Possible temporal counterexamples to Berkeley’s Master Argument

In Three Dialogues Between Hylas and Philonous, George Berkeley questions the common belief that there exist things independent of the mind. I will focus on his Master Argument. First, I will discuss the concepts of mind-independence, conceivability, and conception. These concepts are analyzed in a rather austere fashion in an attempt to pinpoint exactly what is being referred to. After this, a quick formulation of Berkeley’s Master Argument suffices for an understanding of it. Then, I will insert a temporal element into the argument, and use it to propose three types of possible counterexamples. It will be shown that while two of these types fail, one of them succeeds in refuting the Master Argument.

The issue Berkeley addresses in the Master Argument part of the Dialogues is whether there is any good reason to believe that mind-independent things exist. What does ‘mind-independent’ mean? It describes things outside of the mind, the existence of which do not depend on the mind, such that they can exist by themselves, “independently of an unperceived by any mind whatsoever.”[1] For example, my ideas are in my mind and dependent on it, so my ideas are not mind-independent things. If you believe that there exist trees outside of your mind that would exist even if your mind stopped functioning, then you believe trees are mind-independent things. Berkeley would say that your belief is flawed, as we see later.

A technicality of my formulation of mind-independence above is that by ‘things’, I did intend to have a broad scope; in general, things could be ideas in the mind such as thoughts, or material objects such as trees. But for mind-independent things in particular, could they be ideas or material objects? I leave the question on material objects undecided for now, as it is the subject of the Master Argument. The question on ideas can be answered now. There cannot exist mind-independent ideas because ideas are things that exist in and depend on the mind. It is impossible to conceive of ideas that just ‘float around’ without a mind to conceive of them; even if you somehow claim that you can, then you must think ideas are mind-independent things by virtue of being possibly unconceived (even if you are conceiving of them), and the Master Argument applies.[2] Since I have just ruled out ideas as possible mind-independent things altogether, you may think that we should specify the terminology from the broader “mind-independent things” to the narrower “mind-independent material objects.” However, the second term is actually redundant: material objects are, by the intended reference of common usage, supposed to be mind-independent. When we talk about material objects such as trees, we try to refer to the trees “themselves” rather than any associated mind-dependent things such as our ideas of the trees. Thus, it suffices to refer to them as “material objects” alone. Even so, I shall keep using the term “mind-independent things” to keep the meaning explicit and because “material objects” can misleadingly connote the existence of them.

So much for mind-independence. The question remains about whether there exist mind-independent things. If there are such mind-independent things out there, then intuitively we should be able to conceive of (think about) them. In other words, if they aren’t conceivable (perceivable, or thought about),[3] then there doesn’t seem to be any reason to believe that they exist. Let’s analyze the concepts of conception and conceivability. Berkeley doesn’t seem to give definitions for these concepts, so I will attempt to formulate them based on my understanding (conception) of his argument. For something to be conceived at some specific point in time, there must be an idea of it at that time. For something to be conceivable, it must be possible for there to be a (coherent) non-contradictory idea of it. Conceivability is a timeless concept; it can be determined a priori. For example, I can conceive of the idea of a tree that is (exactly) ten feet tall, so the idea of a tree that is ten feet tall is conceivable. When I actually do conceive of it, such as right now, the idea of a tree that is ten feet tall is (being) conceived. However, I can’t conceive of the idea of a tree that is not a tree, which means the idea of a tree that is not a tree is inconceivable. So this idea is always unconceived.

As a first technicality about my formulation of conceivability, you may wonder why conceivability seems to be more about the content of the conception rather than the mental process of conception itself. That is, my formulation of conceivability relies more on the concept of non-contradictory ideas than what actually goes on in the mind. I don’t see (conceive of) any problem with this. It may very well be the case that conceivability is more about the content than the process. After all, conceivability is a difficult concept to grasp (conceive of), more so than the content, which is the entire purpose of conception. So, treat this as a functional formulation of conception.

As a second essential technicality, note that in my example above, I talk about conceiving of ideas such as ideas of trees, rather than conceiving of material objects such as trees themselves, which are mind-independent things. This is because we are not sure mind-independent things exist yet, so it is not a good idea to refer to them as if they exist. When I use words like “you” in this paper, I am actually referring to my idea of you, the reader, though of course it would be tedious if I spelled this all out.[4] So, I restrict the use of this terminology to concepts immediately relevant to Berkeley’s argument. Here, we have that the conception of an idea involves the idea of that idea, because conception itself means having an idea of its content. In fact, we could use “ideate” as a synonym for “conception,” but that would be harder to read.

