Michael Barkasi

philosopher and cyclist

Engaging in conversation

Most of us have many conversations each day: a quick “good morning” to our cohabitants, small talk at the office water cooler, a phone call to your mother to catch up, chatting with your children about their day over dinner, haggling with a client over price, a tax meeting with your accountant, some flirtatious banter with a captivating stranger, and a round of pleasant reminiscing or cathartic venting over a beer with a friend. As these examples suggest, the purposes of our conversations can vary widely. We make small talk merely to grease the wheels of social interaction, we ask questions to gather information, we flirt to signal sexual interest, and we vent to sooth frustration. We express love to our family, make promises and form contracts, entertain each other, and do many other things through conversation.

Although conversation is an integral part of our lives, there’s one aspect of it which we don’t do very well. I’ll call this aspect engagement. While talking—conversing—with others is something humans do by nature, what comes naturally is unreflective, passive, or merely reactionary conversation. We don’t naturally engage.

What is engagement? While I don’t have a rigorous definition in mind, there are some behaviors which clearly count. The most basic is asking questions. Not all forms of questioning are engagement. I’m not engaging if I ask random questions unrelated to what you’re saying. But I would be engaging if I asked questions to check my understanding or to resolve ambiguities. Actively seeking out ambiguities and potential misunderstanding would themselves be other forms of engagement. The former you might do by thinking up multiple interpretations of what’s been said, while the later you might do by listening for the unexpected from your interlocutor.

Here’s an example: Jill: It’s bad to smoke. Fran: Cigarettes or marijuana? Jill: marijuana. Fran: But it doesn’t harm anyone. Jill: No, I don’t mean immoral, I mean bad for your health. First Fran engages by asking for clarification. Jill later catches a misunderstanding and corrects it. It’s an overly simple example, but hopefully it gets across the idea.

Water-cooler small talk provides a good contrasting example of unengaged conversation. Jill: Did you see the game last night? Fran: Oh yeah, it was great—but it ran way too late. Jill: Yeah, I was up too late as well; now I’m beat. In that exchange Fran never asks what game Jill is talking about. Maybe she knows, maybe she doesn’t. Maybe she has one in mind, but it’s not the one Jill meant. Perhaps Jill has misunderstood Fran’s comment about the game running late. She assumes Fran meant that it kept her up, but perhaps she meant instead that it kept her from her work.

When engagement matters

Of course, no one intends water-cooler small talk to involve any real engagement. Repeating cliché conversational tropes in an almost scripted manner without much regard for what’s being said is just how it goes. It even involves bullshitting, saying things without regard to truth because they keep the conversation going. In fact, it’s the bullshitting, cliché conversational tropes, and overall lack of substance which makes this sort of small talk work as a form of socializing for professional colleagues within the work environment.

So some forms of conversation have purposes which preclude engagement, or at least don’t depend on it for success. But there are lots of other conversational contexts where success does depend on engagement. Management deliberates about how to solve a problem that could sink their company. You go to a friend and need real advice on how to solve a serious personal crisis. Public officials debate the contours of public policy issues such as taxation, abortion, or immigration. Scientists wrestle with the correct interpretation of their data. In these sorts of cases—and many others—engagement matters.

We might characterize these cases as conversations which aim to solve a problem, get at the truth, or produce good decisions. Let’s call these cases where we aim to get it right. Why does engagement matter in cases where we aim to get it right? It matters because getting it right requires deep understanding, depth of thought, and productive reasoning. Each of these three things, in turn, require engagement.

Deep understanding

Engagement is necessary for deep understanding. Understanding—including the avoidance of misunderstanding—is something you build up through engagement.

People often don’t recognize ambiguity, misunderstanding, or shallowness. It’s a long story about why we miss them so often, but the crux of it has to do with how our own thinking works. We all have some stock of concepts in our head with which we think: our conceptual framework. As long as the words someone says activate some concept of ours, we won’t notice that there might be other concepts they intended (we won’t notice ambiguity), that we might have activated the wrong concept (we won’t notice misunderstanding), or that there might not be much depth of understanding to our concept (we won’t notice shallowness). We mistakenly interpret the ease with which we access our own concepts as clarity, understanding, and depth.

