In Book I Part II of A Treatise of Human Nature, Hume defends what has seemed to many scholars to be a puzzling thesis about time: unchanging objects do not exist in time. Hume acknowledges that his view conflicts with “the common opinion of philosophers as well as of the vulgar”; and he is certainly correct: we simply do not believe that objects come in and out of time as they change and stop changing (T 1.2.3.11).1 We assume that the concept of time applies to unchanging objects. Yet Hume refers to this assumption as a “falsehood,” (T 1.2.3.11) and he returns to his confounding “no change-no time” thesis at various points in the Treatise, only to further insist on it (e.g. T 1.2.5.29, T 1.4.2.29).
Hume, like Locke, Malebranche, and Berkeley before him, identifies changing objects as the source of the idea of time;2 unchanging objects cannot give rise to the idea of time. Hume argues that because an idea can only be applied to the objects that can produce the idea, it “inevitably follows,” as he puts it, that “the idea of duration … can never in any propriety or exactness be apply’d to … anything unchangeable” (T 1.2.3.11). Unchanging objects cannot be said to endure.
Hume identifies a “fiction” in the workings of the mind that leads us to suppose that the idea of time can be applied to unchanging objects; we shall refer to this fiction as the “endurance fiction.” Hume identifies other fictions at the foundation of several fundamental beliefs, such as those concerning the identity of objects. In general, wherever we apply an idea to an object that cannot cause the idea, there is a fiction lurking in the mind.
The background of Hume’s discussion of time, a discussion that relies heavily on his prior treatment of space, is the Newtonian conception of space and time. In his treatment of space, Hume explicitly criticizes the “Newtonian philosophy” (T 1.2.5.26n12(App.)). Just as he rejects the Newtonian concept of absolute space, it is evident that his arguments concerning the idea of time have the corresponding concept of absolute time as their target.3 Scholars have also noted that, given the theological uses that philosophers like More and Clarke confer on these concepts, Hume likely perceives them as pernicious even beyond the context of philosophy and science (see, e.g., Russell 2008, ch.9; Cottrell 2019a, 83–85). Indeed, in terms of its theological implications, Hume’s extensive critique of the notions of empty spaces and changeless times seems of a piece with his extensive critique in the same part of the Treatise of another “priestly dogma” about space and time—the doctrine of infinite divisibility (E 12.18).4
This philosophical background, however, does not provide a solution to the puzzle of why Hume would deny that unchanging objects endure. Moreover, the account Hume presents of the endurance fiction, which is supposed to explain the false belief that unchanging objects endure, is itself rather perplexing. The account he puts forward of “those appearances, which make us fancy we have that idea [of time without change]” (T 1.2.5.29) appeals rather vaguely to ordinary experiences of “stedfast objects.” It is precisely in his descriptions of these experiences that, in the eyes of many scholars, Hume’s account of the endurance fiction seems to unravel, and with it his argument for the conclusion that the ordinary belief is false. Scholars allege that, when we inspect the experiences in question, it becomes apparent that Hume could have criticized the Newtonian concept of absolute time without impugning the ordinary belief that unchanging objects endure. In fact, the alternative thesis that unchanging objects do endure seems to fit effortlessly with Hume’s descriptions of the experiences. If so, it seems Hume goes out of his way to defend a rather implausible thesis, and for no apparent good reason. As Lorne Falkenstein points out, it is as though Hume is “contorting” a psychological explanation to defend a view that is not even prima facie plausible to begin with (2017, 48).
Our aim in this paper is to address these interpretive puzzles. To this end, we contextualize Hume’s position with more precision. We argue that Hume’s target is not simply the general concept of absolute time but, more specifically, Locke’s approach to this concept in An Essay Concerning Human Understanding. Scholars have taken little notice of Locke in relation to Hume’s view on unchanging objects, but we think Hume’s arguments in fact respond to Locke’s specific claims about the application of the idea of time. We show that, when set in this context, Hume’s position on unchanging objects is much more cogent than the scholarship suggests.
First, we explain Hume’s theory of time and his account of the endurance fiction (section 1). We then review the main interpretive challenges in the literature (section 2). Finally, we show that these challenges can be addressed by reading Hume’s view on unchanging objects as a response to Locke (section 3).
1. Time: The impression, the idea, and the fiction
Hume argues that we acquire our idea of time from the experience of “a succession of changeable objects” (T 1.2.3.7, T 1.2.3.8, T 1.2.3.9, T 1.2.3.11).5 More precisely, Hume argues that our idea of time derives or is produced by the succession of perceptions in our minds.6 There is the succession of sense impressions as we hear a train passing. There is the succession of impressions of reflection as we experience changes in emotional states, say from fear to anger; and finally, there is the succession of ideas as we remember waking up in the morning. Hume emphasizes that these successions are made up of finite parts, and he devotes the first two sections of Book I Part II of the Treatise to arguing that space and time cannot be infinitely divisible.
For Hume, an impression of time just is a succession of impressions. When we are listening to “five notes played on a flute,” the impression of time is not anything over and above the successive sounds of the notes (T 1.2.3.10). In the same way that there is no impression of existence without it being the impression of an existing something or an existing object (T 1.2.6), there is no impression of time without it being a succession of perceptions.
In advancing this argument, Hume distinguishes between perceptions and external objects to emphasize that, provided that these need not always correspond to one another, it is the succession of perceptions that is responsible for the idea of time:
It has been remark’d by a great philosopher, that our perceptions have certain bounds in this particular, which are fix’d by the original nature and constitution of the mind, and beyond which no influence of external objects on the senses is ever able to hasten or retard our thought. If you wheel about a burning coal with rapidity, it will present to the senses an image of a circle of fire; nor will there seem to be any interval of time betwixt its revolutions; meerly because ‘tis impossible for our perceptions to succeed each other with the same rapidity, that motion may be communicated to external objects. Wherever we have no successive perceptions, we have no notion of time, even tho’ there be a real succession in the objects. From these phænomena, as well as from many others, we may conclude, that time cannot make its appearance to the mind, either alone, or attended with a steady unchangeable object, but is always discover’d by some perceivable succession of changeable objects. (T 1.2.3.7; Hume’s emphasis).
