John Searle on Realism

John Searle gives a fascinating lecture on realism in modern philosophy at a conference at the University of Bonn on 26–28th March, 2012.



I think there are two points where Searle goes wrong: his unwillingness to accept that (1) some emergent properties – like the human mind – are so remarkable that they make a new ontological category, and (2), because of (1), we have good reason to consider Popper’s three worlds ontology as a realistic metaphysics.

On Popper’s three worlds ontology, see here and here.

The Argument for Idealism versus Realism

I’ll try once again below to evaluate the argument for idealism, but first properly describe the argument for idealism and in formal terms.

Let us just note that legitimate forms of argument must use the following forms:

(1) Deduction, yielding necessary truth (e.g., various types of syllogism)
(2) Induction, yielding probable truth, and further divided into

(1) induction by simple enumeration;
(2) inductive argument by analogy;
(3) statistical syllogism, and
(4) induction to a particular.

(3) Inference to the best explanation/abduction, yielding probable truth.

One should note that “inference to the best explanation” is frequently considered a form of inductive argument, but some classify it as a sui generis form of reasoning. Whatever the classification one choices, however, the fact remains that it is (1) an important and legitimate form of reasoning and (2) firmly in the same general category as induction as a method of argument that yields non-necessary truth.

The idealist and indeed the realist can appeal to direct personal experience as a posteriori evidence of the following:

Proposition 1: the only things I have direct and immediate experience of are

(i) my own mind, and
(ii) ideas that are presented to the mind, further divided into

(a) ideas over which I have control (imaginations, thoughts expressed in language, etc.) and
(b) ideas over which I have no control (some perceptions of the senses).

So from this the idealist can establish a posteriori the following proposition:

There exist ideas in my mind over which I have no control (perceptions of the senses).

But, furthermore, the idealist or realist can continue and construct an inductive argument, to support the idea that other minds exist.

To do this, an inductive argument by analogy can be used, and such arguments have the following general form:

a, b, c, d all have the properties P and Q
a, b, c all have the property R
Therefore d probably has the property R

where a, b, c, d are things or events, and P, Q, and R are properties of the former things.

Therefore we can construct the following argument:

“a, b, c, d, … n have the properties P, Q, R, S
a has the property of T
Therefore b, c, d, … n probably have the property of T,”

where a = myself
b = a thing that appears to be a person with a mind
c = a second thing that appears to be a person with a mind
d = a third thing that appears to be a person with a mind, etc. to
n = 14th thing that appears to be a person with a mind.

P = speaks language,
Q = seems to have emotions
R = responds to pain
S = can report descriptions of objects seen
T = a conscious mind.

Of course, you could extend the analogy to every person you meet and strengthen the argument, but the induction by analogy is clear: it is probable that the humans we experience who claim to have other minds really do have conscious minds in the same way any individual person who experiences consciousness directly does.

Note that this inductive argument has no necessary truth: it has only probable truth, and certainly does not exhaust the list of possible explanations for our experience of things that claim to be minds like us.

For example, the idealist is also subject to the Cartesian evil demon argument: how can the idealist know that all his perceptions of objects he cannot control are not caused by some malevolent force that is deceiving him? After all, he has already given an inductive argument that it is probable that other minds exist, but that argument has no necessary truth. It might be that only some of the things that claim to be other minds really have minds, or that there are only two minds: myself and the evil demon.

We could list other possible reasons:

(1) an evil demon is deceiving me;

(2) I do have a single mind, but I am dreaming or hallucinating all other minds (by some process internal to my mind that I do not understand), and other minds do not exist.

(3) I am some entity like a “brain in a vat” whose consciousness is given to me by some unknown minds for reasons unexplained.

This does not exhaust the possibilities. You can construct increasingly outlandish ones that nevertheless might be possible as long as they are at least logically possible.

So how can the idealist justify the view that his inductive argument by analogy is the best explanation of the things that seem like other minds, when such minds might not exist and be delusions? How can the idealist (or realist) answer the radical skeptic? This is not a trivial point, but a fundamental one.

