Absence and Nothing
The Philosophy of What There is Not
STEPHEN MUMFORD
1
Great Clarendon Street, Oxford, OX2 6DP, United Kingdom
Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries
© Stephen Mumford 2021
The moral rights of the author have been asserted
First Edition published in 2021
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above
You must not circulate this work in any other form and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data Data available
Library of Congress Control Number: 2021911519
ISBN 978–0–19–883153–2
DOI: 10.1093/oso/9780198831532.001.0001
Printed and bound in the UK by TJ Books Limited
Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.
5.4
7. Perception
7.5
7.6
8. Empty Reference
8.1
8.2
8.3
8.4
8.5
9. Negative Truth
11. Negative Belief
Preface
I am grateful to the Leverhulme Trust in granting me a Major Research Fellowship from 2016 to 2019 specifically to complete this work. Leverhulme remains willing to fund blue-skies research that has no immediate practical benefit. The present book falls firmly into that category.
Problems relating to absence and nothingness have troubled me since 1989 when I read Armstrong’s A Theory of Universals. He discusses at one point whether there are negative properties, such as being non-red, non-circular, uninitiated, sugar-free, and so on. I immediately shared Armstrong’s instinct that such ‘features’ did not deserve a place in reality. The problem was, however, that when Armstrong provided specific arguments against negative properties, they all seemed question begging. As an unknown PhD student, I wrote a letter to Armstrong (in the days before email) on the other side of the world and explained my worries. To my delight, a few weeks later I received a detailed reply in which Armstrong admitted that his arguments had begged the question and he then offered some further considerations. Negative properties did not seem particularly natural and perhaps should be excluded from our ontology on that basis. I couldn’t rest content with this explanation either but had to shelve my concerns and get on with my PhD, which was on an unrelated topic. Sometime later, doctorate secured, I revived my interest in negativity and presented a paper in which I attempted a new argument against negative properties. This contained an elementary error so I shelved it indefinitely. If Armstrong’s arguments had failed, mine had failed spectacularly.
Negativity would not go away, however. Over time, I noticed a number of other philosophical problems that concerned what there was not: problems such as causation by absence and truthmakers for negative truth. The latter was a problem I took from George Molnar. In one of the final papers he published, he articulated the problem in what is now widely acknowledged as the definitive formulation. His paper ends on a sorrowful note, however. No one had a solution to this problem and he could find none either. I took this as a challenge and published a couple of small pieces on the subject, with which I was not entirely happy. Still more problems of absence and nothingness arose: problems such as perception of absence, nonentities, holes and shadows, negative facts and the act of denial. In 2005, I first had the idea of collecting these together for a monograph in which there would be a common solution offered. I have a notebook from that date in which I plan out the contents for A Book About Nothing, which later became Nothing Really Matters. Neither of those was written because I became
immersed in other projects and then unfortunately fell into university management. The idea was to have a single, systematic approach that would dissolve these many related problems and show that nothingness was not a feature of ontological reality but only a feature of our ways of thinking or speaking. Until I examined the detail, however, I could not be sure that it would all work out that way.
The plan was once more shelved as other matters took priority. Only as I reached the end of my period in management did I think once more that I ought to revive my interest in nothing in particular. Having dwelt on absences and nothings for twenty-six years, I was able to make the case to Leverhulme that the book really would not be completed without their valuable support. As Colin Howson wryly observed, they decided I was good for nothing.
I started the detailed work, which took some unexpected turns. Most notably, I recognized the benefit of situating my approach within the Parmenidean tradition. I realized, in doing so, that my concern was with the very first problem seriously discussed in Western philosophical history. That shouldn’t really come as a surprise. The topic, after all, is the very division between Being and Nonbeing. There can be nothing more fundamental than that. Parmenides was largely correct. What exists is everything. What does not exist is nothing. But Parmenides rather overextended this thesis, arguing that we cannot even think about what there is not, and that change is impossible. I was never going to accept that. My sympathies were only softly Parmenidean, I realized, and this could inform an approach to the whole subject matter. Let us test whether we can see through a Parmenidean project, explaining some long-standing philosophical problems, without ever having to invoke an absence or nothingness as part of reality. I will leave the reader in suspense over whether a soft Parmenidean approach is successful in the final reckoning. One further discovery, however, was that we cannot resolve all these issues doing metaphysics alone. An understanding of empty reference, denial, and negation was also necessary. Another possible title, The Metaphysics of Nothing, also had to be rejected, then. Only the first half of this book is metaphysics. The second half is epistemology, philosophy of language, and logic.
What have I got to show for over thirty years thinking about nothing? I have certainly learnt a lesson in humility. These philosophical questions are bigger than all of us, and it is very unlikely that a simple response will work. A whole lifetime is barely enough to complete this kind of work. Nevertheless, I hope that the present book offers at least some progress and might allow others to see the issues in a more illuminating light. I’m happy if what I have done enables or encourages others to do better.