The “great philosopher” is Locke. Locke distinguishes between a rapidly spinning object and the appearance of that object as a static circle to criticize the Aristotelian view that motion per se (that is, independently of the succession of ideas in the mind) is the source of the idea of time ( Essay II.xiv.6–10). Hume deploys the example similarly, to emphasize that it is not the fact of changes in the external world that generates our idea of time, but rather, it is the perception of those changes that produces the idea of time. We think Hume’s observation that there can be a lack of correspondence between sense impressions and external objects is important in this passage, and we return to it later. 7
Having given his account of the impression of time, Hume turns to the question of whether time can be conceived “without our conceiving any succession of objects, and whether it can alone form a distinct idea in the imagination” (T 1.2.3.9). And this, for Hume, is a question about separability. Hume argues that to know whether any objects, “which are join’d in impression, be separable in idea, we need only consider, if they be different from each other; in which case, ‘tis plain they may be conceiv’d apart.” Hume maintains that because our idea of time is the idea of succession (they are indistinguishable), we can’t separate them (T 1.2.3.10).
Time and succession for Hume are not simply “join’d in impression” in the way the shape and color of an apple can be; rather, they are altogether indistinguishable from one another. The impression of time when listening to “five notes play’d on a flute” just is the succession of the sounds, not anything over and above (T 1.2.3.10). Given that our idea of time derives from experience, our idea of time is the idea of succession.
Having established the position that our idea of time is inseparable from the idea of succession, Hume takes up the most controversial aspect of his treatment of time. “I know,” he remarks, that “there are some who pretend, that the idea of duration is applicable in a proper sense to objects, which are perfectly unchangeable; and this I take to be the common opinion of philosophers as well as of the vulgar” (T 1.2.3.11).
Hume portrays this common view of the vulgar and the philosophers as false (he refers to “its falshood” T 1.2.3.11). It is false because the idea of time “can never be convey’d to the mind by any thing stedfast and unchangeable.” And for Hume, “it inevitably follows from thence, that since the idea of duration cannot be deriv’d from such an object, it can never in any propriety or exactness be apply’d to it, nor can any thing unchangeable be ever said to have duration” (T 1.2.3.11). We shall call this argument the “propriety argument.”
The propriety argument relies crucially on a premise about the “proper” or “exact” application of an idea to an object, namely, that an idea applies only to the objects from which the idea derives. As Hume puts it, an idea “can never without a fiction represent or be apply’d” to any other object (T 1.2.3.11). A “fiction” for Hume is the act of applying an idea to the kind of object that cannot give rise to the idea.8 In the endurance fiction, we apply the idea of time to an unchanging object — an object that cannot give rise to the idea of time. Similarly, we engage in a fiction when we apply the idea of space to something invisible and intangible (T 1.2.5.5–14), or the idea of simplicity to a compound object (T 1.4.3.5). In all these cases, the object in question cannot generate the idea the mind applies to it.
Hume’s demand that we limit our application of ideas to the objects that produce them might seem unreasonably strict. Yet, it is a demand we all commonly agree on. Consider the idea of shape or the idea of color. We can’t acquire the idea of shape from the experience of pain or the idea of color from sound. Thus, it is improper to ascribe, say, the idea of trapezoidal shape to a stomach pain, or the idea of red to the sound of birds chirping.
Hume follows the text at T 1.2.3.11, about the proper application of ideas, with a promissory note to explain how the mind comes to engage in the endurance fiction, that is, how “we apply the idea of time, even to what is unchangeable” (T 1.2.3.11). The explanation appears in the following polemical paragraph:
But tho’ it be impossible to shew the impression, from which the idea of time without a changeable existence is deriv’d; yet we can easily point out those appearances, which make us fancy we have that idea. For we may observe, that there is a continual succession of perceptions in our mind; so that the idea of time being for ever present with us; when we consider a stedfast object at five-a-clock, and regard the same at six; we are apt to apply to it that idea in the same manner as if every moment were distinguish’d by a different position, or an alteration of the object. The first and second appearances of the object, being compar’d with the succession of our perceptions, seem equally remov’d as if the object had really chang’d. To which we may add, what experience shews us, that the object was susceptible of such a number of changes betwixt these appearances; as also that the unchangeable or rather fictitious duration has the same effect upon every quality, by encreasing or diminishing it, as that succession, which is obvious to the senses. From these three relations we are apt to confound our ideas, and imagine we can form the idea of a time and duration, without any change or succession. (T 1.2.5.29)
This is Hume’s account of how we “imagine we can form the idea of a time and duration, without any change or succession.” And his strategy is to explain “those appearances, which make us fancy we have that idea [of time without change].”
By way of illustrating the appearances in question, Hume introduces the example of the experience of “considering” a “stedfast object” from five o’clock to six o’clock. But how exactly is the experience of a stedfast object constituted? The problem is that Hume leaves this experience significantly under-described. From Hume’s depiction, we cannot even tell whether the experience is one where we contemplate the object continuously from five to six o’clock (as many scholars suppose), or whether our contemplation is interrupted by other sense impressions, for instance, if we go for a run between five and six o’clock.
We can gain some insight into Hume’s example of the experience of a steadfast object when we turn our attention to the examples of experiences he presents in the case of the fiction of the vacuum.9 They are all structurally equivalent. In the case of the fiction of the vacuum, he offers two examples. The first involves “two visible objects appearing in the midst of utter darkness” (T 1.2.5.15). Hume takes pains to set up the example in a way that stresses that there is nothing visible between the two objects: “we must suppose, that amidst an entire darkness, there are luminous bodies presented to us, whose light discovers only these bodies themselves, without giving us any impression of the surrounding objects” (T 1.2.5.8); the darkness that surrounds the objects is “a perfect negation of light, and of every colour’d or visible object” (T 1.2.5.11). The second example is “that of a man, who feeling something tangible, leaves it, and after a motion, of which he is sensible, perceives another tangible object” (T 1.2.5.13). Hume, again, constructs the example to highlight that there is nothing tangible between the tangible objects: an “utter removal” of every thing tangible (T 1.2.3.7).10 In both cases, there is an absence of certain objects — namely, visible and tangible objects, the types of objects that give rise to the idea of space — bookended by certain sensations. In the example of the steadfast object, Hume’s aim is also to depict an absence bookended by two appearances — namely, the absence of changing or different objects, the type of objects that can give rise to the idea of time.
