Misuses of Meaning
A communist, a bishop, and Sam Harris walk into a bar...
A warm beer is better than a cold beer. After all, a warm beer is better than nothing, and nothing is better than a cold beer.
- unknown
“Arguing semantics” is often considered a distraction – an alternative to substantive discussion. In a sense, this understanding is entirely sensible. If we think of the meaning of a word as something agreed on by a language’s speakers, debating semantics is no more than a failure to agree, and “winning” the argument is no more than agreeing to phrase claims in the way you like best – it reveals no facts about the world. But this is not the whole picture. There are fundamental questions within the field of semantics – what are meanings, and where do they come from? When we say a claim is true “in virtue of its meaning,” what is the significance of this, and how can we check? To what extent are longstanding philosophical disputes attributable to semantic confusion?
If you care about the truth, these are important questions. Our beliefs, claims, facts, questions, and arguments all rest on a tacit semantic philosophy – an unspoken and infrequently-examined understanding of how meanings should be navigated and manipulated. Our other avenues of philosophical inquiry – metaphysical, ethical, epistemological, political – depend on this underlying architecture of meaning, perhaps to a surprising extent. If you want to know the truth of a claim, you must first understand what it means. And if you want to explain truth itself, you must first be able to explain meaning itself to at least some satisfactory extent.
The aim of this article is to illustrate the importance of a robust theory of meaning by setting out some examples of prominent thinkers whose arguments are weakened by confusion on this precise point. In future articles, I will paint a more positive picture – setting out existing theories of meaning and identifying common threads, persuasive elements, and opportunities for synthesis. But since the need for such a theory is by no means obvious in the first place, I think it is worthwhile to start with these “problem cases.”
A. The Problem of Substantive Definition
The fallacy of equivocation is a failure, within an argument, to indicate that the same word is being used with multiple meanings. This is an easy mistake to make, and one which can be difficult to identify. Philosophers accuse each other of this constantly. The simplest form of equivocation is to make two statements, which use the same word in two different ways:
Noisy children are a real headache. Two aspirin will make a headache go away. Therefore, two aspirin will make noisy children go away.1
A particular form of equivocation, a practice I will be referring to as “substantive definition,” occurs when a claim is made based on a word’s ordinary intuitive meaning, but given a formal “definition” which conceals this:
ALICE: The government shouldn’t collect taxes. Taxation is theft!
BOB: How’s that? It doesn’t seem like theft to me.
ALICE: By theft, I just mean ‘property changing hands without consent.’ Taxation is theft by definition.
Alice’s argument hinges on the tacit premise that theft is bad. The average person does, intuitively, endorse the claim “theft is bad” as a general principle – but the average person does not employ Alice’s definition of theft. For the average person, if taxation does not count as theft, it is perhaps because theft’s badness is part of its meaning – if we consider some morally non-objectionable instance of “property changing hands without consent” (for instance, capturing weapons in the course of a defensive war, collecting on a debt, repossessing stolen goods, paying a speeding ticket, and so on), the average person is far and away more likely to say “that’s not theft” than to say “sometimes, theft is fine.”2
With this approach, Alice saves herself the hassle of demonstrating that “theft,” as defined, is in fact wrong. There may well be an excellent argument for this – but Alice hasn’t made it! Substantive definition is quick and cheap. It’s effective in debate clubs and courtrooms, but occasionally we find arguments along comparable lines in philosophical works as well – as a way to get, basically, “one fact for free” by way of definition. I don’t care for this.
B. Harris and Goodness
In Sam Harris’s The Moral Landscape, he defines “good” as “that which supports well-being.”3 This strikes me as an unfortunate choice, because “it is good to support well-being” is a substantive claim - indeed, the core claim of the entire book! It is not remotely a matter of the term’s meaning. This is a substantive definition. It may well be the case that wellbeing is good – it may even be the case that an account of wellbeing is, as a matter of fact, an exhaustive account of goodness. But if we define “good” as “that which supports well-being,” various uncontroversial claims like “good things should happen,” “people should be good,” and so on, become thorny and difficult to demonstrate, using only that definition. These claims, I’d say, are true in virtue of the meaning of the word “good” – this is precisely why they are so uncontroversial. Definitions are supposed to be boring and uncontroversial in this sense. We should not consider ourselves able to arrive at facts about the world by definition, because our definitions are a linguistic choice, and facts certainly are not.
