not the case. In order to account for an awareness of the volume of space around us, as opposed to the location of a single reflective surface, this type of process would have to be scaled up: we hear multiple echoes located at all of the reflective surfaces around us, and perceiving these echoes informs us of the locations of the surfaces, which allows the volume of the surrounding enclosed space to be calculated. Making a noise in a reverberant environment, however, provides us with an awareness of the size of the room without revealing the location of the surrounding surfaces: clapping one's hands in a cathedral will inform you that you are in a large space, but not whether you are located closer to the left-hand or right-hand wall, or whether the ceiling is lower at one end of the hall than at the other. If reverberation and echo experiences really are of the same type then we might expect a clearer awareness of where the reflective surface in the room are, and any account that tries to reduce one to the other is weakened if it cannot provide an explanation as to why this is not the case. Note that the problem here is not that reverberation provides an imprecise awareness of where walls and ceilings are—hearing an echo, or seeing a reflection, might provide only rough information as to the location of a reflective surface—but that we can become aware of a volume of space without any awareness whatsoever of the locations of reflective surfaces. This point is reinforced when we consider that it is still possible to hear reverberation, and to form beliefs about the size of a space, when listening in mono. If you compare the old mono recording of Be My Baby with the mono version of A Hard Day's Night by The Beatles, the former still sounds like it was recorded in a larger space than the latter was, despite the recordings not carrying any information as to the locations of the surfaces in relation to the performers. These difficulties are possibly not fatal for this strategy. We have seen that reverberation involves sound waves being reflected from surfaces multiple times in quick succession, causing them to blend together. It could be argued that this process somehow leads to information about location being lost but information about volume being retained. However, it is not obvious why this would happen, as can be seen if we consider a scaled-up version of the mirror analogy. If hearing a sound echo once is equivalent to looking at a mirror and seeing a single reflection of an object, then the equivalent in the case of reverberation would be one's standing in a room in which the floors, ceiling, and walls are mirrored. While this might be a disorienting experience (as standing in a fairground hall of mirrors can be), there is no reason to think that we would lose track of where the walls are while at the same time retaining awareness of the size of the room. Despite these problems, the intuition might remain that because reverberation experiences and echo experiences involve the same sort of physical process—reflected sound waves hitting the ear after direct ones—they are still essentially the same type of perceptual experience. In response to this, we can turn briefly to a psychoacoustic phenomenon known as the precedence effect, which shows that the speed at which reflected sound waves follow direct ones to the ear can lead to quite different types of perception. Experiments in this area often involve subjects standing in an anechoic room and being presented with pairs of identical sounds (such as clicks), one played slightly later than the other, on speakers at different locations. The idea is to simulate a simplified case of direct, and then indirect, sound waves hitting the ear. The most important result for our current concerns is that, when the interval between the presentation of the first and second sound is sufficiently short (between 5 ms–50 ms depending on the type of sound),[^10] the second sound is not discernible but is instead heard to be 'fused with the original or leading sound to yield the percept of a single source' [Litovsky et al. 1999: 1636].[^11] It is only above this threshold that the second click leads to the representation of a second individual—a sound or an echo. Moreover, in the cases where the two sounds are heard as fused, the perceived location of the source of the sound is largely determined by the location of the first sound, and the location of the second sound is suppressed [ibid.: 1634]. What this suggests is that, when the interval between direct and reflected sound waves is short enough to elicit experiences of reverberation, the reflected sound waves are not heard as separate echoes with their own location, but instead only influence how the original sound is perceived. I take this to be sufficient to show that reverberation and echo experiences are of fundamentally different kinds. ## 4. Reverberation