It is at this point that we need to invoke the type of reasoning called inference to the best explanation (Harman 1965; Vogel 1990; Scruton 1994: 20), a form of induction or (alternatively) a legitimate non-deductive reasoning like induction.

That is, if you can provide good reasons why your theory offers a better and systemic explanation of the evidence, and show why the rival theories do not (or show that they explain very little at all), then you are rationally justified in choosing the theory that is the best explanation.

First, the evil demon. If the skeptic refuses to give reasons for why the demon caused experience x or y, or why the demon consistently causes us to have belief z, and says merely that the demon just causes them and nothing more, we already have a good reason for believing that the “demon theory” makes for a poor explanation. Even when ad hoc explanations are offered in response to such questions, things are not much better. By contrast, we can construct better explanations for so many experiences involving other minds using the theory that they exist and are not there to deceive us. We can even predict many types of behaviour of other minds based on our theory.

Furthermore, “demon theories” really do have no systemic explanation of why and how the demon deceives. After all, many experiences do not turn out to be obviously deceptive or inconsistent in the way you would expect if some force was really trying to deceive you.

Idea (3) above has less explanatory value than (1), for it openly admits that things happen for unexplained reasons.

If we move to (2) above, we can argue that dreams and hallucinations have different properties than what we call conscious, non-dreaming existence: dreams are usually more fragmented, less clear, and often inconsistent, and lack the constancy we find in “waking” conscious existence. Does the dreaming hypothesis offer a powerful explanatory theory for many aspects of the consistency of waking life, as compared with the hypothesis that waking consciousness contains other minds like ours? Not really.

By inference to the best explanation, one can argue that the behaviour, properties and consistency of our experience of other minds suggest that they do exist and other explanations are inferior on present evidence.

Note how, as we have seen, even the idealist must invoke a non-deductive inference to the best explanation to choose between competing theories and overcome the skeptics.

Now let us proceed to the next proposition that the idealist must accept:

“There are ideas over which I have no control with a causal origin external to me.”

The idealist must accept this, as it follows on logically from his rejection of the sceptical “dreaming” theory above.

For it is possible that, although I exist as a real mind, my entire conscious life might be a dream generated by some internal process inside my mind that I do not understand. Why? The reason is that it is clear, by direct personal experience, that we do not understand many aspects of our cognition and thought (e.g., why do we make Freudian slips, or sometime forget things we easily remember on other occasions?). There is no reason a priori why our latter “dreaming” explanation is necessarily false.

So the idealist is committed to the view that some external cause is the explanation for those ideas over which we have no control (e.g., constant perceptions of the senses).

How does he provide a rational justification for this? Again, it is by inference to the best explanation: the other theories positing a mere unknown internal cause of the ideas over which we have control are judged to be inferior explanations.

We now have three propositions:

Proposition (1): There are ideas over which I have no control (perceptions of the senses).

Proposition (2): It is probable that the humans we experience who claim to have other minds really do have conscious minds in the same way any individual person who experiences consciousness directly does.

Proposition (3): There are ideas over which I have no control with a causal origin external to me.

But notice how these propositions can be accepted by any realist too, who will have used the same type of arguments to defeat the Cartesian skeptic. It is at this point that the idealist and indirect realist part company.

For the question is now:

What is the external cause which is the causal origin of those ideas over which we have no control?

Some of the earlier sceptical theories are still relevant. But the list of possible explanations or theories is larger.

If we have given an inductive argument for the probability of an external cause, then no logically possible external cause can be ruled out a priori.

Here the skeptic will demand that his theories at least be considered as possibilities, but also the theistic idealist, non-theistic idealist, realist and panpsychist will add other theories too:

(1) an evil demon is the external cause (skeptic);

(2) we are “brains in a vat” whose consciousness is given by some unknown minds for reasons unexplained (skeptic);

(3) a god of classical theism is the mind that is external cause (theistic idealist);

(4) a “super-mind” that is not the god of classical theism is the external cause;

(5) we live in a purely mental world, but every object of our perception is invested with a mind and hence there are multiple external causes for all constant objects we perceive (a variant on panpsychism);

(6) there are non-mental objective external objects that are the causal origin of our perceptions of them (indirect realist).