Harris accurately notes that Moore’s open question argument is aimed at this precise question – of whether goodness can be defined in terms of some other thing. I don’t quite agree with his characterization of the argument, however:
Moore argued that goodness could not be equated with any property of human experience (e.g., pleasure, happiness, evolutionary fitness) because it would always be appropriate to ask whether the property on offer was itself good. If, for instance, we were to say that goodness is synonymous with whatever gives people pleasure, it would still be possible to worry whether a specific instance of pleasure is actually good.4
By Harris’s account, the open question argument is a sort of skepticism that attacks the certainty of claims about the nature of goodness. But by my reading, this is not actually what Moore is criticizing! Moore is making a methodological point – criticizing prior ethical naturalists’ tendency to engage in this precise form of substantive definition!
When A says ‘Good means pleasant’ and B says ‘Good means desired,’ they may merely wish to assert that most people have used the word for what is pleasant and for what is desired respectively. And this is quite an interesting subject for discussion: only it is not a whit more an ethical discussion than the last was. Nor do I think that any exponent of naturalistic Ethics would be willing to allow that this was all he meant. They are all so anxious to persuade us that what they call the good is what we really ought to do.5
The mistake, by Moore’s reckoning, is having the debate at all on semantic terms. It is simply the wrong arena – and I agree! An analysis into the meaning of goodness must be distinguished (quarantined!) from an analysis of its nature, and inquiry into the facts about it! Any native English speaker knows what “good” means. “Goodness is wellbeing” is simply not true, even if “wellbeing is good” is! Meanings and definitions are not where we should look for the “meat” of a dispute. We must resist misusing definitions in this way – as tempting as it is to get our “one fact for free.” Moore, in pointing out this danger, informed us in 1903 that this pitfall exists. A century later, Harris says, “this verbal trap is easily avoided” – and then falls right in!
C. Berkeley and Existence
Substantive definition is not confined to the field of ethical inquiry. For an example in metaphysics, Bishop George Berkeley, an early empiricist occasionally considered the “Father of Idealism,” engages in comparable misconduct in his Treatise Concerning the Principles of Human Knowledge, where he claims that only things which can in some sense be perceived, exist:
I think an intuitive Knowledge may be obtained of this, by any one that shall attend to what is meant by the Term Exist when applied to sensible Things. The Table I write on, I say, exists, that is, I see and feel it; and if I were out of my Study I should say it existed, meaning thereby that if I was in my Study I might perceive it, or that some other Spirit actually does perceive it.
[***]
I can as well doubt of my own Being, as of the Being of those Things which I actually perceive by Sense: It being a manifest Contradiction, that any sensible Object should be immediately perceived by Sight or Touch, and at the same time have no Existence in Nature, since the very Existence of an unthinking Being consists in being perceived.6
For Berkeley, then, it is simply a fact about existence’s meaning that it “consists in being perceived” – he collects his one fact for free! But this is terrible! If “to exist” means “to be perceived”7 , the truth of a claim like “there are no nonexistent things” – which we might ordinarily consider true in virtue of existence’s meaning – is an open question! If imperceptible things do not exist by definition, there might still be imperceptible things – things which are, but don’t “exist”! As in The Moral Landscape, taking this claim seriously robs the remainder of Berkeley’s writing of all substance – after all, if it really were true by definition that existence consists in being perceived, what could be gained by writing a whole treatise belaboring this point? Harris at least is proposing wellbeing as a definition of goodness, and arguing for it – but to hear Berkeley tell it, his substantive definition is simply already the case!
Importantly, the problem here is the methodology employed in arguing these points – not the claims themselves, necessarily. Berkeley makes substantive arguments in defense of his core claim elsewhere – as does Harris in The Moral Landscape. But these other substantive points are cheapened by this methodological defect. At best, these points are redundant, and the core claims of a work survive when they are ignored. At worst, they are crutches which, kicked out of the way, expose destructive holes in any argument’s core thesis.