as a Property of a Sound The precedence effect suggests a third way in which reverberation might affect what we auditorily represent and therefore a third way in which empty space might be derived indirectly from auditory experience. Rather than reflected sound waves leading to the representation of a separate individual—a sound or an echo—they instead alter the auditory properties that the original sound is perceived as possessing. In particular, we might think that they alter the perceived timbre of the sound. We saw above that reflections under a certain temporal threshold are heard as 'fused' with the initial sound, but this should not be taken to mean that they are not detected at all [Litovsky et al.: 1999]: > We emphasize the fact that echo threshold is not the threshold of detectability of the lag; lead-lag sounds and lead-only sounds can be distinguished easily based on overall sound quality … listeners are usually able to distinguish between trials in which a lag is present and trials in which the lead is presented alone. If reflected sound waves affect 'overall sound quality', this could serve as a basis for inferences about empty space. An analogous case would be the way that old-fashioned transistor radios add a particular tinny timbre to any and all sounds that are played through them. If we hear the sound of a guitar with this characteristic tinniness, we can deduce that what we are hearing is not a real life guitar but instead one played through a transistor radio. Part of what gives these radios their character is the small size of their speaker: the smaller the size, the tinnier the sound. We can imagine someone who has listened to a great many different types of radio being able to work out the size of the speaker by the degree of tinniness that the sounds playing are heard to have. Similarly, sounds in reverberant environments could be heard to have a certain 'reverberatory timbre', and, from this variation in quality, beliefs about the size of the room can be formed. Unlike the previous two strategies, this account does not try to explain experience of reverberation in terms of the perception of a separate particular, and it therefore avoids some of the difficulties that they have: there is no need to be concerned about whether the perception of reverberation is illusory or not, awareness of the volume of space need not depend on awareness of the locations of the enclosing surfaces, and it fits better with research on the precedence effect. Despite these advantages, it is not plausible to think that reverberation can be reduced to only variations in timbre, as, even if reverberant environments affect the timbre that sounds are perceived as having, this is not the only way they affect auditory experience. Compare the experience of a cello playing a sustained note in a small but not particularly reverberant room, with the same note being played in a large hall in which the reverberation is more apparent. Arguably, the timbre that the sound is heard as having is altered, depending on the room, but it is wrong to say that this is the only difference: when a single sound discernibly reverberates, our experience represents multiple elements at multiple locations. That is, we would hear the cello to be in a particular region of our egocentric space—in front and to the left, for example—but the reverberation of the sound is not heard as coming from that same location. It could instead be described as coming from all around, or as having no location. The two strategies outlined in sections 2 and 3 can accommodate this phenomenon quite easily. Because both hold the perception of reverberation to be the perception of a particular that is distinct from the original sound, hearing a sound reverberate can be thought of in terms of hearing a sound to be at one location and hearing the second sound or echo that follows it to be at another. A strategy that tries to reduce reverberation to timbre, on the other hand, explains the apparent locatedness of reverberation much less well, because it posits only one particular—the sound itself—that can only be heard as having one location. The only way to maintain the claim that just one sound is heard would be to say that single sounds can be heard as having spatial parts that can be heard as located in relation to one another. In this way, we could say that we hear the 'cello' part of the sound to be located in one place, and the reverberatory part to be located at another. This is, however, a high price to pay, as it is difficult to make sense of a single sound having different spatial parts. As Nudds says, '[h]earing simultaneous sounds as having distinct spatial properties is sufficient to hear them as distinct sounds, so we cannot simultaneously hear distinct parts of a single sound as standing in spatial relation to one another' [2009: 81]. This problem is compounded if we take reverberation to be heard as having an indeterminate location, as being located all around us, because it is even more difficult to think of a single sound as having both determinately and indeterminately located spatial parts. ## 5. Reverberation as Hearing Spaces We have seen that auditory awareness of spaces cannot be explained in terms of the perception of extra sounds, echoes, or changes in timbre. We should therefore take seriously the idea that space is heard, not inferred.