Socrates' Conception of Definition
by James R. Beebe
Dept. of Philosophy
University at Buffalo
Copyright ã 2003
Outline of Essay:
I. Socratic Definitions
II. Real, Stipulative and Lexical Definitions
III. Against Real Definitions
IV. A Socratic Response
V. Conclusion
References
Appendix: An Objection to the Eternal Truth of Definitions
In the Euthyphro, as in many of Plato’s dialogues, we find Socrates engaging other people in a discussion of some philosophically important concept. In the Euthyphro the target concept is piety. In the Meno, Socrates is searching for the answer to the question “What is virtue?” Other Socratic topics of interest include: What is moderation (Charmides)? What is friendship (Lysis)? What is courage (Laches)? What is beauty (Hippias Major)? As you can see from this list of topics, Socrates was most interested in ethics. Each of these dialogues ends with Socrates and his interlocutors failing to find a satisfactory answer to the central question that was posed. No one is able to formulate a definition that is both explanatory and immune to counterexamples.
Some students get the impression that Socrates liked to ask questions of the form “What is F?” simply because he got a kick out of annoying people with counterexamples to their answers. These students do not think that Socrates was ever really interested in answering the questions he posed. They think Socrates is simply playing games with his fellow Athenians. Other students think that Socrates is serious in his quest for good definitions but that there will always be problems with any proposed definition. There will always be a counterexample or some defect in any definition, so it is pointless to try to formulate them in the first place.
Socrates, however, would have strongly disagreed with both of these interpretations of his philosophical activity. Socrates had a clear idea about what a good definition should look like and was firmly convinced that there were answers to all of his “What is F?” questions. In this essay I will explain what kind of definition would have satisfied Socrates’ demands. As we will see, Socrates’ standard for a good definition is so high that it is difficult to meet.
I. Socratic Definitions
We can figure out what Socrates thought about good definitions if we look carefully at his criticisms of Euthyphro’s proposed definitions. Any time Socrates says, “Your proposed definition is not good because it fails to do x,” we can infer that Socrates believes that any good definition should do x. Similarly, if Socrates says “Your proposed definition is bad because it does y,” we can infer that Socrates believes that any good definition will not do y. In other words, we can take all of Socrates’ criticism of Euthyphro and turn them into a list of requirements that any good definition must satisfy.
Euthyphro’s first attempt to define of piety was simply to point to his own supposedly pious action. When Socrates asked, “What is piety?” Euthyphro replied, “What I’m doing right now.” This answer, however, doesn’t explain anything. It does not tell us what it is about Euthyphro’s action that makes it pious. Socrates wanted to know whether Euthyphro’s action was pious because of the sincerity of Euthyphro’s conviction, because the gods have commanded that certain actions are pious or because of something else. Pointing to a single instance of supposed piety does not answer these important questions.
Moreover, the Socratic question “What is piety?” is not ultimately focused on any particular instance of piety. The question is asking about something general: What is it that makes any and every pious action pious? Not only can one not provide an adequate definition of some concept simply by indicating one instance of it, a complete list of all instances of the target concept would also not do the job. Suppose you had amnesia and had to relearn all sorts of basic concepts all over again. If you were to ask me “What’s a grandmother?” it wouldn’t be very helpful for me to parade before you the one and a half billion women in the world who are grandmothers. You would not be able to figure out what property all and only those women have in common simply by being confronted with them. Not only would I not succeed in communicating the meaning of ‘grandmother,’ engaging in such an activity is unnecessary. I can simply tell you that a grandmother is anyone who is the mother of a parent. When we are seeking a definition of a concept, we are looking for the set of defining features of that concept. We are not asking for a list of examples.
The set of defining features of a concept will pick out a combination of properties that every instance of that concept possesses. This particular combination of properties will also be something that only instances of that concept possess. If the combination of properties listed in a definition are possessed by some things that do not fall under the target concept, then the definition is not a good one. In addition, if each of the properties that appear in a definition are not possessed by every instance of that concept, then this is not a good definition either. There are two strategies, then, for testing the adequacy of a definition:
(i) Try to think of an example that clearly falls under the concept in question but does not satisfy the requirements of the proposed definition.
(ii) Try to think of an example that satisfies all of the requirements of the proposed definition but does not fall under the concept being defined.