Other possibilities could be added of course, but these make a useful list.

Can the idealist rule out any of these other possibilities a priori without simply begging the question by already assuming idealism is true? I do not think so.

All must be regarded as at least possibilities.

And at this point it is induction and inference to the best explanation that must be invoked to decide which one is true.

The idea that the indirect realist does not use legitimate logic or induction is untrue, for we need only consider the induction by analogy arguments that can be invoked by the realist or the idealist:

(1) the idealist:
We observe directly things that claim to be minds that probably have minds (e.g., people). On analogy, there might be an unobserved super-mind that is the external cause.

(2) the realist:
We observe directly in our minds objects that appear to have no minds and probably do not have minds (e.g., tables, chairs, rocks, books). On analogy, there may be unobserved non-mental objects that are the external causal origin of our perceptions of these things.

These are both inductions by analogy: in this step, the realist and idealist are on the same epistemological ground.

But who is right? We now need to move beyond this to inference to the best explanation.

Does the idealist theory of the super-mind explain the vast set of experiences and evidence we see around us?

E.g.,

(1) what are fossils under the idealist theory? Presumably they are mere ideas that existed in the super-mind in the past.

(2) but this just raises the question: why did the super-mind create in its thought a world 541 to 485 million years ago devoid of human minds with Cambrian aquatic animals and plants? To what purpose?

For some unexplained reason the super mind imagined all these animals. Then thought of the idea of “mass extinctions.” But then thought of vast numbers of new animals in later eras, only to cause more “mass extinctions” by its ideas. Why?

The questions that the idealist cannot give coherent answers to multiply to huge numbers quickly:

(1) did dinosaurs have minds?

(2) when did animals first have real minds? Did the “mass extinctions” cause suffering to animals if they had minds?

(3) why did the super-mind “destroy” the dinosaurs?

(4) why are some humans colour blind?

When pressed, there is little the idealist can say in response precisely because his theory has poor explanatory power. These questions remain mysterious puzzles.

But, as I have shown in the last post, the indirect realist materialist has an interpretation of the natural sciences that has great explanatory power and can give coherent and powerful explanations of all these questions and thousands more, and is superior to the idealist theory.

The universe really exists as a material external world and requires no god or mind for its existence and operation. Nor will it do for the idealist to say that idealism just follows natural science in its explanation of the world but stripped of matter: for natural science says the universe is an independent entity and has autonomous natural laws that do not require any god or super-mind to create events and objects in it. That is the whole thrust of modern science since Newton!

All life on earth exists in an external world and evolved by Darwinian natural selection: the result of (1) chance mutation or combination of real DNA and (2) the deterministic process of survival of fitter animals.

DNA is a real external object that creates animals by genetics and biochemistry.

And of course once realism is accepted it is but a short step by inductive evidence to the probable conclusion that minds are causally dependent on brains.

There is a vast research literature about animal minds: they are dependent on the complexity of brains that really exist inside the organisms as external objects in the real world. We have a probable inductive theory about why the dinosaurs died out: a chance contingent event caused their destruction when an asteroid hit the earth. Precisely the type of chance event that can happen in a real external universe like ours with real asteroids flying around our solar system.

Human beings experience colour blindness because of genetic defects and disease, and so on, and so forth.

In short, an inference to the best explanation supports realist materialism as the most powerful explanatory theory. It is rational, powerful and supported by evidence.

BIBLIOGRAPHY
Harman, Gilbert H. 1965. “The Inference to the Best Explanation,” Philosophical Review 64: 88–95.

Moser, Paul. 1990. “Two Roads to Skepticism,” in M. D. Roth and G. Ross (eds.), Doubting: Contemporary Perspectives on Skepticism. Kluwer, Dordrecht. 127–139.

Scruton, R. 1994. Modern Philosophy: An Introduction and Survey. Penguin Books, London.

Vogel, Jonathan. 1990. “Cartesian Skepticism and Inference to the Best Explanation,” The Journal of Philosophy 87.11: 658–666.