[^12] In this final section, I shall outline two ways in which empty spaces could be auditorily represented, and I will deal with a potential objection. It was mentioned in section 1 that, as well as sounds, echoes, and auditory properties, sound sources are considered by some to be represented in auditory content. I suggest that the different models of auditory source perception that have been proposed can also serve as models for auditory spatial perception. Hearing often leads to beliefs about sound sources, such as their location, size, and material composition. For example, we might learn from hearing that a large metal bell has been struck to the left of us. Very often, the formation of these beliefs is effortless and immediate, without the need for conscious inference. As with auditory awareness of space, this could be a case of unconscious inference, with a sound's timbre leading us to judge that an object of a particular type has been stimulated. However, many contemporary accounts of audition have it that we can in fact hear the sources themselves. There are two different ways in which sources could be perceived, both of which can be adapted to explain space perception. The first, we might think of as an 'immediate' approach to source perception. Recall that, according to Kulvicki's stable disposition view, sounds are considered stable dispositional properties of objects and that the auditory system represents material objects themselves, not disturbances of the air that they have produced: in the same way that we see an illuminated object to have a colour, we hear a stimulated one to have a sound. This might be thought of as a quite revisionary view of audition, as it leaves behind the intuitions that what we hear are sounds and that sounds are individuals distinct from material objects. A less revisionary approach to this issue is to maintain that we hear sounds, and that sounds are distinct individuals, but that we perceive sources mediately, through hearing sounds. O'Callaghan, for example, argues that we hear sounds as 'constituent parts of everyday audible events', in the same way that we see the facing surfaces of objects as parts of a larger individual [2011: 19]. It is important to realise that mediately perceiving something is different to making inferences about that thing from what is perceived. A mediate theorist of vision does not say that we see the front of the door and we infer the presence of the back, but that in seeing the front we see the whole door: the representation of the facing surfaces leads to the rest of the object being included in perceptual content as well. Similarly, those who endorse mediate perception in audition have it that the auditory representation of a sound leads to the auditory representation of its source. I will not take a stand here on mediate perception in vision, or on whether sources can be heard, or on how they are heard. However, immediate and mediate approaches to source perception suggest two ways in which empty space might be auditorily represented. First, we might think that when reflected sound waves hit our ears they lead to representations of space *simpliciter*, similarly to Kulvicki's approach to object hearing. We hear directly the space surrounding us, unmediated by the perception of anything else. Like the stable disposition view, this could be thought fairly revisionary as it entails that sounds, understood as distinct individuals, are not necessary to explain the hearing of spaces. A second, less revisionary, approach would be to say that volumes of space are heard through hearing sounds in much the same way that mediate theories hold sources to be heard. Those who maintain that we hear sounds as parts of events should consider the possibility that we hear sounds as parts of events taking place in a particular volume of space. Regardless of whether an immediate or mediate approach is adopted, auditory representation of spaces avoids the problems faced by the timbre view examined in the previous section. We saw there that reverberation could not be reduced to changes in the timbre of sounds, because reverberant environments affect spatial aspects of auditory experience as well: sounds are normally heard as occurring in a particular region of space, but reverberation is heard as coming from all around us. As with the distinct sound and echo views, the approach involves multiple individuals