Finally, relevant to this talk about ideas of ideas is the widely-held claim that if something is conceivable, then it is possible that it exists. The claim basically follows from my formulation of conceivability. If something can be thought about, then it is non-contradictory and hence possible that it exists, even if the actual chance that it exists is conceivably low. After all, it is the content of some idea, which exists, so there’s nothing inherently wrong with thinking that it could exist in the mind or somewhere else if possible. Conversely, however, it is unclear that what is possible is necessarily conceivable. Still, if it is not possible to conceive of something, then this should be a major concern to any belief that the thing exists. For example, since I can conceive of the idea of a tree that is ten feet tall, it is possible for that idea to exist. Since the idea of a tree that is not a tree is inconceivable, this suggests that that idea does not exist. Berkeley applies this claim to mind-independent things to show that there is no good reason to believe in their existence.

The Master Argument, as hinted above, uses the claim that mind-independent things cannot be conceived. Since conception is a mental activity, what is mind-independent must be able to exist unconceived. Otherwise, if what is mind-independent is always conceived, then it is always in some mind, which does not show at all that it is mind-independent. Berkeley challenges us to conceive of a mind-independent thing like this, that is able to exist unconceived. So, we must conceive of this thing as being unconceived. In less words, we must conceive of the unconceived, which is a contradictory idea. By my formulation of conceivability, there cannot be a contradictory idea in any conception, which means it is inconceivable to conceive of the unconceived. This means we can’t conceive of the unconceived, a failure to meet Berkeley’s challenge. Berkeley concludes from this line of thought, without so many steps, that “you can’t conceive how any corporeal sensible thing should exist otherwise than in a mind,”[5] which implies that there is no good reason to believe in the existence of mind-independent things.

Now that the Master Argument is explained in terms of my formulation of the relevant concepts, I will consider possible counterexamples to it. If I can show that I can conceive of something being unconceived, then that thing serves as a counterexample to refute it. To clarify the Master Argument, let’s add a temporal element to it, as it is ambiguous about the time aspect of conception. To match Berkeley’s challenge, we must conceive of something that is unconceived… right now, or at some other point in time? For the former, it seems impossible to conceive of something as being unconceived right now; as soon as I conceive of it, it cannot be unconceived at the same point in time. For the latter, maybe I am able to conceive of a thing that was unconceived at some time other than right now. Berkeley seems to consider only the former case, but it is worth considering the latter in case it is relevant. I shall propose three types of such “temporal counterexamples.” The first two fail to actually be counterexamples, but they serve as motivation for the third one, which succeeds in refuting the Master Argument.

As a first possible temporal counterexample, it is reasonable to assert that the idea of a computer was unconceived back in the early centuries. However, this does not refute Berkeley’s argument. So what if the idea of a computer was unconceived back then? It didn’t even exist anyway! If the idea of a computer did not exist, then there is no good reason to believe that a computer itself existed. As a technicality, you may wonder why I talk about conceiving the idea of a computer here, rather than conceiving of a computer itself. As neither were conceived, it makes more sense to talk about conceiving the idea of a computer because I don’t even know what a computer is at this point, though I certainly have the idea of a computer. Berkeley runs into trouble by talking too much in the common language of his opponent, which gives the wrong impression about what he means. If you are not satisfied with this explanation, you could re-read the beginning of this paragraph by replacing “the idea of a computer” with “a computer,” and the reasoning would remain valid. So, conceiving of the idea of anything being unconceived before it existed does not pose a threat to the Master Argument. This applies also to the conception of the ideas of things unconceived in the future before they exist, if that’s possible. These form an infinite number of ideas, constituting the first type of possible counterexample.

As a second possible temporal counterexample, I can conceive of a world in which there are no minds, so nothing can ever be conceived within that world. Suppose I can somehow have the idea of there being things inside that world. Then those things are unconceived in that world and therefore, technically mind-independent. But (the idea of) a world without any minds is not at all like (the idea of) the world I am in, which contains a mind, namely my mind. The idea of that world containing mind-independent things does not imply in any way that (the idea of) my world contains mind-independent things. Also, given this possible counterexample it is still not right to say that mind-independent things exist anywhere because those things in that world are ultimately my ideas, and I’ve already shown how ideas cannot be mind-independent things. Even though those things are unconceived in that world, they are conceived by my mind right now, and the claim about them existing anywhere must consider whether they exist in both that world, my mind, and wherever else. In general, this reasoning applies to the conception of ideas that contain unconceived things. As another example, how I manage to discuss unconceived things in this paper without having to commit to their existence is that I am conceiving not of them, but of ideas of them. It is important to distinguish the different levels of conception. Ideas of ideas of unconceived things constitute this second type of possible counterexample.

As a third type of possible temporal counterexample, let’s consider a more ‘recent’ version of the first type. This involves conceiving of something as being unconceived at some time other than right now, but while it exists. I can conceive of a computer at a time shortly after it was created, but before many people knew about it. Reasonably, it is possible that the computer was unconceived at that time. Suppose (conceive that) it was unconceived at that time. There’s nothing contradictory about this conception because my conception of the computer right now has no effect on its being unconceived at that time. This temporal separation allows for this possible counterexample to escape the Master Argument. Since the computer existed at that time, and it was unconceived at that time, this shows that it was a mind-independent thing at that time. This seems like a suitable counterexample to Berkeley’s challenge, except for the extra temporal element. Now I show that the extra temporal element disappears if we analyze what it means for the computer to be mind-independent at that time.