Having acknowledged the Leverhulme Trust, I should also thank those who have supported me personally, especially over the last three years when I was working intensely on the book. That includes my family, whom I relocated from
Nottingham to Durham, and all my colleagues in Philosophy at Durham. Being on externally funded leave meant that I was not contributing to the life of the department in all the ways I would have liked. I should thank Oxford University Press, especially philosophy editor Peter Momtchiloff, for once again having faith in my work and making a commitment to seeing it published. Peter’s readers then gave me many detailed comments. Appointed readers and referees often get a bad press, and it is almost always a thankless task. It is important for me to point out, then, that the two anonymous readers of the beta-version gave me the most helpful, incisive, and constructive comments I’ve ever had. In countless places I have followed their suggestions, and it has made for a vastly improved book. Also working on behalf of the press, I must thank my copy editor, Phil Dines, who had to wrestle with, in places, a difficult text, and push me on some deeper-thanyou’d-think lexical-philosophical matters, such as whether we should distinguish ‘what-is-not’ from ‘what is not’. Copy editors are among the many unsung heroes who turn an author’s almost-ready thoughts into a presentable book.
I would like to thank all those who attended talks I gave on topics related to absence and nothing and for the comments and criticisms received. As this has been over such a long period, there are too many venues to list and names to name, but I hope that, if you’re one of those people, this volume does justice to your concerns and you find the answer to your question herein. I am grateful to Liva Rotkale for assisting me with the Latvian emphatic denials used in chapter 10. Some of the material in chapter 10 has appeared in an earlier form as ‘Negation and Denial’ in the Cambridge Handbook of Philosophy of Language, P. Stalmaszczyk (ed.), and appears here with permission of the editor and publisher.
Finally, this book was completed during the pandemic at a time that has been difficult for us all. In my own case, it was an extremely tough year. As I write these words, the crisis is still far from concluded. Friends and family have helped me through, and I will never forget the importance of that. My hope is that some good emerges from this nightmare and that we will re-evaluate the ways we live to produce a better and more caring post-pandemic world.
Soft Parmenideanism
1.1 Nothing really matters
It matters to us how things are. But how things are not can be equally important. It matters when something you want is not there. In a variety of circumstances, it matters that a drink is sugar-free. A migraine sufferer might want any pudding as long as it is not chocolate. Anyone with an allergy or intolerance will have a similar type of negative preference. It could matter to someone that a particular train does not stop in Wakefield, if they were planning to get off there. Bereavement matters too. When a loved one is no longer around, the feeling of their absence is profound. Some fear their own non-existence more than anything else. In other cases, things not being a certain way can be entirely mundane. Someone might regret not being six-feet tall or not being a better chess player. But some cases of what is not are of great, even cosmic importance. Because the spread of Covid-19 was not stopped, the UK suffered over 130,000 deaths. This omission itself could be explained in terms of an absence: the lack of an effective government strategy. It matters enormously that the planet Earth is not closer to the sun than it is. It is likely that there would be no human life on the planet if it were. Equally, it matters that we have not recently been hit by a giant asteroid: the sort of impact that might have led to extinction of the dinosaurs. If nothing matters so much, we ought to have at least a basic understanding of how it matters. Unfortunately, as soon as we try to understand what-is-not, we are immersed in the deepest of deep philosophical perplexities. If what-is-not is an absence of something, does it mean that absences are parts of reality? How could they matter otherwise? But if they are real, it seems as if we are saying that what-isnot is a part of what-is. That looks downright self-contradictory. Suppose, then, that we straightforwardly reject the reality of absences, voids and nothings, and any kind of entity that looks ‘negative’. Can we still give a complete account of the world? For example, in some cases of causation, it looks as if it matters not only what is present but also what is absent. If you had no air to breathe, you would die. What would kill you? The most obvious cause would be a lack or absence of oxygen. It is very difficult to explain the death in other, ‘positive’ ways: in terms of what was present. Various things were there. But their presences don’t entail the absence of oxygen, which it seems would have to be cited as an extra fact about the situation, in addition to the facts of all that was present. Furthermore, it seems that the presences of
all those things are not what explain your death. The absence of oxygen looks essential and indispensable in explaining the effect under consideration. These questions, we will see, are not trivial. They are easy to state but hard to answer. The distinction between being and non-being must be understood for us to make progress, and perhaps there is no more fundamental distinction in the whole of metaphysics than this. There will be much more we must understand besides.