More Problems with Idealism

By “idealism” I refer to either

(1) George Berkeley’s (1685–1753) theistic subjective idealism (or immaterialism) that invokes a god of classical theism as the “greater mind” that creates both finite minds and the ideas given to minds, or

(2) a non-theistic idealism that is agnostic about what the “greater mind” actually is.

It is interesting that Berkeley’s idealism was clearly directed both against (1) the radical skeptics of his day and (2) the materialist atheists (Grayling 2005: 166–167).

For the empiricists of Berkeley’s era had opened up profound epistemological problems by stressing the differences between

(1) appearances (our conscious experiences and sensations), and

(2) reality (some posited external world).

In the terminology of Berkeley, there is thus a gap between (1) human “ideas” and (2) “things” (of the external world) (Grayling 2005: 168). Our internal sensations (or ideas) do not give us direct access to the external world, so how can we have knowledge of it? The latter was the question asked by radical skeptics.

Berkeley’s answer was to simply argue that the material “things” of the external world do exist (Grayling 2005: 168): only ideas and minds exist. Therefore we have direct experience of ideas – the only things that exist apart from minds.

Under Berkeley’s idealism, god is the external cause of finite human minds and the consistent and non-illusory ideas we have. For, given that there are many consistent ideas not under our control, Berkeley admits that the order seen in such consistent ideas must have some systematic cause: hence it is god (Mackie 1982: 71).

One of Berkeley’s maxims is the idea that “to exist is to be perceived” (or in Latin esse est percipi), and thus our existence and the ideas of our conscious life are causally dependent on god’s “perception.” Thus, for example, Berkeley accepts that the table in your kitchen still exists when nobody is in the room perceiving it, because it is still perceived by god.

By contrast, the non-theistic idealist does not necessarily need a god of classical theism, just a “super-mind” or “uber-consciousness” that is the cause of our finite human minds and our ideas. In what follows, I will refer to the non-theistic idealist’s greater mind as the “super-mind.”

I argue that these are crucial points in any debate with idealists:

(1) Epistemologically speaking, the idealist is in the same fundamental position as the indirect realist materialist: the idealist posits an external and directly unobserved cause of our given consistent ideas and perceptions. For neither (1) the external world of realism nor (2) the god or “super-mind” of idealism is directly observable or directly experienced in immediate conscious life.

Now I stress that this is not necessarily a problem for the idealist, but it is an important observation that means that in this respect the idealist position is not epistemologically superior to the realist one;

(2) Epistemologically speaking, the idealist – unless he can provide a valid and sound deductive argument or arguments – must use inductive arguments on the basis of indirect evidence to prove his hypothesis of a god or “super-mind.”

But that means, epistemologically speaking, that the idealist and the realist are also on the same ground, since the indirect realist’s belief in an external world is also an inductive inference on the basis of indirect evidence.

So again in this respect the idealist position is not epistemologically superior to the realist one, but on a par;

(3) Berkeley’s argument that “an idea can be like nothing but an idea” is unconvincing;

(4) Berkeley’s theistic god is a clear weak point in this theory, and it is not a simpler theory than realist materialism;

(5) on many issues, idealism of any type has weak explanatory power and often no explanatory power at all.

I think I have already explained (1) and (2) above, so let us proceed to (3), (4) and (5) below.

(3) An idea can be like nothing but an idea?
Berkeley makes this argument in The Principles of Human Knowledge, section 8.

J. L. Mackie shows why the indirect realist position is not refuted by Berkeley:

“… that an idea can be like nothing but an idea, collapses when we see what sort of likeness or resemblance the materialist view requires. This is just, for example, that when I am in a perceptual state whose content is describable by saying that I seem to be seeing something square, there should be something that is (approximately) square. Or again, that when it is with me as if I were seeing two objects of similar shape and size, say two billiard balls, there should really be two similarly shaped and sized objects in the neighbourhood. The likeness or resemblance required is just that where the content is as of an X, the reality should be (roughly) an X. There simply is no a priori impossibility or even implausibility about something other than an idea being like an idea in this sense.” (Mackie 1982: 72).

Perfect or exact resemblance between our sensory states and objects of the real world is unnecessary for the indirect realist position.