Does it mean anything to say that the computer was mind-independent at some time only? It is reasonable to assert that the computer is either mind-dependent or mind-independent at every point in time while it exists. Berkeley may be forced to admit that the computer was mind-independent at that time by my above reasoning, but he may try to maintain that it is mind-dependent right now because I am conceiving of it right now. If he tries to maintain this, he must then admit in general that a thing is mind-independent every time it is unconceived, but mind-dependent every time it is conceived. However, the mind-independence of a thing isn’t conceivably the sort of property to change frequently. I claim that if something can be shown to be mind-independent at some time only, this suggests that it is mind-independent for all time. My argument for this is as follows.

In the third possible counterexample, the computer being mind-independent at that time means it had no such connection to the mind at that time, so why would it suddenly have that connection as soon as I conceive of it? To account for sudden changes in the mind-independence or mind-dependence of everything that is conceivable, Berkeley must provide some sort of theory in addition to the Master Argument. I cannot conceive of such a theory; it is inconceivable that a thing that is mind-independent at some times necessarily becomes mind-dependent at all other times as soon as it is conceived. Since this is inconceivable, there is no good reason to believe in it. A good reason to not believe in it is that conception alone does not change any property of its content, other than its conception. For example, when I conceive of the idea of a tree, the idea itself does not change, other than perhaps a change in property from “unconceived” to “conceived.” This does not imply a change in another property, from “mind-independent” to “mind-dependent.” After all, we know that the idea of a tree can’t be mind-independent from my earlier formulation of mind-independence. This example of the idea of a tree shows that a thing can go from “unconceived” to “conceived” without going from “mind-independent” to “mind-independent.” This ends my argument that something being mind-independent at some time suggests that it is mind-independent for all time. Now I conclude based on this that the third possible counterexample is an actual counterexample that refutes the Master Argument.

Therefore, by the third possible counterexample, while I can’t conceive of anything being unconceived (and hence mind-independent) right now, I can conceive of things being unconceived (and hence mind-independent) at some other time, which suggests that that such things are mind-independent all the time. This is a successful counterexample in showing that mind-independent things exist.

As the final part of my analysis, I compare this third type of counterexample with the two previous ones to show how it succeeds where they fail. The first type failed because the computer didn’t actually exist at the time conceived. The third type rectified this issue by conceiving of the computer at a time at which it actually exists. This may sound like the existence of the computer is presupposed, but there is nothing inherently contradictory about conceiving of the existence of this particular thing, so the Master Argument does not prevent such a presupposition (recall that what is conceivable is at least possible). The second type failed because of its indirectness in conceiving of the idea of a thing that is unconceived (right now). The third type rectified this issue by conceiving the thing that is unconceived (at some other time) itself, not just the idea of it. This makes it so we can talk about the content of this conception directly.

The three types of possible temporal counterexamples proposed above are intended to cover the intuitive notion of the kinds of things that are possibly unconceived. It has been shown that one of them is a successful counterexample in showing the existence of mind-independent things (including material objects). My formulation of the concepts used in the Master Arguments was thorough to try to describe (my conception of) Berkeley’s intention in order to adequately lead to this assessment.

Notes

  • [1] Berkeley, 429.
  • [2] You may note that the idea of this sentence is an application of the Master Argument itself. I had originally begun the sentence with “ideas can’t just ‘float around’…,” but that would more likely raise the question of how I know that.
  • [3] Berkeley seems to use terms related to ‘conceive’ and ‘perceive’ interchangeably, which does not affect his argument. Szabo (466) and Priest (61) agree. Atherton (138) disagrees.
  • [4] You may object to this by claiming that “you” and “my idea of you” are entirely different things, that I am confusing between the two. You are right that they are different, which is why I use “you” to refer to my idea of you, and “my idea of you” to refer to “my idea of my idea of you,” etc. Where the terms are not placed inside quotation marks, the reference is intended to be obvious by context.
  • [5] Berkeley, 430.

References

  • Atherton, Margaret. The Empiricists: Critical Essays on Locke, Berkeley, and Hume. Rowman & Littlefield, 1999.
  • Berkeley, George. Three Dialogues Between Hylas and Philonous. In Ariew, Roger and Eric Watkins. Modern Philosophy: An Anthology of Primary Sources. Hackett, 1998.
  • Priest, Graham. Beyond the Limits of Thought. Oxford University Press, 2002.
  • Szabo, Zoltan Gendler. “Sententialism and Berkeley’s Master Argument.” The Philosophical Quarterly 55 no. 220 (July 2005), 462–474. http://www.blackwell-synergy.com/doi/pdf/10.1111/j.0031-8094.2005.00411.x