Hume appears to be setting up the examples in such a way as to exclude the possibility of the relevant impressions: visual, tactile, and succeeding impressions. Of course, the relevant impressions can be excluded in different ways. As we saw earlier, objects spinning rapidly appear to be at rest; therefore, not only impressions of objects at rest but also impressions of objects spinning rapidly could be employed to illustrate the absence of changing impressions between the two appearances.
As we saw in the five-o’clock-to-six-o’clock passage, Hume outlines “three relations” between the steadfast object and “that succession, which is obvious to the senses.” To understand these relations better, we must once again consider the three relations as they appear in the case of the vacuum: The relations that hold between “that distance, which is not filled with any colour’d or solid object” (the gap between the visible and tangible objects in the examples just discussed) and “that distance, which conveys the idea of extension” (T 1.2.5.18).
First, Hume remarks, the objects separated by an absence of visible and tangible objects “affect the senses in the same manner” as objects separated by visible and tangible objects. The light rays reflected by the objects hit the eyes at the same angle regardless of whether there are other visible objects between them. Similarly, the sensation of motion in moving one’s hand from one object to the other is the same regardless of whether there are other tangible objects in between (T 1.2.5.15). With respect to the steadfast object, Hume likewise suggests that the two appearances affect the senses in the same manner regardless of whether there are changeable objects between them: “the first and second appearances of the object, being compar’d with the succession of our perceptions, seem equally remov’d as if the object had really chang’d” (T 1.2.5.29). The mind experiences the two appearances as “removed” whether or not there are changes between them.
Second, the objects separated by an absence of visible and tangible objects “are capable of receiving the same extent” as objects separated by visible and tangible objects (T 1.2.5.16). What Hume means is that the gap between the two objects could be filled with visible and tangible objects without any change to the objects themselves or their effects on the senses: “an invisible and intangible distance may be converted into a visible and tangible one, without any change on the distant objects” (T 1.2.5.16). Correspondingly, the gap between the two appearances of the steadfast object could be filled with changes without any change to the appearances themselves: “experience shews us, that the object was susceptible of such a number of changes betwixt these appearances” (T 1.2.5.29).
Third, the objects separated by an absence of visible and tangible objects have the same effects with respect to qualities that are sensitive to distance as objects separated by visible and tangible objects. Hume mentions several qualities of objects that “diminish in proportion to distance,” such as heat, cold, light, and attraction (T 1.2.5.17). Thus, for instance, the farther two objects are from each other, the weaker will be the attraction felt between them, other things being equal. Hume’s point is that these qualities would still diminish in the case of the objects separated by an absence. The qualities are still sensitive to the gap between the objects, and diminish accordingly, regardless of whether that gap is filled with objects or not. Correspondingly, in the temporal case, “the unchangeable or rather fictitious duration has the same effect upon every quality, by encreasing or diminishing it, as that succession, which is obvious to the senses” (T 1.2.5.29). Consider a time-sensitive quality, for instance, speed. The longer the time it takes for an object to move from one point to another, the lesser its speed, other things, such as the distance between the points, being equal. Now suppose that one moves across the room in between the first and second appearances of the stedfast object. Hume’s point is that one’s speed would still diminish in proportion to the gap between the two appearances: the greater the gap, the lesser one’s speed. As in the case of the vacuum, where a non-spatial distance takes on the same effects as a spatial distance, in the case of the stedfast object something that is not a succession and thus not temporal takes on the same effects as a genuine succession or instance of time.
Having presented these three relations, Hume appeals to two closely related psychological principles:
We may establish it as a general maxim in this science of human nature, that wherever there is a close relation betwixt two ideas, the mind is very apt to mistake them, and in all its discourses and reasonings to use the one for the other (T 1.2.5.19).
Resembling ideas are not only related together, but the actions of the mind, which we employ in considering them, are so little different, that we are not able to distinguish them … we may in general observe, that wherever the actions of the mind in forming any two ideas are the same or resembling, we are very apt to confound these ideas, and take the one for the other. (T 1.2.5.21)
The first principle states a sufficient condition for the mind mistaking two ideas, namely, that the ideas be closely related. The second principle states another sufficient condition: that the ideas be formed by the same or a similar action of the mind (and Hume suggests in the same paragraph that ideas tend to be thus formed when they are closely related).
Thus, the three relations noted earlier, together with the above-mentioned psychological principles, conspire to render the mind “very apt” to mistake the idea of an absence of visible and tangible objects with the idea of a distance filled with visible and tangible objects. As a result, we are led to suppose, first, that the former as well as the latter constitutes an idea of space, and second, that we can easily conceive of space devoid of visible and tangible objects. We are able to talk of vacuum solely on the basis of this mistake, and not on the basis of having an idea of vacuum (T 1.2.5.22). Accordingly, the mind is “very apt” to mistake the idea of an absence of change with the idea of a succession of objects. As a result, we suppose that the former as well as the latter constitutes an idea of time, and we proceed to talk of stedfast objects, even though we have no idea of time without change.
Hume’s conclusion is that, when we think we are conceiving space and time separately from the objects that they are tied to in experience, we are only mistaking or confounding an idea of the absence of the relevant objects with an idea of space or time. The idea of the stedfast object is not an idea of time — we only mistakenly assume that it is due to its close relations to an idea of time.
2. Interpretive challenges
Scholars have struggled to interpret Hume’s epistemic assessment of the endurance fiction, specifically, his claim that the application of the idea of time to a stedfast object is lacking in “any propriety or exactness” (T 1.2.3.11).11 Some interpret Hume’s reference to the “falshood” of the fiction at face value and maintain that, for Hume, it is false that steadfast objects endure (Baxter 2008, 21, 103 note 17; McRae 1980, 120, 123–24; and Bennett 2001 vol.2, 356–57). At the same time, this position seems embarrassing. Bennett describes it as “preposterous” (2001 vol. 2, 358). If by “stedfast objects” Hume understands ordinary objects at rest — lampposts, buildings, rocks, etc. — he is implying that it is false to believe that these objects exist in time. Ainslie (2015) opts instead to qualify Hume’s position,12 but it is not clear the proposed qualifications are consistent with Hume’s emphatic claim that the ordinary belief is false.