All we need is (1) an approximate similarity and/or (2) a relationship of causal dependence from objects in the external world running to some of our sensations, which does not exclude that, in a modern scientific sense, the object of the external world that is the causal origin of some perception has many properties not observable to us.

(4) Berkeley’s theism is fraught with difficulty
Berkeley’s idealism is based on a theistic conception of human minds: that humans are created by a god, and (presumably) fundamentally different from animals.

First, a problem here is: do animals have minds? How does the idealist offer an answer to this question? It is not clear what Berkeley thought (Cummins 2005: 190).

Secondly, as we have seen, for Berkeley god is the external cause of the regularity in many of our ideas.

So Berkeley needs to prove the existence of god. But the history of attempts to use deductive arguments for the existence of god is one of utter failure (and I rely on Michael Martin’s [1990] survey and critique here), so Berkeley’s argument must fall back on inductive arguments.

The problems with Berkeley’s actual argument for god are well known (Mackie 1982: 64–80). For one thing, the cause of the consistency of the ideas beyond our control might not be unitary. Why can’t it be explained by many “super-minds” or panpsychism? (Mackie 1982: 71).

Even if we put all those difficulties aside, there is still a severe problem: the hypothesis of a theistic god is much more complex and more extravagant than a realist materialist hypothesis.

Berkeley, of course, argued that his idealist hypothesis with a god was the simpler explanation invoking Occam’s razor. But this contention is unconvincing.

Consider the attributes of the god of classical Judeo-Christian theism, with additional ones required by Berkeley:

(1) god is a personal and conscious mind;

(2) god is omnipotent;

(3) god is omniscient;

(4) god is perfectly good;

(5) god constantly maintains the existence of finite minds and their ideas in a process we can’t understand but which appears miraculous.

I submit to you that these attributes make the god invoked by Berkeley an extremely complex entity, possibly the most complex thing you could postulate.

In fact, there is still debate about whether properties like omnipotence, omniscience, and perfect goodness in one being can even be logically possible (Martin 1990: 278–316).

By contrast, look at the attributes of a real physical universe as postulated by science:

(1) the external universe of matter and energy is not conscious nor it is a personal entity, and in this respect is a simpler entity than a theistic god;

(2) following from (1) the universe is not omnipotent, omniscient, nor perfectly good, and in this respect is a simpler entity than a theistic god;

(3) the universe consists of matter and energy and its relations governed by a limited set of regularities we call “physical laws,” which can be discovered by humans through inductive and experimental science. Thus the external universe is a simpler entity than the miraculous powers of a theistic god.

(4) all complex things in the universe like life and minds have been built up and created over many billions of years by relatively simple and well understood processes like Darwinian evolution by natural selection. The existence of minds does not require some inexplicable miracle, but is explicable under a realist materialist view.

One can add that the realist materialist view also explains evil and suffering in a far better way than theistic idealism, amongst its many merits.

In short, the realist materialist position is the simpler and better explanation – and, under the principle of Occam’s razor, the preferable one.

Furthermore, even Berkeley’s master argument, if it is really taken to be a devastating argument against the existence of a real external world of objects, would apply to other minds and to the “super-mind”:

“Secondly, if the [sc. Master] argument were sound, it would prove too much. Although … [sc. Berkeley] states it with regard to sensible objects like houses and books, their nature as sensible objects plays no part in the argument; so, if it held at all, it would hold equally against the suggestion that I can conceive that there are minds that are not conceived by me; thus it would establish solipsism, in which I would reject as not even conceivable the existence not only of material things but also of other minds and spirits, including God, except in so far as they are conceived by me. This line of argument is, therefore, not only fallacious but also, in these two ways, misdirected.” (Mackie 1982: 68).

That is, think of a mind that you do not perceive now. There is a vast amount of empirical evidence that other minds exist even though you do not perceive them.

But if we cannot conceive of a mind without it being perceived, it follows that in any initial thinking through idealist philosophy, you have no good reason to think any minds exist that you do not immediately perceive. But, says Berkeley, god perceives them, so they do exist. But Berkeley’s argument works just as well against god or the “super-mind”!

If we have no direct perception of the “super-mind,” who perceives the “super-mind” and causes its existence? And who perceives the “super-super-mind,” and so on?