1.2 The first argument
Nothingness, non-being, was the first philosophical problem in the Western tradition. The oldest such surviving text in which there is sustained philosophical argument is Parmenides’ poem On Nature, which discusses non-being in the section entitled ‘The Way of Truth’ (Parmenides Fragments). Graham (2010: 4) makes a credible point when he says that the main division in Greek philosophy should not be between pre- and post-Socratics but between pre-Parmenideans and post-Parmenideans. Plato’s dialogue The Sophist can be understood as a direct response to Parmenides. In one place, the Eleatic visitor, granting a claim that looks anti-Parmenidean, confesses that he might be guilty of patricide (Sophist: 241d3–4). Parmenides was from Elea, and the Eleatic visitor is intended, either literally or figuratively, to be Parmenides’ son.
Parmenides argued that there is no non-being. Indeed, we should not even speak of it. Gorgias (On What-is-Not: 741) soon dissented, claiming that what-isnot exists. If we look at more recent philosophical history, non-being plays a significant role in the philosophies of Hegel, Heidegger, and Sartre. It is not just a view of the Continental tradition, however. Russell (1918: 206–8) admitted that there must exist general facts that are essentially negative. Despite being a professed naturalist, Armstrong (2004: 54–7) followed Russell in accepting higherorder negative facts, which his facts of totality are. Among contemporaries, Schaffer (2004) believes that absences can be causes, which seems to commit to their existence and their powerfulness. Hommen (2013) and Zangwill (2011) accept that there are negative properties, such as being non-red. Barker and Jago (2012) allow negative instantiation as a likely part of reality. And Priest (2005) states that some non-existent things nevertheless are. It seems that there is currently no shortage of analytic philosophers willing to whisper sweet nothings in our ears. Often they will depict themselves as holding a radically unorthodox view. The above list shows, however, that the acceptance of negatives has become mainstream. Take, for example, Sorenson, who describes himself as ‘a reluctant apostle of negative metaphysics’ (Sorensen 2008: 16), committing to the reality of negative features in the world and our ability to perceive them. Nevertheless, he begins by assuming that ‘Philosophers and physicists alike have a strong conviction that reality is positive’ and speaks of a natural ‘Aversion to negative things’
(Sorensen 2008: 16). On the contrary, a philosophy of ‘negativism’ is currently prevalent. It is virtually conventional wisdom.
Despite this ongoing acceptance of negative entities of various kinds, I will defend the spirit of Parmenides’ position, if not the letter. There is no nothing. Absences and nothingnesses have zero reality.
I am, however, going to adopt a stance that I call soft Parmenideanism. I use this label with a dual meaning. Soft Parmenideanism is my name for a methodology that is Parmenidean in tone but, unlike Parmenides himself, sets out what conditions would have to be met in order for a negative entity to be accepted into our ontology. The hard Parmenidean, presumably, is someone who would never countenance negative entities in any circumstances, which is another way of saying that a rejection of negative entities is their most basic and fundamental commitment. Below, I shall be explaining the soft Parmenidean methodology and justifying its adoption.
Soft Parmenideanism can also be used as a label for a position that is held. Such soft Parmenideanism would be accepted as an ontological thesis. It is soft because it accepts some of what Parmenides said—basically, the rejection of negative existents—but it does not accept all of what Parmenides said. The soft position rejects the more hardline views of Parmenides: such as that motion and change are not real (perhaps the view most associated with the Eleatics) and that it is impossible to think about what is not. As Gallop (1984: 24) shows, it is in any case dubious that these stronger claims followed from Parmenides’ reasonable starting point that what-is-not is not anything. Hard and uncompromising Parmenideanism is certainly entertaining, as Della Rocca (2020) proves, but it is not the view I will be defending here. Taken overall, the present book makes the case for a soft Parmenideanism, both as an ontological thesis and as a philosophical methodology.
We need to understand Parmenides’ original position and why so many have sought to overturn the parts that seem obviously true. Some might think it trivially so. I duly start with a look at Parmenides’ specific claims and his arguments for them. I will then go on to outline an approach of soft methodological Parmenideanism that will be applied in the rest of this book. It follows the spirit of Parmenides’ position, though it leaves open the possibility that the approach might not deliver the hard conclusions for which Parmenides is notorious. It accepts that there are conditions under which some form of negative entity would have to be accepted as real. The rest of this book will then examine whether there are in fact cases in which those conditions are met.
1.3 Soft ontological Parmenideanism
I shall come to my proposed Parmenidean methodology shortly, but first I wish to describe, and to an extent justify, my soft Parmenideanism as an ontological thesis. This is a view concerning what there is and what there isn’t.
We must begin with Parmenides’ text, specifically ‘The Way of Truth’, which is accepted as the part of the poem in which Parmenides expresses his own philosophy, though he has a character voice the words. I will highlight seven main claims that I think should be taken from Parmenides before explaining why some of them, but not all, might be accepted. Another reader of Parmenides could produce a different list of the Presocratic’s most important claims. Della Rocca (2020: 1), for instance, argues that the rejection of all distinctions is a key claim in Parmenides. Parmenides’ philosophy is particularly open to interpretation since it is written in hexameter verse and the exact meaning is debatable. I admit also that I have approached Parmenides’ text with particular interests and purposes in mind, as the rest of this book evinces. I believe exactly the same can be said of Della Rocca’s book, of course.