For example, we might be tempted to include the property ‘being four-legged’ in a definition of ‘dog.’ But possessing this property is not necessary for being a dog. We’ve probably all seen three-legged dogs. They are still dogs, so being four-legged is not a defining feature of doghood. No definition of ‘dog’ that includes the property ‘being four-legged’ can be a good one. This is an example of the first strategy for testing definitions.
As an illustration of the second strategy, consider the following example (due to Aristotle) of a bad definition of ‘human being.’
(1) x is a human being if and only if x is a featherless biped.
If I pluck a chicken, then I’ll have a featherless biped which is not a human being. Since it is possible for something to possess the properties described on the right hand side of (1) without being a human being, (1) is not a good definition.
Good definitions must also not be circular. Consider the following definition.
(2) x is a pious action if and only if x is a pious action.
This is true, but it’s obviously uninformative. Some definitions can be circular in other ways. We might, for example, follow Euthyphro in offering the following definition of piety.
(3) x is a pious action if and only if x is dear to the gods.
When asked what it means for something to be dear to the gods, we cannot turn around and define ‘is dear to the gods’ as follows.
(4) x is dear to the gods if and only if x is pious.
That would be traveling in a small circle.
II. Real, Stipulative and Lexical Definitions
Socrates believed that there were real answers to “What is F?” questions. He thought the answers to these questions would be definitions that express eternal truths about the world. In other words, Socrates believed that there were ‘real definitions’ for all philosophically important concepts. Real definitions are eternally true statements that capture the essence of the concepts they define. It is important to note that Socrates did not think he knew what those eternally true definitions were. He simply believed that they were out there to be found. Even though people have had very different opinions about what virtue, justice or courage consist in, Socrates believed that there had to be facts about what virtue, justice or courage really are.
Not everyone has followed Socrates on this score. Richard Epstein (2001, p. 3), for example, denies that definitions express truths at all. It’s not that Epstein believes that all definitions are false. He does not believe that definitional claims can be either true or false. William Charron concurs. Writing about proposed definitions of death, he claims,
The best definition of human death would not express a truth or be a true statement about the nature of death: no definition proposed or adopted can be defended on the ground that it is true, nor criticized on the ground that it is false with respect to what death is. Nonetheless, the search for an acceptable, if not true, definition of death should not be abandoned. (Charron 1975, p. 981)
Epstein and Charron believe that definitions are expressions of choice that should be judged by their usefulness or convenience rather than by their truth.
Epstein and Charron think that definitions are neither true nor false because people can make words mean whatever they want them to mean. Sometimes people make up new words and give them definitions just to be funny. Here are some examples:
(5) elbonics: n. The actions of two people maneuvering for one armrest in a movie theater or airplane seat. (Murphy 2002, p. 389)
(6) pupkus: n. The moist residue left on a window after a dog presses its nose to it. (ibid.)
Suppose someone were to ask, “Is that the real definition of ‘elbonics’?” or “Is that what ‘elbonics’ really means?” or “Is that definition of ‘pupkus’ eternally true?” These questions don’t seem to make a lot of sense. The words ‘elbonics’ and ‘pupkus’ can mean whatever their creators want them to mean. Those who coined these words have absolute sovereignty over their meaning. Their definitions are not constrained in any way by how the world is. Anyone asking about the truth of these definitions would seem to be confused. Philosophers call definitions like the ones found in (5) and (6) ‘stipulative definitions.’ Stipulative definitions are created when people bring new words into being by some kind of creative act.
People are continually coining new terms—especially in the areas of business and information technology. For example, the terms ‘downsized,’ ‘internet,’ ‘e-business’ and ‘instant messaging’ did not exist a few years ago. Every time a new word is invented, a new stipulative definition has been created. New words are coined because it is convenient to have handy ways to refer to new things . If people were to find the words ‘elbonics’ and ‘pupkus’ convenient to have around, they would become part of our everyday repertoire of words. If no one finds them useful, they will slip into oblivion. (Note: Even if these words were considered utterly worthless, their definitions would still not be false.)