In addition, it is not clear that Hume’s argument for this unusual position (that is, the propriety argument) withstands closer scrutiny. To see why, we must turn to a related interpretive debate concerning the “appearances” Hume identifies as giving rise to the endurance fiction.
Hume presents these “appearances” with the case of considering “a stedfast object at five-a-clock” and regarding “the same at six” (T 1.2.5.29). But how exactly are these appearances constituted? In terms of Hume’s own taxonomy for experience, what impressions comprise the experience of “considering a stedfast object”? According to the standard interpretation, the experience cannot be comprised of a single unchangeable impression of the object that lasts from five to six o’clock. This is because, for Hume, the mind is unable to hold a single perception for any length of time longer than an instant (where the length of time is measured by changes in the background). Instead, it is argued, the experience of the object must be comprised of qualitatively resembling (or even qualitatively identical) but numerically distinct impressions (Price 1940, 46–47; Stroud 1977, 102–3; Waxman 1994, 206–7; Rocknak 2013, 123–55).13 Against this interpretation, Baxter (1987; 2001; 2008, ch. 3) has proposed that the mind’s impression of the object from five to six o’clock is in fact comprised of a single unchangeable impression. As such, this impression is indeed perfectly succession-less. Yet, it co-exists with other successions. The mind imagines it to be successive because it confounds co-existence with succession with actual succession.14
The problem for the propriety argument is this: if the impression of the steadfast object is comprised of numerically distinct impressions, then it would seem that such an impression is, after all, comprised of a succession; there would be nothing improper in applying the idea of time to such an impression. Baxter’s reading is more consistent with Hume’s assessment that the application is improper: the impression and idea of the object from five to six o’clock are perfectly succession-less, and, consequently, we cannot “without a fiction” apply the idea of time to it (T 1.2.3.11). Despite this interpretive advantage, however, the position that Baxter attributes to Hume has an “air of inconsistency” that Baxter acknowledges: two perceptions can be equally simple in time — they can both be perfectly succession-less — and yet one can co-exist with a longer succession than the other, and in this sense be “longer” than the other, despite having no temporal parts (Baxter 2008, 38).15
The crucial question for us is, why insist in the first place that the five-to-six-o’clock object is not comprised of a succession and that it is a mistake to attribute time to it? Falkenstein (2017) raises this critical question. He observes that Hume could have theorized the experience of the five-to-six-o’clock object as the standard interpretation suggests — that is, as comprising a succession of qualitatively identical impressions. He could then have construed the experience of the five-to-six-o’clock object, and the idea to which it gives rise, as truly temporal. He would then have had a more consistent and intuitive view, or, at the very least, he would have “nullified the contortions of T 1.2.5.29” (2017, 48).
We may press Falkenstein’s objection further. Recall that, in his reference to Locke’s illustration of a rapidly spinning object, Hume observes that there can be a lack of correspondence between external objects and their impressions. Sometimes, we do not detect change in objects that are changing. Objects that change undetected have the capacity to produce the idea of time. A tomato that happens to be inside an utterly dark cave has the capacity, when in daylight, to produce the idea of red. If so, Hume’s dictum against applying ideas to objects that cannot produce them does not rule out applying the idea of time to objects that change undetected, just as it does not rule out applying the idea of color to objects in the dark. On the plausible assumption that all ordinary objects continuously change on some level or other, it is not improper to apply the idea of time to ordinary objects at rest. If so, why does Hume insist that the belief that these objects endure is false?
To sum up these challenges: Hume could have maintained that ordinary objects and our experiences of them continuously change. Had he done so, he would not have had to take issue with the seemingly unproblematic belief that ordinary objects at rest endure. It is thus al-together baffling why Hume opts to defend the cumbersome and not even plausible thesis that time does not exist in the “stedfast object” from five to six o’clock.
3. Hume and Locke on the application of the idea of time
To render Hume’s position coherent, we need to explain both why he takes issue with the belief that ordinary objects at rest endure and what the “falsehood” of the belief consists in. We believe we can offer these explanations by interpreting Hume’s position as a response to Locke’s specific views on the application of the idea of time.
By way of showing that Hume is thinking of Locke in his discussion of time, it is worth noting the ways in which Hume’s discussion echoes Locke’s. Hume endorses Locke’s view that the successiveness of ideas is the root source of all temporal concepts (Essay II.xiv.3). He also borrows some of Locke’s specific arguments for this view, for instance, the argument that sleep removes the awareness of time by removing the successiveness of ideas (T 1.2.3.7; Essay II.xiv.4). As we saw earlier, Hume explicitly references Locke’s distinction between the successiveness of perceptions and motion in the external world (T 1.2.3.7; Essay II.xiv.6–10). In the same passage, Hume praises Locke’s analysis of the temporal “bounds” of our perceptions (T 1.2.3.7; Essay II.xiv.6–10). Locke examines multiple instances of discrepancies between external-world succession and our experience of it — cases where the external-world succession is too quick to be detected (like the spinning object) and cases where it is too slow (like the motion of the Sun in the sky and the shadows of sundials) (Essay II.xiv.6–11). From these examples, Locke concludes, first, that our ideas are continuously changing at a relatively constant speed (Essay II.xiv.9, 13), and second, that external-world successions that either exceed or fall short of this speed beyond certain “bounds” will not be detected (Essay II.xiv.9).
There is a great deal to be said on the extent to which Hume follows Locke in his discussion of time. While the traces of Locke are noticeable, the differences are perhaps even more striking. An important difference is that Locke distinguishes between three temporal concepts: succession, duration (the distance between two parts of the succession), and time (the measurement of duration) (Essay II.xiv.3, II.xiv.17). We do not find the same distinction in Hume. Instead, Hume uses the three terms interchangeably.