All those provisos declared, we can say that Parmenides is reasonably attributed with the following views:
Pi. Nothing is not (A)
Pii. There are no degrees of being (A)
Piii. Nothing comes from nothing (A/R)
Piv. Non-being is unknowable (A/R)
Pv. Reality is a single plenum (A/R)
Pvi. Non-being is unthinkable and unnameable (R)
Pvii. Neither motion nor change is possible (R)
As an indicator of what is to come, I have followed each claim with an (A), where I think we should accept it, an (R) where I think we should reject it, or an (A/R) where I think that we can either accept or reject it, depending on how it is interpreted and our other commitments.
Let us take it, then, that a hard ontological Parmenidean is someone who accepts all of Pi to Pvii. Such a position would fully deserve to be called hard. The view would be not just that there is no nothing but also that we cannot even speak or think of what is not. It would follow, of course, that one could not even consistently deny that nothingness is something. If I was a hard ontological Parmenidean, this could be a very short book. None of the discussions that follow, concerning negative properties, causation by absence, negative truth, perception of absence, and so on, could meaningfully be articulated.
The soft Parmenidean, I suggest, is someone who holds to a proper subset of Pi to Pvii, so rejects at least one of them. This immediately suggests the question of how many claims are to be held and allows that there could be degrees of Parmenideanism. Some soft Parmenideans might be harder than others, if they reject just one of Pi to Pvii, for instance, while another rejects five of them. My own soft Parmenideanism, as indicated above, is towards the softer end of the scale, given that I see it as a firm commitment to only two of Pi–Pvii. Furthermore, I commit to the Parmenidean claims that I take to be the most obvious, almost
undeniable, and which cause relatively little difficulty with the rest of our philosophy. But my Parmenideanism is not completely soft. I would be willing to accept Piii, iv, and v if they are interpreted in a certain way or restricted in a specific manner. That could then lead to accepting all of Pi to Pv in some form or other, which would be a fairly robust Parmenideanism. As the above menu indicates, however, I do think that it is a perfectly tenable position to hold only Pi and ii while rejecting Piii, iv, and v. It can be inferred from this that I do not take Pi and ii to entail iii to v.
So that this discussion does not remain forever abstract, I should say something of what I take each of Pi to Pvii to mean, give some textual justification for attributing such views to Parmenides, and provide preliminary reasons for why I think each claim should be accepted or rejected. (The references to Parmenides’ text are given by the number of the fragment followed by the line number, all taken from the Gallop 1984 edition. Hence 6.2 refers to fragment 6, line 2)
Pi. Nothing is not nothing is not, that is what I bid you consider (6.2) For never shall this prevail, that things that are not are; (7.1)
Not all of Pi to Pvii carry equal weight, since it is hard to see how anyone who rejects Pi can claim to be any kind of Parmenidean at all, even if they held most or all of the remaining claims. This is Parmenides’ most important and significant claim, since it concerns the division between being and non-being. Indeed, I have arranged Pi to Pvii in an order where they become progressively less and less essential to Parmenideanism, as I am conceiving it.
How is Pi to be understood? In its simplest form, which is how I will take it, Parmenides is saying that there is no nothing. There is only what-is. There is only what exists, hence non-being is not something. The point ought, then, to be obvious. Perhaps it is trivial and hardly worth stating. It rejects what Gallop (1984: 24–6) calls the Homeric equivocation (after Homer, Odysseus IX.366–7 and 398–412). If you say ‘there is nothing in my pocket’ you cannot be taken to mean that there is indeed something in your pocket, namely ‘nothing’, since nothing is not a thing. Similarly, in Homer, to say ‘Nobody killed him’ does not mean that someone did kill him, namely ‘Nobody’. Nor, in Parmenides, when the goddess accuses humans of ‘knowing nothing’ (6.4), she does not mean that there is an object of knowledge called ‘nothing’ that we know. Such inferences are also parodied by Lewis Carroll, through his character of the White King, who misunderstands the statement ‘I see nobody on the road’ (Carroll 1871: 234). We should not, therefore, attempt any kind of positive description of the nature of nothingness, such as we find in Heidegger (1929), for instance.