Sometimes people copyright or trademark the words they coin. Legendary basketball coach Pay Riley, for instance, owns the rights to the term ‘threepeat.’ In 1988 Riley was coach of the Los Angeles Lakers, who had just won their second championship in a row. Riley and Laker guard Byron Scott were sitting around a swimming pool in Hawaii trying to formulate a catchy phrase that would motivate the Lakers to win another championship in 1989. The phrase they came up with was ‘threepeat,’ and Riley immediately trademarked it. When the Chicago Bulls threepeated twice during the nineties and the Lakers completed a threepeat in 2002, Riley received millions of dollars in royalties from the sale of t-shirts with the word ‘threepeat’ written on them (McNeal, 2002). Usually, the formulators of new words do not make it into the history books, but if you think you’ve coined an important new term, you might want to consult a lawyer.
Another common type of definition is a ‘lexical definition.’ Lexical definitions are descriptions of how words are used by particular groups of people at particular times and places. Most of the definitions you find in a dictionary are lexical. Unlike stipulative definitions, these can be true or false. When, for example, I tell you that some scientists once used the word ‘phlogiston’ to mean that substance the presence of which is necessary for combustion, I am telling you something true. Even though there is no such thing as phlogiston, it is nevertheless true that some people used to employ this word in the way I described. Lexical definitions report truths about people’s habits of speech rather than truths about the world. If I have described the linguistic practices of the phlogiston-believers correctly, then my lexical definition is true. Because lexical definitions do not aim to describe reality correctly, they cannot be used to resolve philosophical, political, religious, or ethical disagreements. This is why you should never quote a dictionary while writing an argumentative essay.
Because words pass into and out of common usage, dictionaries employ lexicographers to monitor the frequency of word use. If a word becomes commonly used for the first time, it will receive its own new dictionary entry. Similarly, if a word falls out of common usage, it will get deleted from the dictionary. Collegiate dictionaries—more than other types of dictionaries—are typically more willing and eager to include new words in their latest editions. However, even the stoggiest, stuffiest, most uppity dictionary of them all, The Oxford English Dictionary, has included the following new words in its most recent online edition (available at: http://www.oed.com):
(7) d’oh: int. Expressing frustration at the realization that things have turned out badly or not as planned, or that one has just said or done something foolish. Also (usu. mildly derogatory): implying that another person has said or done something foolish (cf. DUH int.).
Quotation: “Look out, you dern fool! You’re gonna cut off your...” “D’oh!!!” (1996 A. FEIN et al. Simpsons Comics strike Back! 14/2)
(8) phat: adj. slang [orig. U.S., esp. among African-Americans; prob. a respelling of FAT; cf. the similar use of big, large, etc., as terms of approbation]
(a) Of a person, esp. a woman: sexy, attractive. (b) Esp. of music: excellent, admirable; fashionable, ‘cool.’ Particularly associated with the hip-hop subculture.
Who would have thought that Homer Simpson would make a significant contribution to the English language? Other recent additions to the OED online include ‘chick flick,’ ‘lava lamp’ and ‘24-7.’ (According to an ABC News report, the term ‘bootylicious’—popularized by Destiny’s Child—has made it into the latest edition of Webster’s Collegiate Dictionary (Jennings and Brewster, 2003). However, a search of the online version of the Eleventh Edition of Webster’s Collegiate Dictionary (available at http://www.m-w.com) failed to validate this report. I was relieved not to have found it.)
Stipulative definitions can become lexical definitions when the stipulations proposed catch on and become widely accepted. At their inception stipulative definitions are neither true nor false, since they neither: (a) accurately describe how things stand in reality, nor (b) accurately describe how words are actually used in practice. Once stipulative definitions become accepted, they can, however, become true descriptions of the behavior of certain groups of people. Think of a stipulative definition as a proposal for how to use a new word. If the linguistic proposal is accepted and people begin to follow the usage that is proposed, it will then be appropriate to offer a lexical definition describing the linguistic behavior of the people in question. If certain segments of the society use a word in one way, and other segments use it in another, it will not be possible to offer a single lexical definition that covers everybody’s linguistic usage. For example, some people in contemporary U.S. society use the word ‘patriotism’ to describe a virtuous devotion and loyalty to one’s country. Other people use that same word to describe a pernicious form of nationalistic and unthinking bias in favor of one’s country. So, we may need to offer one lexical definition for one group and another lexical definition for another group. Notice that in the lexical definition of ‘phat’ above hip-hop subculture was mentioned. Lexical definitions always describe the behavior of particular groups of people at particular times and places. This is most readily seen in the case of slang words like ‘phat,’ but it is actually true of all words.