The upshot is that if you’re merely listening passively to what someone says, you’re apt to misunderstand them and not even notice. You might misunderstand because there is ambiguity and you haven’t heard the meaning they intend. Even if there’s no ambiguity, you might misunderstand simply because you’ve mistakenly activated the wrong concept. To catch these forms of misunderstanding, you have to try actively to think up alternative interpretations (so that you catch ambiguity) and watch for unexpected conversational moves or claims (so that you catch misunderstanding). As you converse you have to follow up with questions resolving the ambiguity or misunderstanding. Thinking up alternative interpretations, watching for the unexpected, and asking questions are all forms of engagement; so, it’s through engagement that you avoid misunderstanding.

Depth of thought

Much of our thinking is done through the medium of language: we think via spoken external and private internal speech. (There are non-linguistic forms of thinking as well, e.g. thinking about your favorite food by imagining its taste or visualizing it in your mind’s eye.) In some sense what we say out loud—and also what we say to ourselves in our heads—is just an expression of our thoughts. For example, someone might say they can’t articulate their thoughts in words. This is because we think in concepts, which aren’t merely the words we say. But while thoughts don’t merely reduce to our actual acts of public and internal speaking, the two are very tightly related. How so?

Concepts are informed by both our sensory experiences and information we take in through language. My concept of gold, for example, is informed both by my experiences of seeing and touching pieces of gold and by what I’ve been told about it: e.g., that it has a certain high dollar-value per ounce, that it has a certain atomic structure, and that in history people sailed the oceans searching for it. Some of our concepts—concepts of things which unlike gold cannot be directly experienced—are entirely informed by what we say about them. (Take, for example, “abstract” concepts, like mathematical concepts, or concepts of justice and morality.) Even those which are grounded in experience, such as my concept of gold, are often filled out or enriched by information we learn linguistically. The upshot is that most of our concepts are informed or in some way constituted by the words we use to express them. Concepts are rarely fully independent of the words we use to express them.

So for the most part the thoughts we’re able to think are the thoughts we’re able to put into words. So the depth of those thoughts will be limited by the depth of the words we use. Thoughts made up of words with rich, subtle, or complex meaning will themselves be rich, subtle, or complex. But if the words you use lack substance or depth of meaning, your thoughts will as well.

So why is engagement necessary for deep thought? If you want to think deep thoughts, just make sure you use words with rich, subtle, or complex meaning. The problem with words is that beyond a certain point they only have as much meaning or substance as we pack into them. Words derive meaning from two sources: their long-standing conventional use within a language, and the more immediate context of their use. The word ‘now’ is a good example. Conventional use has it that ‘now’ refers to the time it’s uttered. But the time referred to by any particular utterance of ‘now’ depends on when it was uttered. If I say ‘now’ at 2:21pm on August 14th, 2020, I mean 2:21pm on August 14th, 2020. If I say ‘now’ at a different time and date, I mean that different time and date. Although the story is more complex, something similar goes for any other word: the exact meaning as I utter that word in that instance will depend on both its conventional meaning and any particularities it acquires in the present context of use.

For an example, consider the phrase “the good life”. What do I mean by this? Maybe you have some rough idea of what I might mean, e.g. a life that would be pleasing. Maybe there are many things I could mean. In fact, what I have in mind at the moment is the particular concept denoted by that phrase as explicated by Plato and Aristotle. They have a lot to say what which lives are “good”, and what we even mean here by “good”. If I kept going with this explanation, I’d fill out a meaning of the phrase “the good life” which had all the substance, depth, and subtlty of that historical conversation going back to Plato and Aristotle. Similarly, consider the word “atom”, which roughly refers to the smallest individual unit of an element. But, if you’re a physicist writing about atoms in a physics paper, your use of the word “atom” draws on all the technical definitions, computations, and experiments that physicists have associated with that word. The point is that physicists have engaged in a long history of use of the word “atom” which has built up a depth to the concept going well beyond what’s captured by its everyday conventional meaning.