The most significant difference for our purposes concerns their views on the application of temporal concepts. Consider the following sequence of passages in Locke:
Indeed a Man having from reflecting on the Succession and Number of his own Thoughts, got the Notion or Idea of Duration, he can apply that Notion to things, which exist while he does not think; as he, that has got the Idea of Extension from Bodies by his Sight or Touch, can apply it to distances, where no Body is seen or felt. (Essay II.xiv.5; boldface ours)
Wherever a man is, with all things at rest about him, without perceiving any motion at all; if during this hour of quiet he has been thinking, he will perceive the various ideas of his own thoughts in his own mind, appearing one after another, and thereby observe and find succession where he could observe no motion. (Essay II.xiv.6; boldface ours)
For supposing it were 5639 Miles, or millions of Miles, from this place to the remotest Body of the Universe … as we suppose it to be 5639 years, from this time to the first existence of any Body in the beginning of the World, we can, in our Thoughts, apply this measure of a Year to Duration before the Creation, or beyond the Duration of Bodies or Motion … (Essay II.xiv.25; boldface ours)
In the first passage above, Locke indicates that we can apply the idea of duration beyond the limits of our experience; we can say, for instance, that dinosaurs roamed the Earth for 180 million years, or that a solar eclipse in the year 2290 will last 20 minutes and 15 seconds. In the second passage, Locke pushes this application of the idea of duration further: we can apply the idea “where we observe no motion.” Locke indeed goes on to stress through the remainder of the chapter that the perception of motion is necessary neither for the idea of duration nor for the measurement of duration, and that we can apply ideas of duration and of measurements of duration to absences of motion (see, e.g., II.xiv.19 and II.xiv.24). By the end of the chapter, as the third passage indicates, Locke is ready to say that we can apply the idea of duration even outside the domain of the world and of any bodies whatsoever : to “duration before the Creation, or beyond the Duration of Bodies or Motion.” 16
What is driving Hume’s argument, we suggest, is precisely this inertial transition in Locke’s thought: from applying the idea of time to unobserved objects, to applying it to an absence of motion, to applying it to an absence of all bodies. Hume most certainly has a problem with this line of reasoning. When taken to its logical limit, it implies a notion that for Hume is the pinnacle of inconceivability: the existence of time as an entity onto itself, without any objects that occupy it. This notion is the notion of absolute time, and Hume addresses it directly. He comments that “the indivisible moments of time must be fill’d with some real object or existence, whose succession forms the duration, and makes it be conceivable by the mind” (T 1.2.3.17). Yet, the passages from Locke give Hume reason to not only address the idea of absolute time but also pinpoint where Locke goes wrong in his progressive application of the idea of duration. Locke is too willing to apply the ideas of duration and time to an ever-broader range of situations. In response, Hume diagnoses precisely the point at which an application of the idea of time becomes illegitimate. And the diagnosis is none other than his “propriety argument” and his account of how the endurance fiction arises.
The propriety argument tells us that Locke’s crucial misstep is applying the idea of duration to an absence of change. Hume identifies this as the misstep by applying his empirical approach to ideas. When we trace our idea of time to impressions, we realize that it is changing impressions that are the source of our idea of time; our idea of time is in fact inseparable from the idea of changing objects. Our application of the idea of time goes wrong the moment we apply the idea to an absence of change — in other words, the moment we begin to engage in the endurance fiction.
Hume’s account at T 1.2.5.29 supplements the propriety argument by explaining the “appearances” and the psychological processes that lead us to apply the idea of time to absences of change. As we saw, Hume’s account appeals to the experience of watching an object from five to six o’clock, yet Hume leaves this experience significantly under-described. The incompleteness of Hume’s description is responsible for many of the worries in the literature. Yet, when we consider Hume’s target in this discussion, we can make out the outlines of the experience Hume has in mind: it is an experience whose characteristic feature is an absence of change, such that, when this experience is considered qua absence of change, it cannot be said to be an experience of succession or time, even though it can be easily confounded with an experience of succession or time. Hume evidently considers such an experience commonplace enough to claim that the endurance fiction “almost universally takes place” (T 1.4.2.29). Thus, Hume envisions the experience both as an ordinary experience and as an experience that fulfills the philosophical parameter his argument requires, namely, the absence of change. He sees the experience from five to six o’clock as fitting this description.
We agree that Hume is envisioning the experience in question rather naively, but we do not think his naive conception is contorted or misguided. All that Hume expects is the capacity to notice an element of changelessness in our ordinary experiences. At least at first sight, this expectation seems reasonable. As you sit at your desk writing, you notice many such elements: the table, the walls, the pictures on the walls — these all strike you as experiences of objects that do not change. Perhaps the experience of an absence of change involves abstracting or isolating an element like the table from the changing elements in its environment. Perhaps it also involves a degree of idealizing — one might argue that the experience of the table is technically changing (for instance, if our visual impressions are never stable but instead buzz or flicker) and we only idealize it to be unchanging. Nevertheless, whether by abstraction, idealization, or maybe even both, we do in fact have experiences that can be adequately described as experiences of an absence of change.
Having noticed these absences of change in our experience, and having described or labeled them as such, Hume argues that we proceed to confound these absences with actual change or succession on account of the “three relations” between an absence of change and an actual succession (section 1). The endurance fiction is the product of this conflation. It is beside the point to ask whether our experience of the absence is constituted by multiple qualitatively identical impressions or whether it is constituted by a single unchanging impression. However constituted at the ontological level, at the level of an ordinary experience — a level where we do not distinguish between the table and our impressions of it (T 1.4.2.31) — what we notice is an absence of change. It is this absence that we then confound with actual change. Hume is correct in claiming that we mistakenly confound it with change because the stipulated datum — the absence — is in fact not an instance of change.
The apparent “preposterousness” of Hume’s claim that it is false to consider stedfast objects as enduring disappears when we appreciate what Hume is really after. As we noted, Hume is trying to thwart the line of reasoning that leads to the notion of absolute time by ruling out the application of the idea of time to unchanging objects. If we read Hume charitably, he is saying that it is a mistake and a falsehood to apply the idea of time to objects that we stipulate to be perfectly unchanging. If you conceive of the table as perfectly unchanging, you are engaging in a fiction if you go on to believe that the table endures. If the table is perfectly unchanging, then it cannot endure because it does not involve a succession of different objects. Suppose, however, that upon further reflection you realize that the table does in fact change. One of the implications of Hume’s discussion of the spinning object is that objects in the world can and typically do change even when we do not perceive those changes (T 1.2.3.7). When we pause to reflect on these objects, we often learn that they do in fact change in ways that we cannot perceive, and this belief is supported by experience, especially intersubjective experience. We might even learn that, at the fundamental physical level, the objects are split into numerically distinct time slices. If so, it would no longer be a mistake or a falsehood to attribute time to the table, because the table is no longer conceived of as unchanging. The falsehood is applying the idea of time to an object that we are stipulating to be unchangeable, where we hold this stipulation fixed. It is especially improper to apply the idea of time to a universe entirely devoid of objects as Locke does.
