Truth is always an afterthought

Dissecting the marriage of convenience between truth and language

From Narrow To General AI
9 min read16 hours ago

By Yervant Kulbashian. You can support me on Patreon here.

“Common sense” is an ironic term.

In everyday use it means “what we all understand to be true” or “that which needs no discussion”. It’s ironic, then, that the average person only uses the term “common sense” when they’re speaking to someone who apparently doesn’t agree with what they are saying. “It’s just common sense!” they exclaim in exasperation. When everyone agrees, when there is no doubt, the phrase “it’s common sense” is unnecessary and doesn’t show up. It only appears when it is manifestly not the case. Using the phrase is therefore an implicit expression of its opposite.

So why use it, if it is never accurate to do so? The answer is also fairly obvious: you use it when you are trying to convince someone that they should adopt a belief. When your interlocutor’s questions are starting to annoy you, you make one final attempt at peer pressure, something like “everyone already believes it, so I shouldn’t have to prove it to you”. The phrase goes so far as to suggest that the listener themselves believes it, and is merely denying it through stubbornness — it straddles an awkward line between “you already agree” and “you don’t believe me”. What you really should say in such cases is: “I didn’t expect to ever have to defend this point”. But since you apparently don’t want to put in the effort to dissect your own argument, that is not what you say.

This makes the statement a kind of verbal paradox. It is one of many instances of using language to enforce your own version of how things should be by simply claiming that it is already true. Language here is not presenting a faithful mirror to reality, but is a means of uniting two contradictory poles; that is, making something you want to be true to actually be true. We regularly see emotionally-laden, wishful-thinking statements follow this pattern, where the statement outstrips the evidence. For example:

  • in political discussion: e.g. “the president agrees with my take” or “no sane person would vote for her”,
  • when presenting opinion as fact: e.g. “Cats was an awful movie” or “society generally behaves rationally”,
  • when providing hope: e.g. “we’ll get through this” or “things always work out in the end”,
  • when expressing religious conviction: e.g. “he is the Messiah” or “God looks after all of us”.

In all these examples, the wish for something to be true is expressed as if it already is so.

Though we can readily spot the ideological bent in the above cases, it is not easy to know where to draw the line between the two modes of speech, i.e. between saying something because you believe it is true, and saying it because you want it to be true. The nature of language inherently blurs this distinction, simply because all acts of speaking take time and energy. So every verbal expression, every communication implicitly raises the question: why speak at all? Why spend the energy to say something, when you could just as well stay silent?

The answer is already given in the word “communication”, which means “to make unified”, as in to align many people to one way of thinking. You are trying to bring about a common sense. Anytime you speak, you do so to make others think a certain way — regardless of whether it is true. This motivated nature of speaking makes the relationship between language and truth rather shaky, and it is difficult to know where in the process “truth” can be located.

At one end, words by themselves are neither true nor lies; they are merely culturally arbitrary sounds. Presumably it is what they represent that is “true”. So the best one can say about so-called true statements is that they are representations of some real remembered experiences. Yet here we hit our first roadblock. Thoughts may be haphazard, real, imaginary, chaotic, invented, fantasies, accidental, etc., and there is no guarantee that the thoughts you base your words on will necessarily be “true” in any sense. Nor can anyone determine just by looking at the content of a thought in isolation whether it is true or false. What if you are wrong or misinformed? What if it’s the result of wishful thinking? Even direct memories cannot be relied on to be an objective record of reality. Thoughts are just thoughts; they have no inherent quality of “truthiness” until you take the time to evaluate that they are indeed true.

And everyday introspection shows that being truthful is not really about the content of the original thoughts, nor the words you say, but rather about how you decide to connect your thoughts to what you say. The search for truth must begin by looking at the decisions that generate words from thoughts, when you decide what you will say, when you may choose to lie, or simply not speak at all. Whenever you argue that someone’s statements are lies, as opposed to an honest mistake, what you mean is there was something in their intentions to speak that made it a lie. Truth is therefore a motive, a “truth-telling intention” that directs your speech. To find the source of truth, then, we must investigate your motives.

Which brings us right back to the earlier question: whether or not you are being truthful, why speak at all? The fact that something is true is, by itself, not a reason to speak. Rather the reason for speaking is something more specific, like: “it is important that they know about this fact”, or “I want to look smart”, or “they would be interested in this”, or “they will avoid making a giant mistake”. In all these cases — whether or not you tell the truth — the act of speaking is directed at some outcome; it is not merely a mirror-reflection of thoughts in your mind.

That outcome is the reason that “it is imperative they know this now”; it is the reason you formulate your words in a native language they will understand, the reason you use the tone and emphasis you do. It is the reason you employ whatever tactic at your disposal to get the point across, including deliberate exaggerations… and sometimes even conscious lies. Whatever motive is driving you to speak in the first place — and there must be such a motive — will also determine what you say, and how you say it.

As you may realize, both truth and lies fit nicely into this template: both are intended to achieve an effect on others. There is no such thing as a purely “truth-telling” intention without extenuating motives. Any motive for speaking invites the possibility of bending the truth to serve that need. Even a person who desperately wanted to be seen as honest may decide to lie if that will achieve the goal. So no meaningful distinction can be made between truth and lies at the level of intentions.