being represented simultaneously and so can explain the spatial aspects of reverberation experiences in a similar manner: we hear the sound (or perhaps the vibrating object) to be at one location, and the volume of space around us to be at another, less determinate, location. It could be objected here that, even if we allow that that sound waves can lead to the representation of individuals other than sounds, echoes, and sources, we should not assume that empty space is the extra individual. It could be that something else is represented and then inferences about volumes of space are made from that. Only Nudds considers the possibility that auditory representation might extend beyond sounds and sources, and in doing so he provides two alternatives for what might be perceived instead of space [2010: 286]: > This pressure wave interacts with other objects in the environment and is differentially reflected by surrounding surfaces. These interactions alter the pressure wave in characteristic ways. The structure of the vibration that is transmitted to the ears by the pressure wave therefore carries or embodies information, not only about the objects and events that produced the pressure wave, but also about the environment within which the events that produced it occurred. Because sound waves interact both with other objects and surrounding surfaces, it could be that these—not space—are auditorily represented. Indeed, the latter possibility in particular suggests another way in which awareness of space could be arrived at indirectly: we do not hear space; we hear the walls themselves and we infer space from them. However, when we consider the poverty of information that audition supplies about these things, it is implausible to think that either—the walls or the space—is auditorily represented. As mentioned, one reason to think that we do hear source objects is that, when the sound waves that they produce hit our ears, we come to know about some of their properties instantly and without conscious inference. But we do not come to know about non-source objects in the vicinity with anything like the same richness: a chime in a church will not inform you of how many seats there are in the hall, or where they are placed, or whether they are wooden pews or metal stools.[^13] As we saw in section 3, reverberation provides no knowledge of where the surrounding walls are, and although the material constitution of surfaces can have an impact on the character of auditory experience it is nevertheless minimal when compared to what we learn about source objects. Moreover, as we also saw in section 3, if we are unable to know the locations of surrounding walls we are unable to explain how we become aware of the space that they enclose. For these reasons, I take empty spaces themselves to be more plausible candidates for auditory representation. Because space is an absence of objects, the types of property that could be ascribed to it are—obviously—minimal: it does not, for example, make sense to talk of the material constitution of space, but what can be attributed is an amount or a size. It is perfectly natural to ask, 'How much space is there in the hall?' or 'How big is the hall?', and this, I argue, is exactly the type of property that is attributed to represented spaces. ## 5. Conclusion Given that experiences of sounds reverberating cannot be reduced to experiences of hearing extra sounds, echoes, or properties of sounds, there is good reason to think that empty space can be heard. This contradicts Nudds's claim that auditory experience does not represent empty places, and that our auditory experience of space is 'exhausted' by our awareness of spatial relations between ourselves and sound sources. Instead, reverberation leads to the auditory representation of empty places, albeit ones with less determinate locations than those represented in vision, and that we hear sounds to stand in spatial relations to these. Moreover, the comparison with source perception shows two ways in which space could be auditorily represented: those who accept that sources are heard should consider the possibility that spaces are heard in the same way.