Hume is not asserting that ordinary objects, which to us often seem unchanging, do not endure. He is instead calling attention to a contra-diction between two beliefs we commonly hold: that these objects are unchanging and that they endure. Hume’s argument is that, if we are trying to reason properly,17 we cannot retain both beliefs — at least one of them must be surrendered. Maintaining that most or even all objects endure while abandoning the belief that they do so without changing would be consistent with Hume’s demand. And this is, of course, how we conduct ourselves in ordinary life; when we reflect further, we do not insist that objects that appear to be at rest are unchanging all the way down to the subatomic level.
Recall that a fiction consists in applying an idea to an object that cannot cause that idea (section 1). For Hume, an object that is truly unchanging cannot produce the idea of time, and this is because it cannot cause a succession of perceptions. Thus, it is a fiction to apply the idea of time to such an object.18 The fact that the unchanging object cannot cause a succession of perceptions and hence the idea of time helps to reinforce our previous observation that the question of how our impressions of the object are constituted is a distraction. Even if our impressions of the unchanging object were successive, the object itself would not be the cause of that succession. Think of a movie film camera recording an unchanging object, say a lamppost. The lamppost is the cause of the image recorded, but the lamppost is certainly not the cause of the rolling of the movie film camera. The rolling of the movie film camera has a different cause. Hume would stress that, even if our impressions of the unchanging object were changing, it remains the case that applying the idea of time to the object is a fiction, insofar as the object cannot be the source of those changes.
To put it another way, imagine a camera that works like those automatic lights that turn on only when they register movement or change. In this case, the object recorded influences or controls the recording itself. You can play with the camera, moving your hand in front of it causing the camera to turn on and off as you move the hand, stay still, and initiate a new movement. In this case, the camera is activated by the changing object; the changing object causes the recording of the camera. Hume’s argument, as we understand it, is that a camera that was activated only by change would never be activated by an unchanging object. The reason it is a fiction to apply the idea of time to an unchanging object is precisely that such an object does not have the capacity to activate the experience of time.
In this section, we have offered an interpretation of Hume’s position on the “falshood” of the belief that unchanging objects endure. As we saw, scholars find Hume’s motivations for this view perplexing. We have argued that Hume develops this view in order to diagnose the crucial misstep in our application of the idea of time. It is Locke’s progressive application of the idea that prompts this diagnostic effort from Hume. The misstep, as Hume construes it, is stipulating that an object is unchanging and simultaneously attributing time to the object. The falsehood consists in the conjunction of these beliefs. Hume’s view is not as implausible as it might seem at first sight, because he denies not that the ordinary objects we perceive to be unchanging endure, but rather, that “perfectly unchangeable” objects endure (T 1.2.3.11). We fall into the fiction when we regard ordinary objects at rest as both “perfectly unchangeable” and enduring; we do not fall into the fiction when we acknowledge that the objects do in fact change.
Before concluding, we would like to address an additional objection to Hume’s account of the endurance fiction. Some scholars have argued that Hume’s account of the endurance fiction and his account of the fiction of identity create a vicious circle. While explaining the belief in the external world, Hume addresses the question of how the mind can conceive of an object as an identity — as something that “is the same with itself” (T 1.4.2.26). At first sight, the question presents a puzzle. When we conceive of something as an identity, we conceive of it neither as a single object (since there would be no relata for the relation in this case) nor as a multiplicity of objects (since there would be no identity) but as something in between singularity and multiplicity. The question is how we can conceive of such a medium (T 1.4.2.28). The details of Hume’s solution to this puzzle are beyond the scope of this paper. What is crucial for our purposes is that Hume relies on the endurance fiction to solve it. In his view, the endurance fiction is prior in the order of development to our ability to conceive of an object as an identity: it is by first imagining that time changes while an object does not that we are then able to conceive of the object as a medium between singularity and multiplicity. Clearly, then, Hume’s account of the endurance fiction cannot presuppose, as a prerequisite to the formation of that fiction, that the mind already possesses a notion of identity; if it did, the two accounts would be circular. Barry Stroud, however, insists that the accounts are circular in just this way (1977, 104). Similarly, James O’Shea writes, “what is this ‘stedfast object,’ ‘the same’ at five and six o’clock, if not something with identity over time?” (1997, 193). O’Shea argues that, unless the mind views the appearances at five o’clock and six o’clock as forming an identity, the endurance fiction would not arise.