The only option left for finding “truth” is to look at how you connect your thoughts to the intended words. Regardless of the motivation, truthful statements, one presumes, create a faithful connection between the source thoughts and the words you decide to say. Unfortunately, every act of planning speech follows the motive that drove you to speak; thus planning is necessarily motivated as well. When you work out in thought what you will say, you are still considering the effects of your words on others’ behaviour, and picking those that will reasonably achieve your aims. Consequently, every linguistic thought that runs through your mind ends up being a plan or rehearsal for how you would achieve a desired effect.

Once that’s done, once you have a plan, what more is there left to do? Where exactly in these steps does “truth” make its grand appearance? It is not in the original thoughts, since those are diverse and haphazard, and aren’t guaranteed to be true. It is not in the intention, since that is always aimed at achieving some useful goal. It is not in the planning, since that simply derives the set of words to achieve your aim. And it is not in the words themselves since those are only noises. Nowhere in the series of steps does the elusive artefact called “truth” ever show up.

And yet we can still somehow sense, at least for ourselves, when we are lying. We also know the feeling of pride and clear conscience that comes from telling the truth. Where does this intuition come from?

“Truth” is in fact an afterthought, a self-critical appraisal of what you planned to say. For example, as you put together your intended words you may decide to evaluate them from the perspective of how others would refute you or argue against you. Or you may be worried about how friends will be harmed if they follow your instructions. Either way it is not the production of the words themselves — how you connect thoughts to linguistic plans — that defines “truthfulness”. Truth is a secondary appraisal, a hesitation caused by some foreseen negative consequence of what you will say.

We learn this lesson early on when we discover, as children, that we can’t simply say whatever we want without concern for consequences. For example:

Parent: “Did you wash your hands?”

Toddler: (planning what to say) If I say no, then I will have to wash my hands. If I say yes I won’t. So I’ll say yes (achieves goal).

Toddler: “Yes.”

Parent: “Then why didn’t I hear the water running? It makes me upset when you tell lies.”

Toddler: … (something went wrong, my parents are displeased).

Or perhaps:

Parent: “Will Zoe's dad drive you to the party?”

Toddler: (planning) If I say yes, I’ll get to see Zoe. I’d like that (achieves goal).

Toddler: “Yes.”

Parent: “OK, then I don’t need to drive you”

Toddler: …(something went wrong, no one is taking me to the party).

This negative feedback is the root of your understanding of truth. It represents a miscalculation in verbal planning, which you only notice when something undesirable happens. Going forward, these failures spur you to second guess and reformulate what you plan to say. They also give rise to that murky conscience associated with lying.

Truth must be taught; it is not innate. In the examples above, when the toddler first thought of saying “yes” to the questions, he had no qualms about doing so, no prick of conscience or doubt. He wasn’t really lying because he had no conception of what a lie was. His linguistic thoughts were merely a set of plans to realize his wishes. In fact, he was likely happy with his problem-solving abilities, in being able to use language to achieve a goal. Saying “yes” was as normal and useful as answering “yes” to the question “do you want ice cream?”

Even for adults, words — in action and in thought — remain tools aimed at achieving an effect. “Truth” as we find it is a complication of this same system, a form or self-regulation that arises when the world pushes back against our wishful thinking. This push-back causes an internalised sense of foreboding, which motivates the zealous efforts we make at being “honest”. The righteous social pride and warm glow we feel at being honest is our pride and inner confidence at having resolved this tension.¹

Truth and language have always had a shaky and problematic relationship. Words are a notoriously unreliable medium for storing complete truth. But this is only a problem if you assume the relationship should be stable. Instead, it is a marriage of convenience, of utility, lasting as long as needed to achieve an effect on the listener. Once a set of words have had their effect, the listener can discard them like a used tissue.

From this perspective the paradox in the term “common sense” is no paradox at all. There are many reasons why you may want to employ the phrase “it’s common sense” to sway others. You might say it because you think it will be to the listener’s benefit to agree with you. Or perhaps you simply don’t want to argue the case, preferring to shut the discussion down there. Like all speech, the phrase serves a practical function. Taken this way, “it’s common sense” is no more ironic a phrase than anything else you may say.

What the example of “common sense” helps us see is how both speech and thinking itself must be separated from the necessity of “truth”. No one ever said anything simply because it was true. And since linguistic thoughts are only plans for what you will say, no one every thought anything simply because it was true either. “Truth” is not an innate mode of contemplation, and to look for a “truth function” in the mind is to misunderstand its nature. It is, in the best of cases, a creative act of utility, of coming up with something useful to say to others. All speech is inherently utilitarian — telling the truth is merely a type of utility.

¹ As some readers may notice, we have here described a link between “truth” and reward-based Reinforcement Learning.

Sign up to discover human stories that deepen your understanding of the world.

Free

Distraction-free reading. No ads.

Organize your knowledge with lists and highlights.

Tell your story. Find your audience.

Membership

Read member-only stories

Support writers you read most

Earn money for your writing

Listen to audio narrations

Read offline with the Medium app

--

--

From Narrow To General AI
From Narrow To General AI

Written by From Narrow To General AI

The road from Narrow AI to AGI presents both technical and philosophical challenges. This blog explores novel approaches and addresses longstanding questions.

No responses yet

Write a response