That is, Berkeley’s philosophy must cut off an infinite chain of explanation by positing “god” or the “super-mind” as a “brute fact” that is inexplicable. But, in that respect at least, it is no better than a materialist viewpoint that posits the universe as the ultimate “brute fact,” the difference being, as we have seen, that the hypothesis of a real universe is simpler and by Occam’s razor a better hypothesis.

(5) Realist Science has great explanatory power and Idealism does not
Our natural sciences have produced a vast set of theories that explain the world we experience, and under the indirect realist interpretation we can posit an external world of real objects of matter and energy affected by forces.

As just a sample, consider the following theories:

(1) the Big Bang theory;

(2) the theory of the development of gases, stars, planets and galaxies;

(3) the theory of the origin of life by biochemical processes;

(4) the theory of Darwinian natural selection acting on self-replicating biochemical compounds and living things, producing a great diversity of life over many billions of year on earth;

(5) the theory that human beings evolved from lower life-forms and that various types of minds too evolved gradually and incrementally as a product of evolution.

If one takes an idealist view, whether Berkeley’s or a non-theistic idealism, none of these theories can be literally true, for there is no external word of matter and energy.

Minds were created by god or the “super-mind” and did not arise by gradual material processes.

But we have a vast amount of empirical evidence available to our senses that can be used in inductive argument to support the truth of these theories.

Our realist materialist science has a strong explanatory power explaining the wealth of data we see, but idealist philosophy – whether of Berkeley or a non-theistic idealism – has weak explanatory power and often no explanatory power at all.

This point is very easy to prove, I think.

We need only think of fossils. Our conscious life presents to us things we call fossils, and by inductive inference we infer that they are the remains of living things long dead. A realistic materialist science explains what fossils are, and indeed can use them to reconstruct a history of life on earth: things that really existed in an external world, including a specific long line of animal life forms from which human beings are descended.

But what explanation can an idealist offer for fossils? As we have seen the idealist faces difficulties answering the question whether higher animals even have minds. If animals do not have minds, then they must be mere “ideas” given to the mind by god or the super-mind. I feel certain that no idealist would content that plants have minds, so how does the idealist explain plant fossils?

What explanation can the idealist give under his theory, except to say it is another mystery?

And why has god or the super-mind given us a wealth of sensory ideas that lead us to inductively infer that our “world” once had living animals and plants now dead? Is god or the super-mind engaged in deception and fraud? If so, why?

Or take another example: why is it that some people have different perceptions of colour? (e.g., colour blindness). Why does god or the super-mind that create minds with different types of colour perception?

A realist materialist gives us a clear explanation: colour blindness is either (1) genetic and a problem with genes or (2) caused by certain diseases.

Again and again, I contend we can find endless puzzling problems from our conscious life that realist materialist science gives a powerful explanation of, but which cannot be adequately explained by idealism.

Update
Philip Pilkington has a response to me on idealism here:

Philip Pilkington, “The Mystery of Matter: A Response to Lord Keynes on Berkeley’s Idealism,” Fixing the Economists, February 27, 2014.
http://fixingtheeconomists.wordpress.com/2014/02/27/the-mystery-of-matter-a-response-to-lord-keynes-on-berkeleys-idealism/

BIBLIOGRAPHY
Cummins, Phillip D. 2005. “Berkeley on Minds and Agency,” in Kenneth P. Winkler (ed.), The Cambridge Companion to Berkeley. Cambridge University Press, Cambridge and New York. 190–229.

Dicker, Georges. 2011. Berkeley’s Idealism: A Critical Examination. Oxford University Press, Oxford and New York.

Grayling, A. C. 2005. “Berkeley’s Argument for Immaterialism,” in Kenneth P. Winkler (ed.), The Cambridge Companion to Berkeley. Cambridge University Press, Cambridge and New York. 166–189.

Mackie, John Leslie. 1982. The Miracle of Theism: Arguments for and against the Existence of God. Clarendon Press, Oxford.

Martin, Michael. 1990. Atheism: A Philosophical Justification. Temple University Press, Philadelphia.