Although Socrates would agree that it is possible to stipulate the meanings of some words and to report on how people commonly use them, he would disagree with the claim that definitions are always stipulations or summaries of common linguistic usage. Socrates maintained that questions like “What is truth?” or “What is virtue?” call upon us to do more than: (a) stipulate answers that we happen to find convenient or (b) report on how people actually use the words in question. In response to those who try to stipulate definitions of, say, virtue and truth, Socrates would say that we are not free to define virtue and truth however we please. We must discover the facts about what virtue and truth really are. The truth about these ideas must be discovered rather than created or stipulated.
In response to those who look to lexical definitions to solve philosophical controversies, Socrates would say that, simply because certain cultures have used the word ‘rights’ in a way that excludes women and minorities does not mean that women and minorities do not have rights. Similarly, facts about how people actually use the word ‘justice’ may not be reliable guides to what justice really is. The masses might be dead wrong about what justice is. Their prejudices and ignorance may influence the way they use the word ‘justice,’ but that doesn’t mean that prejudice and ignorance are part of the true nature of justice. Therefore, empirical reports on how people actually use words (i.e., lexical definitions) will not tell you anything about the real truth of the matter. Socrates was not very interested in how ordinary people use words like ‘justice’ or ‘virtue.’ He wanted to know the truth about justice and virtue.
III. Against Real Definitions
Is Socrates correct in thinking that there are real definitions for all philosophically important concepts? Or are Charron and Epstein correct in their claim that no such eternally true definitions are out there to be found? To help us think through this issue, let’s consider some of the arguments Charron offers in favor of his position on definitions.
One reason Charron thinks that real definitions do not exist is that there has never been any agreement on the definitions of important philosophical concepts after 2,500 years of debate about them. Philosophers no more agree today about the definitions of ‘virtue,’ ‘justice’ or ‘wisdom’ than they did in Socrates’ day. The great number of discarded definitions that litter the history of philosophy makes Charron skeptical that there ever will (or even could) be consensus on these matters. If real definitions are not the sort of things we humans could ever come up with, it can be difficult to see why we should still believe in their existence.
Another reason Charron is skeptical about the existence of real definitions is that he thinks there will always be unforeseen future circumstances that render previous definitions obsolete and inapplicable (Charron 1975, p. 979). For example, death was traditionally defined as the total and permanent arrest of all vital functions, including circulation and respiration (Charron 1975, p. 979). This definition was formulated before anyone imagined the possibility of artificially circulating someone’s blood and using a respirator to help them breathe. If someone is completely brain dead, but that person’s circulatory and respiratory functions are supported artificially, that person—in the eyes of most people today—would actually be dead. The old definition of death got in the way of using donated organs for transplants. Charron (1975, p. 979) writes,
[T]he practice of pronouncing death only in the event of both cardiac and respiratory failure has proved inhibiting and burdensome to the procurement of useful organs for transplantation; transplant surgeons operating in jurisdictions that have not adopted legal definitional adjustments risk civil suit if they select donors whose cardiac function has not totally failed.
Because those who formulated the traditional definition of death could not have foreseen recent developments in medicine, there is no way they could have fashioned a completely adequate definition of death.
Even those who first put forward definitions of death in terms of brain death during the 1950s and 1960s could not have imagined what the future would bring. They defined death as the total loss of all brain function. According to these definitions, someone in a permanent vegetative state, with significant brain damage and no hope of ever regaining consciousness would not be considered brain dead if their brain stem was still telling their lungs to continue respiration. Yet it is people in permanent vegetative states who seem to us today to be the clearest cases of brain death.
The traditional definition of death in terms of the cessation of all vital functions took death to be an all-or-nothing affair. But now we know that some vital functions can cease—because of specific kinds of trauma—while other vital functions can continue. Similarly, early definitions of brain death took it to be an all-or-nothing affair as well. We now know that part of one’s brain can cease to function, while other parts of the brain can continue to function normally. So, physicians and legal experts in recent years have proposed definitions of death in terms of the permanent and total loss of consciousness.
There is no way to tell what future medical and neurological research will reveal about the nature of life and the brain. So, even though current definitions of death may seem superior to older ones, Charron cautions that we should think of current definitions of death as merely contingent proposals that help us to get by right now. There is no reason to think that we can formulate an eternally true definition that will hold up for all time and withstand all future changes in society and technology.