That the Pi claim seems trivial can be questioned to an extent, however. For one thing, it is open to interpretation exactly what Parmenides meant since the ‘is’ in ‘it is and cannot not be’ (2.3) is ambiguous and it is likely that Parmenides
was using it in slightly different ways in different places (Waterfield 2000: 50). The ‘is’ can be interpreted existentially, as when one says ‘the Earth is’, simply meaning that it exists. But ‘is’ can also be used predicatively, as in ‘it is . . . ’. Waterfield also distinguishes a veridical sense of ‘is’ that might be in play since only if x is F, for some attribute F, can we say that x is real. With only limited and imprecise textual evidence, we cannot rule out any of these meanings of ‘is’, so we cannot immediately dismiss Pi as trivial. For the purposes here, I am accepting an existential interpretation of the claim and thus taking it, as indicated above, as a claim about what there is (everything) and what there is not (nothing).
A second reason to think that Pi is not entirely trivial is that Parmenides thinks it worthy of an argument, which he gives at 8.15–16. Gallop (1984: 7) reconstructs it as follows:
[A] 1. Either ‘is’ or ‘is not’ (and not both)
2. Not ‘is not’
Therefore,
3. Is.
The disjunction in 1 can also be used to argue from 3 to the conclusion of 2, of course: that not is not. Now it might be significant in the history of philosophy that Parmenides sought to argue for Pi rather than merely assert it, since most of what we have from his predecessors consists in assertion or rhetoric rather than arguments (see Waterfield 2000: 49). In many ways, then, this is how serious philosophy truly began. What can be taken from the fact that Parmenides offered an argument is that although the claim in Pi might seem obvious and even trivial, Parmenides evidently did not think so since the argument [A] shows that he thought there were assumptions that were even more basic and fundamental than the conclusion expressed in Pi. After all, someone could think it possible both that there is and there is not—there is both something and nothing—and this is what premise 1 of [A] denies by making the disjunction exclusive. Of course, once premise 1 is accepted, premise 2 looks a lot like Pi itself, so we might not be persuaded by [A] as a reason to accept Pi. Perhaps a better argument could be constructed.
There is, however, a third and perhaps the most important reason to doubt that Pi is trivial, which is that many philosophers effectively deny it. As we will see in the remainder of the present book, there are reasonable analytic philosophers who think that we have no choice but to accept certain kinds of negative existence. They do not think that they are denying a trivial truth, for they would usually understand that to be confused, but it is almost certain that they are denying Pi, under its existential interpretation. Even if one thinks Pi is not exactly trivial, one might still think that it is undeniable for other reasons. To put it in familiar terms, one might say that Pi is at the centre of our web of belief (Quine and Ullian 1970). A consideration of soft Parmenideanism as a methodology will
reveal, however, that there could be theoretical grounds for denying Pi if it would leave us with an overall more coherent web of beliefs.
There is a final remark on Pi concerning Parmenides apparent necessitarianism. Parmenides characterizes what-is also as what ‘cannot not be’ (2.3) and what-isnot as what ‘needs must not be’ (2.5). Consequently, it would be possible to ascribe to Parmenides a thesis that what is necessarily is and what is not necessarily is not; that is, a necessitarian view of nature, which denies that there are any contingent truths. This could easily be an eighth Parmenidean thesis on the list, but I do not itemize it separately since I think it is best covered under Pv.
Pii. There are no degrees of being [It] must either be completely or not at all (8.11) [it] all alike is; (8.22)
. . . nor is there a way in which what-is could be More here and less there, since [it] all inviolably is; (8.47–8)
Pii tells us that existence is univocal in respect of degree. Something that exists, exists entirely and absolutely. Nothing can exist to a lesser extent than something else. Existence does not admit of any degrees at all, just as there is no sense of more or less perfect. Hence, the most insignificant speck of dust, if it exists, exists in the same sense and just as fully as the island of Manhattan exists, or the planet Earth, or the whole universe. There is, thus, a stark choice for us when we evaluate what there is. There is no middle ground between existence and non-existence. There is nothing that partially exists or almost exists. One can cut a log in half and the resultant halves exist just as much and in the same sense that the previous whole did.
This thesis, again, might seem very obvious. But there was a time when it was necessary to reassert it. Pii is associated in the twentieth century with John Anderson, for instance, having been spread by his student David Armstrong (see Mumford 2007b: ch. 1). It is seemingly challenged in certain interpretations of Meinong, but, as we shall see (ch. 8, below), this is largely due to Russell’s misrepresentation (such as Russell 1905). Russell’s own statement ‘there is only one kind of being, namely, being simpliciter, and only one kind of existence, namely, existence simpliciter’ (Russell 1903: §427) sounds in accord with Pii. The appearance is superficial, however. Read in context, it is clear that Russell was distinguishing being from existence: both were univocal, but they meant different things. This is not, then, the sort of view a Parmenidean should welcome. We will see that even some contemporary philosophers deny Pii, holding a position named inegalitarianism (for example, Zangwill 2011). Pii rules against any equivocating, prevaricating, or obfuscation. Along with Pi, Pii is a Parmenidean thesis that I accept into my soft Parmenideanism. It rules that negative-being cannot just be a different kind of being, or a lesser degree of it. All I can see against Pii is that it might
also be a trivial truth, but again, that is dubious when some have denied it. We would, I maintain, need very good reasons to abandon it, and they had better not be ad hoc.