Our interpretation might resolve this difficulty. Hume’s account of the “appearances” does not assume that we posit an identity between them; it only assumes that we posit an absence of change. Crucially, the experience of an absence of change is not an experience of an identity. One can experience the table as an absence of change without going through the additional step of positing a relation of identity between various appearances of the table. Hume informs us that this step comes about only later. First, we notice absences of change; then we engage in the endurance fiction and attribute time to such absences; only then do we begin to think of the table in the two alternating ways that Hume describes in T 1.4.2.29 (as singularity and multiplicity) and thereby begin to think of it as an identity.19
4. Conclusion
Hume’s propriety argument and his account of the endurance fiction are far from being preposterous and embarrassing; they are impressively fine-tuned and sophisticated once their precise target is identified. Scholars have already realized that Hume’s attack against the idea of absolute space is much more nuanced than the simple application of his empiricist theory of perceptions to the concept of absolute space. Part of what this paper has meant to accomplish is to show that Hume’s attack against absolute time does not consist in the mere thesis that the idea of time is inseparable from succession or change. Hume’s propriety argument and his explanation of the endurance fiction are essential parts of the same project. Indeed, these texts target the precise point at which Locke’s embrace of Newtonian absolute time becomes possible.20
Notes
- References to Hume’s A Treatise of Human Nature (T) are to the 2011 edition, edited by David F. Norton and Mary J. Norton. Parenthetical citations provide book, part, section, and paragraph number. ⮭
- See, e.g., Locke Essay, II.xiv.3; Malebranche, Search p. 45; Berkeley, Principles sec. 98. ⮭
- Hume follows Huygens, Leibniz, and Berkeley in his disagreement with Newtonian absolutism. Interestingly, however, it was Hume’s empiricist treatment of space and time (and other subjects) that influenced Einstein’s theory of relativity, which finally led to the rejection of Newton’s absolute conception of time. See Slavov 2019, 396. ⮭
- An Enquiry Concerning Human Understanding cited as “E” followed by section and paragraph. ⮭
- We interpret “changeable” in this phrase to mean distinct either qualitatively or numerically. Some scholars interpret “changeable” in this phrase to refer to numerical change (Baxter 1987, 331; Cottrell 2016, 49). Baxter, for instance, argues that “if [Hume] meant qualitative change in our sense of alteration, he would have mentioned a changeable object (singular)” (1987, 331). However, at this point in the Treatise Hume has not yet addressed the concept of identity. Suppose that we have an experience of a lantern changing from red to blue. For Hume, such an experience would give rise to the idea of time. We can conceive of the change in terms of numerical difference (multiple lanterns, one red and one blue), or else in terms of numerical identity and qualitative difference (one lantern, multiple qualities). At this point in the Treatise, Hume might be conceiving of change in terms of numerical difference (hence the use of “objects,” plural) while still being open to the possibility that change can be conceived in terms of qualitative difference. ⮭
- A famous objection to Hume’s theory of space and time is that the ideas of space and time violate the “Copy Principle,” i.e. the principle that simple ideas are always copies of simple impressions (see, e.g., Kemp Smith 1941, 273–4; Waxman 1994, 116; Frasca-Spada 1998, 75; Allison 2008, 51). A discussion of this objection is beyond the scope of our paper, however. We here accept Falkenstein’s response to the objection in his 1997. ⮭
- For a historical discussion of the “spinning object” example, see Larivier and Lennon 2002. ⮭
- Our characterization of Humean fictions is standard in the scholarship (see, e.g., Traiger 1987, 386; and Ainslie 2015, 66). As Cottrell (2016, 50) notes, Hume also uses “fiction” to refer to the mental representations that the imagination creates by applying ideas in this way. ⮭
- Hume’s views on the vacuum are also the subject of scholarly debates; see, e.g., Frasca Spada 1998, ch.4; Boehm 2012; Ainslie 2015, ch.3; Cottrell 2019b. Frasca-Spada and Cottrell focus on the historical context of Hume’s views. While we do not have the space to address these debates here, we think our understanding of Hume’s account of how the fiction of the vacuum arises is uncontroversial. Our focus in this paper is applying Hume’s account of the experiences that give rise to the vacuum fiction to his account of the experiences that give rise to the endurance fiction. ⮭
- This emphasis is evident across T 1.2.5.6–7 and T 1.2.5.13. In T 1.2.5.6–7, Hume gives the example of “a man supported in the air … conveyed by an invisible power,” where, even though the man has a sensation of motion, “‘tis evident he is sensible of nothing,’” meaning any possibility of a tactile sensation has been removed. In T 1.2.5.13, Hume stresses that “the sensation, which arises from the motion” is the same both in the case of the man supported in the air and in the case of the man who feels two tangible objects separated by motion—that is, in both cases the sensation of motion is devoid of anything tangible. ⮭
- A great deal has been written on the epistemic status of fictions in general, especially the fictions involved in the belief in external objects (for a review of this literature, see Cottrell 2016, 50–51). In this paper, we focus on the ‘impropriety’ of the endurance fiction in particular (and by extension the vacuum fiction). We think this focus is warranted because the vacuum and endurance fictions are structurally different from the fictions Hume discusses elsewhere in the Treatise, and so it is not clear to what extent Hume’s views on the epistemic status of the other fictions informs his view on the epistemic status of the vacuum and endurance fictions. ⮭
- Ainslie (2015) qualifies Hume’s position by suggesting that the endurance fiction involves a “constitutive” rather than “epistemic mistake” on the part of the vulgar—the mistake of confounding two objects and imagining a new object on the basis of the conflation, as opposed to the mistake of making false statements about an object (107). On Ainslie’s reading, Hume is “willing to countenance” talk of vacuums and stedfast objects so long as this talk is “properly understood in terms of what we think of when the mind substitutes ideas for one another” (83). We do not have the space to address the details of Ainslie’s reading. However, we think our interpretation in section 3 improves on Ainslie’s by taking Hume at his word when he says that the belief in unchanging enduring objects is false. ⮭
- The passages cited in support for this reading include T 1.4.2.15, T 1.4.6.4, and T 2.1.4.2. ⮭
- Baxter supports his reading by citing T 1.2.3.7, T 1.4.2.29, and T 1.4.2.33. ⮭
- Baxter defends the view against this apparent inconsistency by offering a for-malization of it, but it is not clear that his defense overcomes the objections that have been raised in the literature. See, e.g., Falkenstein 2017 (note 5), and O’Shea 1997. O’Shea argues that, so long as a background of changing perceptions (say, a succession of clock ticks) allows us to distinguish between, for instance, steadfast-object-along-with-tick-1 and steadfast-object-along-with-tick-2, Hume’s “separability principle” implies that the one really is distinct from the other, and hence, that the perception of the steadfast can be distinguished into temporal parts (O’Shea 1997, 199). ⮭
- Gorham and Slowik (2014, 120–24) offer further evidence and discussion of Locke’s embrace of absolute space and time. On their analysis, Locke’s views on space and time evolved from draft A of the Essay in 1671 to the published Essay of 1690. While Locke is initially hesitant about (and in some texts even explicitly denies) the reality of space and time as entities onto themselves, by the time of the publication of the Essay he “overcomes his empiricist hesi-tations about the ontology of absolute space and time” (127). Gorham and Slowik give evidence that this shift is motivated by both theological and Newtonian considerations (120–27). See also Gorham 2020. ⮭
- Hume is not chastising ordinary people for failing to be precise in their application of the idea of time. His note that the endurance fiction “almost universally takes place” suggests that he regards the fiction as innocent and understandable in ordinary contexts (T 1.4.2.29). Hume is rather criticizing those who pretend to or pride themselves of being precise, namely his fellow philosophers and scientists. ⮭
- Here, we are holding fixed the stipulation that the object is unchanging. If one varies this parameter, Hume’s claims about the temporality of the object no longer apply. It is not an objection to Hume that what we are stipulating to be unchanging is really changing: this retort simply changes the parameters of Hume’s argument. Just as it would not be an objection to the claim that “colors do not apply to sounds” that sounds are not really sounds, it is not an objection to Hume’s claim that “time does not apply to unchanging objects” that those objects are not really unchanging. ⮭
- We tend to conceptualize the absence of change as the relation of sameness or identity. Hume himself hints at this conceptualization when he describes the object as “the same” at five and six (T 1.2.5.29). However, Hume does not say that we in fact regard the object as “the same” while in the midst of engaging in the endurance fiction; he only helps himself to describing the object as such in the context of philosophizing about it. When we engage in the endurance fiction, especially during childhood, we simply attribute an absence of change to our experience, without, in addition, positing an identical object. ⮭
- We would like to thank Jani Hakkarainen and two anonymous referees for helpful and insightful feedback on earlier drafts of this paper. ⮭
Bibliography
Ainslie, Donald. 2015. Hume’s True Skepticism. Oxford: Oxford University Press.