In addition to offering arguments against the existence of real definitions, Charron also has some constructive things to say about how we should view definitions. Although Charron believes that lexical definitions are certainly possible, most of what he says about definitions (and, therefore, most of what appears in the rest of this section) is concerned with stipulative definitions. Charron thinks that coming up with a good definition is more like deciding on who the next President will be than it is like discovering a cure for cancer. Charron does not believe that scientific investigation can settle for us this question because the issue of what we mean by ‘death’ would have to be settled before we could develop any experiment to test for death (Charron 1975, p. 983). Someone developing medical tests for death must already have in their minds what death is. Such tests would merely reflect certain people’s conceptions of what death is and could not, therefore, establish what death really is. In other words, a definition is a choice rather than a discovery. He writes,
A publicly defensible definition can be developed if it is recognized that any definition proposed or adopted, whether it reasserts, precises, or radically alters the traditional notion of death or the new brain death definitions, is an expression of choice. (Charron 1975, p. 981)
Charron (1975, p. 983) also calls definitions ‘conventions of language’ and ‘expressions of preference.’
The choice of a definition of death is a choice with ethical and legal ramifications, affecting everyone in society. Charron (1975, p. 985) writes,
[E]nacting a legal definition of the term ‘death’ determines whether a given individual will continue to receive specific immunities and protections and whether other individuals will be obligated or liable in certain ways. As a consequence, defining ‘death’ is an ethical choice.
Because the general public is affected by the way death gets defined, Charron thinks the issue should be decided in the same way that all issues affecting everyone’s needs and interests are decided in a democratic society: by having a public debate about it, with everyone getting to have some kind of input (either directly or via elected representatives) into the final outcome. He explains,
Because it is an act that determines the persons to whom certain basic human protections are extended and rights granted, stipulating a legal definition of the term ‘death’ must meet the ethical requirement of a pluralistic and democratic society. The rule of action must be one that all or most persons are willing to have applied to themselves. If citizens are to participate, as they should, each citizen should ask himself the following questions: Under which kind of permanent incapacity am I willing to have removed from me (i) the rights and protections guaranteed to me as a living person and (ii) the duties and responsibilities imposed upon me for those in my care? (Charron 1975, p. 991)
Thus, Charron’s approach to definitions is thoroughly democratic.
In Charron’s (1975, p. 986) opinion, “basic social arrangements and public ethics should not be determined by one segment of the populace.” So, he maintains that the medical community does not have the right to dictate what the definition of death will be. Nor, he thinks, does the legal community have that right. These groups have not been elected by any democratic process to represent the interests of the rest of us. Hence, they have no political legitimacy in a democratic society. Charron believes that the medical and legal communities will have important and interesting contributions to make to our discussions of death, but they do not have greater authority than those in the general public when it comes to deciding what the definition should be. Regarding the positive role that the medical community can play in the issue of death, Charron (1975, p. 986)writes, “The medical profession must be relied upon to determine reliable empirical tests for the type of incapacity eventually identified by law as death.”
Some people would object to the idea of voting on definitions the way we vote on Presidential candidates. But Charron (1975, p. 984) writes,
The choice of a convention about the term ‘death’ is, of course, not arbitrary; its public defensibility relates to the problems that provoked its consideration and the other human needs and interests that would be affected by its legal adoption.
In other words, ‘decided by a vote’ does not have to mean ‘decided arbitrarily.’ As in all matters of public policy in the U.S., we simply look at who is going to be affected by various proposals and how they are going to be affected, and we decide on which proposal will best satisfy our needs and interests.
IV. A Socratic Response
If Socrates were alive today and could read what Charron has written in critique of real definitions and in favor of viewing definitions as public proposals with ethical and legal ramifications, he would not be convinced. Socrates might say something like this:
Do you really think that death is simply whatever we vote on it to be? That’s crazy! Either you’re dead or you’re not. The way people happen to vote on the definition of death has nothing to do with the objective fact of whether you’re dead or alive. Only facts—not voting procedures—can determine whether proposed definitions are good or bad.
In response to the charge that more than 2,000 years of non-stop disagreement among philosophers make it seem unlikely that there are any real definitions to be had, Socrates would argue that the disagreement may be due to the complexity of the ideas being discussed. A lack of consensus doesn’t necessarily indicate anything about whether there are real definitions or not. After all, scientists have been trying to discover the fundamental physical properties of the universe for more than 2,000 years, and they haven’t succeeded yet, either. Maybe ultimate truth in philosophy, like ultimate truth in physics, is simply very difficult to come by.