D. Lenin and Imperialism
This third example I am less certain of because the original text is in Russian – and substantive definition is the sort of issue that can be exacerbated, or perhaps created wholly anew, by errors in translation. Nevertheless, this appears to fit the pattern. From Vladimir Ilyich Lenin’s Imperialism, the Highest Stage of Capitalism:
If it were necessary to give the briefest possible definition of imperialism we should have to say that imperialism is the monopoly stage of capitalism. Such a definition would include what is most important, for, on the one hand, finance capital is the bank capital of a few very big monopolist banks, merged with the capital of the monopolist associations of industrialists; and, on the other hand, the division of the world is the transition from a colonial policy which has extended without hindrance to territories unseized by any capitalist power, to a colonial policy of monopolist possession of the territory of the world, which has been completely divided up.8
What on earth is there to be gained by defining imperialism as a stage of capitalism? This is simply not what imperialism means. Once again, “capitalism leads to imperialism” is a substantive claim! This is a debate which should not be taking place in the semantic space. If there was a point in time when the Roman Empire happened to have competitive industries, but also conquered neighboring states for the sake of growing an empire – would it no longer “count” as imperialistic? Of course not! If a person thinks that capitalism leads to imperialism, and another person does not, this is far more than a difference of phrasing!
E. Caveat: Conceptual Engineering
This is not to say that all definitions which deviate from ordinary usage are fallacious by default. Conceptual engineering – the practice of refining our understandings of concepts, clarifying existing definitions, addressing edge cases, and “sweeping away the rubbish,” so to speak – is crucial. Most of Plato’s dialogues are examples of this. Socrates and his interlocutors engage in the process of “cleaning up” their concepts by demonstrating contradictions or inconsistencies in the colloquial understanding of justice, piety, etc., and offering an alternative conception which resolves ambiguities and confusions.
For an example of a commendably non-substantive definition, which is still a valuable and substantive philosophical contribution, consider Max Weber’s definition of “state” in Politics as a Vocation:
It is rather the case that in the final analysis the modern state can be defined only sociologically by the specific means that are peculiar to it, as to every political organization: namely, physical violence. [***] In the past the use of physical violence by widely differing organizations – starting with the clan – was completely normal. Nowadays, in contrast, we must say that the state is the form of human community that (successfully) lays claim to the monopoly of legitimate physical violence within a particular territory – and this idea of “territory” is an essential defining feature.9
Rather than leaning on our existing intuitions in lieu of argument, Weber’s definition of “state” seeks to offer an explanatory account of them. It is in fact unique to the things we call states, he claims, that they lay claim to a monopoly on legitimate physical violence. Weber’s definition offers itself as a more rigorous account of the manner in which the term “state” is presently used. Whether or not one agrees with this account, the important feature of this definition is that it is arrived at as a product of careful analysis of the word’s colloquial sense and usage. It does not define any facts about the world into existence. This is essential.
Substantive definition and other semantic maladies plague philosophy and theoretical writing more generally. Once this pattern is noticed, one sees it everywhere. Fundamentally, substantive definition consists in a failure to distinguish between a word’s meaning, and the facts about the thing to which the word refers. A proper theory of meaning allows this disambiguation to be conducted in a rigorous manner. In my view, a reasonable theory of meaning is the first thing every serious thinker needs. The next article10 will set out how such a theory might be developed!
This is the first article of my blog. If you liked it, please do not hesitate to let me know by subscribing and sharing it far and wide!
If you disliked it, please share it far and wide and express how outrageous and controversial you found it. Express your incredulity to as many people as will listen, and make sure to send them a link so they can see the horrors for themselves! Publish a scathing, highly publicized rebuttal! And be sure to subscribe, so you can keep tabs on these dangerous and noxious ideas!
Paid subscriptions are enabled on the blog, but I will not be publishing any articles behind paywalls until I have reason to believe that they are worth real people’s money. Until then, paid subscriptions do no more than express your support and enable my decadent lifestyle. I hope to publish something at least once a week, but I can’t possibly promise this! -G.
Credit for this example, and the quote at the beginning of this article, to Texas State University’s Philosophy Department.
Scott Alexander offers a similar analysis – with particular emphasis on the emotional reaction invoked by a word’s connotation – in his discussion of the non-central fallacy.
Harris, Sam. The Moral Landscape. Free Press, 2010.
Id. (emphasis in original).
Moore, George Edward. Principia Ethica (1903).
Berkeley, George. A Treatise Concerning the Principles of Human Knowledge (1710).
Or “to have the potential to be perceived, or “to partake in perception,” or any comparable permutation.
Lenin, Vladimir Ilyich. Imperialism, the Highest Stage of Capitalism (1917) (emphasis added).
Weber, Max. The Vocation Lectures (1919).
Thanks. Great read. My interest is philosophy of the mind. Poor or missing definitions represent the biggest barrier to sensible discussion.