Piii. Nothing comes from nothing what-is is ungenerated and imperishable (8.3)
In what way, whence, did [it] grow? Neither from what-is-not shall I allow You to say or think; for it is not to be said or thought That [it] is not. And what need could have impelled it to grow Later or sooner, if it began from nothing? (8.7–10)
It is on the basis of these fragments that Parmenides is attributed with the view that nothing comes from nothing. Parmenides might have been the first to articulate the view, though his formulation is not the best known. Lucretius’ ‘nothing can be created out of nothing’ (De Rerum Natura: 31) probably holds that distinction. It is with Piii, however, that matters become more complicated for the soft ontological Parmenidean, since the text is ambiguous between a rejection of causation by absence and another view.
Parmenides argues against the possibility of change, and that includes genesis and perishing (see under Pvii, below), since genesis would involve a change from there being nothing to there being something, and perishing is a change from there being something to there being nothing. Setting aside any such general claim about change and considering in isolation the claim that nothing can come from nothing, then Piii looks a reasonable thesis to hold, certainly if one already accepts Pi. Waterfield explains as follows:
Parmenides denies that anything can come into existence from something that does not exist. Given a state of non-existence, we cannot explain a state of existence since we have no way of moving from one to the other. Since, by definition, what-is-not has no properties, it has no properties that could be taken to explain the generation of what-is. (Waterfield 2000: 52)
Another way of understanding this is that the non-existent has no causal powers, so there is nothing that could give rise to something. We will see (chapter 4), that there are contemporary philosophers who accept the reality of this so-called causation by absence, namely that absences can cause effects. Nevertheless, there is an obvious danger that an acceptance of causation by absence leads to a violation of Pi. The reason is to be found in the Eleatic stranger’s reality test in Plato’s Sophist (247d–e). The stranger says that causal power is the mark of the real: to be real is to bear at least some causal power. If one allows that an absence has a power to produce an effect, then, it entails that the absence is something real. Contrary to Pi, therefore, it is entailed that a nothing is a something. Parmenideans should avoid accepting causation by absence for this reason.
There is a second interpretation of the dictum ‘nothing comes from nothing’, however, that it is possible to reject. If one again sets aside Parmenides’ claim that change is generally impossible, there would be no reason to deny the possibility that there could have been nothing and then there was something, by which I mean spontaneous existence. This is not the same as causation by absence since it is not a case that the previous nothing produces or creates something. It is that suddenly there is something when before it there was nothing. No action of a causal power is required for existence of the something.
To decide which of these two interpretations is correct will depend on how we understand the comes from in ‘nothing comes from nothing’. Does it imply causation? If it does, then we should accept Piii, or at least so I will be arguing below. But if it does not imply causation, and one reads its denial as saying only that there could be something where previously there was nothing, then it seems one could meaningfully jettison Piii, under that reading. Nevertheless, as I have indicated, there is a version of Piii that I think a soft ontological Parmenidean ought to accept.
Piv. Non-being is unknowable
You could not know what-is-not . . . nor could you point it out (2.7–8)
Parmenides’ claim Piv is that non-being is unknowable. This again admits of ambiguity. It might be defensible under one reading but then be dubious under another.
If one takes a superficial reading, interpreting the words at face value, then the claim seems false, which is probably why our first judgement will be against Piv. Among the things known, it is tempting to include items such as that London is not the capital of France, that we do not live on Venus, that 2 + 2 does not make 5, that there are no Arctic penguins, and so on. Sometimes such ‘negative knowledge’ might be important to us, from an evolutionary standpoint, for instance: to know that there is no food at a certain place or that there is no way to jump across a ravine. If Piv amounts to a denial of negative knowledge then, at the very least, we are owed an explanation of how it seems that we can know things of this kind. Perhaps it is possible to explain away the appearances.
Under another reading, however, Piv looks a lot more plausible. To know something, or to perceive it, and thus to point it out, requires the existence of the object of knowledge or perception. It requires that the knower enters into a relation with the thing known and that the perceiver enter into a relation with the thing perceived. This is a factive account, meaning that P can be known or perceived only if P is. Hence, you cannot perceive a pot of gold if it is hallucinated; you can only misperceive it. Nor can you know that you are on the moon if it is not a fact that you are on the moon. You would be mistaken if you thought that you were.
Piv requires further exploration, then. Among the issues we should consider is whether absences can be perceived. Can you see that someone is not in the room?