V. Conclusion
In response to the question, “Is there any definition that could ever be good enough to satisfy Socrates?” the answer is definitely “Yes.” Accurate real definitions would satisfy Socrates. Real definitions, however, are often hard to come by—especially when they concern controversial notions such as virtue or truth. Socrates would not want the tremendous difficulty involved in constructing real definitions to make us lose heart. Socrates spurned his livelihood, his health and eventually his own life in his quest for true definitions of the things that matter most. If he were alive today, he would urge us to join him in his quest.
References
Brewster, Todd, and Jennings, Peter. 2003. “Word Up: How Merriam-Webster Decides Which Booty Belongs in the Dictionary.” (Sept. 23, 2003) Available at: http://abcnews.go.com/sections/wnt/DailyNews/ISOA_Websters020923.html
Charron, William. 1975. “Death: A Philosophical Perspective on the Legal Definitions.” Washington University Law Quarterly vol. 1975: 979-1008.
Epstein, Richard. 2001. A Pocket Guide to Critical Thinking. Belmont, CA: Wadsworth.
McNeal, Stan. 2002. “Adjust or Bust: The Finals are All About Adjustments, but Shaquille O’Neal Keeps the Lakers One Step Ahead of the Nets.” The Sporting News (June 17, 2002).
Murphy, Gregory L. 2002. The Big Book of Concepts. Cambridge, MA: MIT Press.
Appendix: An Objection to the Eternal Truth of Definitions
One of the most common objections to the existence of real definitions is the fact that we seem to be able to make words mean whatever we want them to mean. Since it is evident that we have this ability, it can seem rather odd that someone like Socrates would claim that definitions can express eternal truths. Socrates, however, would say that it is not as odd as it may first appear.
Consider the following definition:
(9) Bachelors are unmarried adult males.
Socrates would claim that (9) is necessarily true. Consider the following objection to this view:
“Bachelors are unmarried adult males” is not necessarily true because the word ‘bachelor’ doesn’t necessarily mean ‘unmarried male.’ Meanings change. The word ‘bachelor’ could have meant ‘married female’ or even ‘David Hasselhoff’s pet sea slug.’ Because we can make words mean whatever we want them to, it is a merely contingent fact about our current linguistic practices that “Bachelors are unmarried adult males” happens to be true. Although it is indeed true, it is merely contingently true. It is not necessarily or eternally true.
In response to this objection Socrates would want us to distinguish between words and concepts. Consider the following sentences:
(10) Snow is white.
(11) Schnee ist weiss.
(12) La neige est blanche.
All of these sentences express the same thought. They simply do so in different languages. Recognition of this fact allows us to distinguish between a thought and the particular sentence someone uses to express that thought. Just as (10), (11) and (12) express the same thought, the words ‘snow,’ ‘Schnee’ and ‘la neige’ express the same concept. Concepts are the building blocks of thoughts. So, we can also distinguish between a concept and the particular word someone uses to express that concept.
When Socrates claimed that definitions were eternally true, what he meant was that the thought expressed by an accurate real definition was necessarily true. While it may contingently true that the sentence used to express that definition only contingently express a thought that is necessarily true, it is nevertheless true that the thought that gets expressed is necessarily true.
In other words, those who object to the possibility of necessary truths confuse the following claims:
(13) The sentence “Bachelors are unmarried adult males” necessarily expresses a truth.
(14) The sentence “Bachelors are unmarried adult males” expresses a necessary truth.
(13) is false, while (14) is true. The reason (13) is false is that it is not necessary that the word ‘bachelor’ means what it means today. ‘Gay’ used to mean ‘happy,’ but it doesn’t anymore. There is no necessary connection between words and the concepts they express. So, (13) is false because the words in that sentence could have been used to express a thought that was false. (14), however, is true. The thought that actually gets expressed by the sentence “Bachelors are unmarried adult males” is a thought that is necessarily true. Sure, we could have used words in a different way. But when we talk about the necessary truth of the thought expressed by the sentence “Bachelors are unmarried adult males,” we aren’t saying something about that sentence. We are talking about the thought, and it is indeed necessarily true. So, this very common objection to the existence of eternally or necessarily true real definitions fails. It rests on a confusion.