Allison, Henry E. 2008. Custom and Reason in Hume: A Kantian Reading of the First Book of the Treatise. Oxford: Oxford University Press.
Baxter, Donald. 1987. “A Defense of Hume on Identity Through Time.” Hume Studies 13 (2): 323–42.
Baxter, Donald. 2001. “Hume on Stedfast Objects and Time.” Hume Studies 27 (1): 129–48.
Baxter, Donald. 2008. Hume’s Difficulty: Time and Identity in the Treatise. New York: Routledge.
Bennett, Jonathan. 2001. Learning from Six Philosophers: Descartes, Spinoza, Leibniz, Locke, Berkeley, Hume, vol. 2. New York: Oxford University Press.
Berkeley, George. 1707–1720/1948–1957. The Works of George Berkeley, Bishop of Cloyne. Edited by A. A. Luce and T. E. Jessop. 9 vols. London: Thomas Nelson and Sons.
Boehm, Miren. 2012. “Filling the Gaps in Hume’s Vacuums.” Hume Studies 38 (1): 79–99.
Cottrell, Jonathan. 2016. “A Puzzle About Fictions in the Treatise.” Journal of the History of Philosophy 54 (1): 47–73.
Cottrell, Jonathan. 2019a. “Hume on Space and Time: A Limited Defense.” In The Humean Mind, edited by Angela Coventry and Alexander Sager, 83–95. New York: Routledge.
Cottrell, Jonathan. 2019b. “Hume’s Answer to Bayle on the Vacuum.” Archiv für Geschichte der Philosophie 101 (2): 205–36.
Falkenstein, Lorne. 1997. “Hume on Manners of Disposition and the Ideas of Space and Time.” Archiv für Geschichte der Philosophie 79 (2): 179–201.
Falkenstein, Lorne. 2017. “Hume on Temporal Experience.” In The Routledge Handbook of Philosophy of Temporal Experience, edited by Ian Phillips, 42–52. New York: Routledge.
Frasca-Spada, Marina. 1998. Space and the Self in Hume’s Treatise. Cambridge: Cambridge University Press.
Gorham, Geoffrey. 2020. “Locke on Space, Time, and God.” Ergo 7 (07): 219–40.
Gorham, Geoffrey and Slowik, Edward. 2014. “Locke and Newton on Space and Time and their Sensible Measures.” In Newton and Empiricism, edited by Zvi Biener and Eric Schliesser, 119–37. Oxford: Oxford University Press.
Hume, David. 1739/2011. A Treatise of Human Nature. Edited by David F. Norton and Mary J. Norton. Oxford: Oxford University Press.
Hume, David. 1748/1975. An Enquiry Concerning Human Understanding, in Enquiries Concerning Human Understanding and Concerning the Principles of Morals. Edited by L. A. Selby-Bigge and P. H. Nidditch. Oxford: Oxford University Press.
Kemp Smith, Norman. 1941. The Philosophy of David Hume. London: Macmillan.
Larivier, Anthony and Lennon, Thomas. 2002. “The History and Significance of Hume’s Burning Coal Example: Time, Identity, and Individuation.” Journal of Philosophical Research 27: 511–26.
Locke, John. 1689/1975. An Essay Concerning Human Understanding. Edited by P. H. Nidditch. Oxford: Oxford University Press.
Locke, John. 1990. Drafts for the Essay Concerning Human Understanding and Other Philosophical Writings. Edited by P. H. Nidditch and G. A. J. Rogers. Oxford: Oxford University Press.
Malebranche, Nicolas. 1674/1980. The Search after Truth. Edited by Thomas M. Lennon and Paul J. Olscamp. Cambridge: Cambridge University Press.
McRae, Robert. 1980. “The Import of Hume’s Theory of Time.” Hume Studies 6: 119–32.
O’Shea, James. 1997. “Fictitious Duration and Informative Identity in Hume’s Treatise.” Manuscrito: revista internacional de filosofia 20.2: 169–211.
Price, H. H. 1940. Hume’s Theory of the External World. Oxford: Clarendon Press.
Rocknack, Stefanie. 2013. Imagined Causes: Hume’s Conception of Objects. Oneonta, NY: Springer.
Russell, Paul. 2008. The Riddle of Hume’s Treatise: Skepticism, Naturalism, and Irreligion. Oxford: Oxford University Press.
Slavov, Matias. 2019. “Hume, the Philosophy of Science, and the Scientific Tradition.” In The Humean Mind, edited by Angela Coventry and Alexander Sager, 388–402. New York: Routledge.
Stroud, Barry. 1977. Hume. London: Routledge.
Traiger, Saul. 1987. “Impressions, Ideas, and Fictions.” Hume Studies 13 (2): 381–99.
Waxman, Wayne. 1994. Hume’s Theory of Consciousness. Cambridge: Cambridge University Press.