Can you see darkness? Can you hear silence? These questions will be the subject of ch. 7, below. We need to understand how thoughts can be about the non-existent (ch. 8). And then we also need to understand how we can believe something not to be the case. Even if we cannot know that something is not, it seems that we can still believe it, and this itself requires an explanation (ch. 11, below). The soft ontological Parmenidean is likely to take at least some of Piv seriously, then, or desire an account that explains the appeal of what Parmenides said.
Pv. Reality is a single plenum it is all one to me (5.1)
Nor was [it] once, nor will [it] be, since [it] is, now, all together, One continuous; (6.5–6) [it] is all full of what-is. (8.24) . . . [it] is completed,
From every direction like the bulk of a well-rounded sphere, (8.42–3)
The idea of reality as a single plenum might be acceptable. Monism has made something of a comeback in recent years, such as when properties are understood as a single interconnected whole (Mumford 2004: ch. 10), or there exists only one thing (Horgan and Potrč 2008), or the one whole has priority over the many parts (Schaffer 2010a, 2010b). The hardline Parmenidean Della Rocca (2020) takes strict monism as the key consequence of Parmenideanism and applies it to a wide range of philosophical areas. The problem, from a Parmenidean point of view, is why that single thing has to be filled at every place: full of what-is and completed. The simplest reason, as we will see below (§3.3), is to think that Parmenides could allow no gaps in reality, as that would mean it included non-being in some places. It is not obvious that such an argument survives in Parmenides’ fragments, but something like this could have been elsewhere, in what didn’t survive (it is quite a philosophical irony that there are parts of Parmenides’ poem that no longer exist: see also Della Rocca 2020: ch. 13). We will see, too, that Parmenides denies the possibility of motion and change, but this is not directly connected with the question of whether what-is is a plenum, since motion is possible in a plenum if parts of it move simultaneously. Even if there are no parts in the plenum, motion might still be possible even for an unvariegated whole (as in the homogeneous spinning disc argument, see Armstrong 1980 and Hawley 2001: 73ff). We cannot know whether Parmenides realized this.
It seems that the soft Parmenidean is at liberty either to accept or reject Pv dependent on their other commitments and what additional arguments are brought to bear. As we will see (ch. 3, below), much will depend on how we regard gaps or holes. Are they essentially non-beings? If they are some kind of negative existent, then we can see why reality would have to be filled at every place. If we can make sense of non-being in a non-reified way, however, then
perhaps we can deny the necessity of the plenum. Likewise, if holes or gaps are not essentially negative.
Pvi. Non-being is unthinkable and unnameable It must be that what is there for speaking and thinking of is; (6.1) Neither from what-is-not shall I allow You to say or think; for it is not to be said or thought (8.7–8)
In an extended passage, Parmenides argues that what is available for thinking and speaking cannot not exist (8.8–18). If we can think about it, then it exists; and if it does not exist, then we cannot think of it or name it. It would appear that there are some startling consequences of this view. One cannot say of some x that it does not exist, since it would come out as self-contradictory (see Quine 1948: 1). It appears that we cannot coherently say that there is no nothing, even though Parmenides appears to have done exactly that.
It might be, then, that we have to reconsider what we can meaningfully say in relation to any kind of non-being, but it is also clear that Parmenides has hit upon a serious philosophical problem, one known as the problem of empty reference. How can one say that Santa Claus wears a red coat, or that Oliver Twist wanted more porridge, or that centaurs have wings, if neither Santa Claus, nor Oliver Twist, nor centaurs exist? How can we say that there are no mermaids or Arctic penguins? There are various types of reference, concerning truth in fiction and myth, negative singular, and general existence claims, all of which would become problematic if no empty terms are allowed.
Much recent philosophy has been devoted to finding a satisfactory answer to questions like these, which I shall consider below (ch. 8). It is clear, however, that Parmenides’ claim is even more wide-ranging than this. It is not just about negative existentials, impossible, mythical, or fictional objects. It applies to all and everything that is not. It rules against any contingent falsehood. Hence, it is supposed to be unthinkable and unsayable that Hillary Clinton is president, since she is not. One would have thought that, even in stating the example, enough has been done to falsify Parmenides’ Pvi. The view might not be so easy to dismiss, however. What we need is an account of negative belief (ch. 11) and probably an account of how it is possible to believe falsely, as Plato considered in The Sophist. Notwithstanding that later discussion, it is clear that the soft ontological Parmenidean has no compelling reason to adopt Pvi on the basis of Parmenides’ arguments. As Gallop (1984: 24) shows, we cannot validly infer from Pi to the view that what does not exist, necessarily does not exist. The latter is needed to show that what does not exist cannot exist (the necessitarianism that I discussed under Pi, above), which in turn, Gallop says, is needed to show that what does not exist cannot be spoken of or thought of. Gallop perhaps assumes here that a merely possible existent can be spoken of and thought of. An argument should
also be given for that, but, it will be shown later, such a view is ultimately acceptable. The consequence of all of this is that in the version of soft Parmenideanism that I defend, Pvi is not included. There is an interpretation of Pvi that would be a more serious matter, however, as we will see below (ch. 8). If the issue is specifically reference, then Parmenides has identified a genuine problem since to refer to x implies that x exists. When x does not exist, we cannot refer to x. But if Pvi is not about reference, then what is it about? (Spoiler: it’s about about.)