[^14] [^15] ## References Bregman, A.S. 1990. *Auditory Scene Analysis: The Perceptual Organization of Sound*, Cambridge, MA: The MIT Press. Casati, R. and J. Dokic 2009. Some Varieties of Spatial Hearing, in *Sounds and Perception: New Philosophical Essays*, ed. M. Nudds and C. O'Callaghan, Oxford: Oxford University Press: 97–110. Kulvicki, J. 2008. The Nature of Noise, *Philosophers' Imprint* 8/11: 1–16. Leddington, J. 2014. What We Hear, in *Consciousness Inside and Out: Phenomenology, Neuroscience, and the Nature of Experience*, ed. R. Brown, Dordrecht: Springer: 321–34. Litovsky, R.Y., H.S. Colburn, W.A. Yost, and S.J. Guzman 1999. The Precedence Effect, *The Journal of the Acoustical Society of America* 106/4: 1633–54. Lochner, J.P.A. and J. F. Burger 1958. The Subjective Masking of Short Time Delayed Echoes by their Primary Sounds and their Contribution to the Intelligibility of Speech. *Acta Acustica united with Acustica* 8/1: 1–10. Martin, M. 1992. Sight and Touch, in *The Contents of Experience: Essays on Perception*, ed. T. Crane, Cambridge: Cambridge University Press: 196–215. Nudds, M. 2001. Experiencing the Production of Sounds, *European Journal of Philosophy* 9/2: 210–29. Nudds, M. 2009. Sounds and Space, in *Sounds and Perception: New Philosophical Essays*, ed. Matthew Nudds and Casey O'Callaghan, Oxford: Oxford University Press: 69–96. Nudds, M. 2010. What Sounds Are, in *Oxford Studies in Metaphysics*, Vol. 5, ed. Dean W. Zimmerman, Oxford: Oxford University Press: 279–302. O'Callaghan, C. 2007. *Sounds: A Philosophical Theory*, Oxford: Oxford University Press. O'Callaghan, C. 2011. Hearing Properties, Effects, or Parts? *Proceedings of the Aristotelian Society* 111/3.3: 375–405. Phillips, I. 2013. Hearing and Hallucinating Silence, in *Hallucination: Philosophy and Psychology*, ed. F. Macpherson and D. Platchias, Cambridge, MA: The MIT Press: 333–60. Rossing, T.D., F.R. Moore, and P.A. Wheeler 2002. *The Science of Sound*, 3rd edn, San Francisco: Addison Wesley. Sorensen, R. 2008: *Seeing Dark Things: The Philosophy of Shadows*, Oxford: Oxford University Press. Yang, X. and D.W. Grantham 1997. Echo Suppression and Discrimination Suppression Aspects of the Precedence Effect, *Attention, Perception & Psychophysics* 59/7: 1108–17. [^1]: Sounds and their properties, echoes, and sources are widely accepted, among philosophers interested in auditory perception, as plausible candidates for what we hear. Sorensen [2008: ch. 14] and Phillips [2013] have also suggested that we hear silence. [^2]: For a detailed summary of the physics of reverberation see Rossing et al. [2002: ch. 23] [^3]: O'Callaghan suggests that a gap of two seconds is required for the 'distinctive multisound echo experience' [2007: 128]. [^4]: Usually within 50–80 ms of the direct sound waves [Rossing et al.: 526]. [^5]: Although the room in which Phil Spector's 'Wall of Sound' songs were recorded was named an 'echo chamber', this is a misnomer—it was actually a room that produced a long, prominent, reverberant field. [^6]: O'Callaghan [2007: 127] says: "When secondary sound waves arrive at the ears ... [b]etween roughly fifty milliseconds and two seconds, the result is an experience as of a primary sound and a distinct but somehow causally related secondary sound or echo. When the arrival delay is greater than about two seconds, the experience is as of two separate and entirely unrelated sounds." He does not elaborate, however, on what it means for a sound or echo to be heard as 'somehow causally related' to the initial sound. [^7]: Anechoic chambers, which have walls, floors, and ceiling designed specifically to absorb rather than reflect sound waves and thus dampen reverberation, are an exception. [^8]: O'Callaghan is quite explicit that 'passing along or redirecting a pre-existing sound wave is not sufficient for producing a sound' [2007: 127] [^9]: Nudds [2009: 73–4] summarises how and why frequency components of sound waves are grouped together. See also Bregman [1990] [^10]: 5–10 ms for clicks [Yang and Grantham 1997]; 40–50ms for speech [Lochner and Burger 1958]. [^11]: Litovsky et al. [1999] reviews psychological and physiological work on the precedence effect. [^12]: This claim can also be put in terms of Bregman's Auditory Scene Analysis. A perceiver is understood as performing 'scene-analysis' when their auditory system determines 'which parts of the sensory stimulation [sound waves hitting the ears] are telling us about the same environmental object or event' [1990: 3]. My claim can be taken as a suggestion that the auditory system also determines which aspects of sensory stimulation are telling us about the surrounding empty space. [^13]: It is possible that humans who use echolocation, such as Daniel Kish, are an exception to this as they are able to determine the locations of non-vibrating objects. This opens up the possibility that they auditorily represent non-vibrating objects and perhaps the determinately located spaces between them. Due to space constraints, I am unable to explore this possibility here. [^14]: I am grateful for the helpful comments on this paper given to me by my research group, as well as the audiences of the MindGrad '14 Graduate Conference in Philosophy of Mind at University of Warwick, UK and XVII Taller d'Investigació en Filosofia at Universidad autónoma de Barcelona, Spain. Particular thanks to Gloria Andrada, Dan Cavedon-Taylor, Peter Fazekas, Kris Goffin, Laura Gow, Grace Helton, Neil van Leeuwen, Bence Nanay, Matthew Nudds, Maarten Steenhagen, and Margot Strohmingeris. [^15]: This work was supported by the FWO Odysseus grant G.0020.12N.