Pvii. Neither motion nor change is possible coming-to-be is extinguished and perishing not to be heard of. (8.21)
The above line is the conclusion of an extended argument running from 8.6 to 8.21. Gallop reconstructs it as follows: ‘If there is genesis, then either the subject arose from what-is-not, or it arose from what-is. But it did not arise from what-isnot; nor did it arise from what-is. Therefore there can be no genesis’ (Gallop 1984: 15). What-is is therefore ungenerated and imperishable. Parmenides offers further argument. What-is could not arise from what-is-not because:
. . . how could what-is be in the future; and how could [it] come-to-be?
For if [it] came-to-be, [it] is not, nor [is it] if at some time [it] is going to be. (8.19–20)
Change involves what is not coming to be or what is ceasing to be, both of which Parmenides thinks are impossible. Change means that something comes from nothing, which he has already concluded against with Piii. For example, if we try to suppose that a tomato changes from red to green as it ripens, this means that its redness was nothing but then comes to be something, hence comes to be from nothing. As Gallop (1984: 17–18) states, the argument against motion is obscure and uncertain. As at 8.38–41, it might simply be the argument against change in general that is being applied now to position: one position perishes and another is generated.
Pvii is not a thesis that I would welcome into my soft Parmenideanism. One reason, almost too obvious to state, is that the reality of change is entirely apparent to us and can be known both empirically, through the evidence of our senses, and rationally through our ability to think different things at different times. Any a priori argument that rules such changes to be impossible has thus doubtless gone awry or assumes some even more counterintuitive premise, such as that time itself is illusory. The only question, then, is whether we have given a good account of change or whether, with Heraclitus (Fragments), we should take its reality as a basic assumption of our philosophy so that we can explain other things in terms of it. In a broadly Aristotelian account, there are powers that exist in potentiality but can come to act. A rich conception of reality, as containing both
potency and act, both being existent, at least has the virtue that it can account for change. The act does not come from nothing: it is explained by what there already is. Change need not be a variety of creation ex nihilo, as Parmenides depicts it. Nevertheless, it is clear that Pvii was an important part of Parmenides’ philosophy since he devoted a number of lines to it in ‘The Way of Truth’. To overturn it is to depart some distance from a hard ontological Parmenidean position. Nevertheless, we should be willing to do exactly that. It is the rejection of nothingness that is attractive about Parmenideanism—since that is reasonable—not the rejection of change, which seems unreasonable. And given that the latter need not follow from the former, there is no reason why we have to accept it.
Here, then, is the soft ontological Parmenideanism that I recommend, and will defend at greater length in the chapters to follow. I accept Pi and Pii: nothing is not and there are no degrees of being. That nothing comes from nothing (Piii) and that nothing is unknowable (Piv) are claims that are to be accepted with certain qualifications and interpretations. Whether being is a single plenum (Pv) is something that can be accepted or rejected irrespective of our other Parmenidean commitments and I shall not go into any further arguments concerning these. (For what it’s worth, I have an attraction to some forms of monism and the idea that spacetime is dense, but I will not be arguing for either here. The discussion of holes might bear on the question of the plenum but not directly on monism, as we see in chapter 3.) That non-being is unthinkable and unsayable (Pvi) is something that I reject, as I do the thesis that change and motion are impossible (Pvii).
1.4 Soft methodological Parmenideanism
I now move to a different way of understanding Parmenideanism: one that builds a philosophical methodology aimed at preserving the commitments of Pi and Pii as far as possible while offering principles for evaluating any challenges that they face. This will provide a new way of understanding a distinction between hard and soft Parmenideanism and a way of proceeding with the arguments to come.
Let me call a strong methodological Parmenidean someone who accepts that there is no nothing, that nothing is not something, that everything that exists is ‘positive’, and that there are no negative entities of any kind, and, furthermore, these theoretical commitments for them are inviolable. The final clause is the crucial one since it then follows that in no circumstances would the strong methodological Parmenidean countenance admission of negative entities into their theory of the world or any part of it. (Note that I prefer to call methodological Parmenideans strong rather than hard.)
It is possible to adopt a softer version of methodological Parmenideanism. Such a position would proceed from the basis that commitment to negative existents is at the very least highly undesirable and ought to be avoided if possible. The