Now, the key point is that depth of meaning is only built up in a term through certain sorts of conversational moves. If you and I are talking and using the terms “the good life” or “atom”, but we use these terms reflexively or without consideration of what they mean, we won’t build up any semantic depth. It’s only by, for example, asking questions about the necessary and sufficient conditions for the good life, or asking questions about how we might experimentally observe atoms, that we enrich the concepts attached to those terms. At least, we need to talk in a way that’s sensitive to these fundamental issues, in a way which respects and recognizes that the terms may not be clear. We need to consider associations between terms, and carefully cut distinctions which unreflexive use might elide over. This kind of reflective use of terms is, again, just another form of engagement.

So, engagement is necessary to think deep thoughts because rich, subtle, or complex meaning in a word is something that’s built up through engagement. The meaning a word acquires from conventional use is usually rather thin. It takes engagement in conversation to build up something more substantial.

Productive reasoning

Often conversation involves giving reasons. I ask you why you believe some claim you’ve just made, or you ask me why I think we should follow the plan I just proposed. We, in turn, answer each other by giving reasons, i.e. giving an argument of some kind. But this kind of reason-giving or argumentation isn’t automatically productive. And here I don’t just mean that giving a reason doesn’t automatically convince your interlocutor (although of course that’s true as well). The kind of productivity I have in mind is epistemic. Giving reasons is epistemically productive only if those reasons actually support your claim or proposed action, i.e. only if they get you closer to the truth or right action.

Now, productive reason-giving requires engagement. As you and I give each other reasons, it’s not enough that we simply hurls reasons at each other, hoping that one of them sticks. If we want our discussion to be epistemically productive, we need to probe each other’s reasons. For example, we have to try to construct counterexamples (examples showing that someone’s premise can be true while their conclusion remains false), or press on whether the give reason (i.e. premise) is itself true in the first place. You give a premise, I press on the truth of the premise, you give an inference, I construct a counter example to challenge the inference, etc.

The idea is that productive conversations require productive reasoning, which requires engagement. I have to pay attention to what you’re saying, absorb it, understand it, and probe it. If I merely shrug off what you say (the reasons you give) without addressing them, I’m not conversing productively with you.

Why engagement is hard

Why don’t people engage much, even when the conversation is the sort that calls for it? While unengaged conversation seems to be natural for most people, the kind of back-and-forth of question-and-answer needed for real engagement just doesn’t seem to be. It needs to be trained. (Social pressure or an unwillingness to ruffle feathers might also keep people from engaging, even when the context calls for it and they are trained to do it.) Training in this sort of engagement is one of the things you’re supposed to get out of university-level liberal arts education, but outside of this context it’s rare to find this training. (Perhaps parents who train their children to be good listeners afford another avenue for training in engagement.) What’s more natural in most contexts is a confrontational approach, where the aim is to “win an argument”, and a lack of engagement, or very shallow engagement, is encouraged.

A related issue is that engaged conversation is difficult. It’s difficult because asking questions is difficult. For example, in substantive conversations about important matters people often make claims the truth of which is not established fact and so is open to doubt. But raising doubts about claims requires asking the right questions—questions which bring out the uncertainty of the claim—and identifying those questions is not easy. The same goes for asking questions which might reveal ambiguity in a term. This requires being able to imagine the two alternative meanings, and being able to articulate those alternatives, neither of which is easy.

Finally, people just may not want to engage, or even be aware of it as a conversational strategy. Consider, for example, the person who’s only talking to you to “prove they’re right”, or to win an argument. If they already think they have the answer and their goal is merely to convince you, then engagement isn’t the natural move. They won’t be inclined to carefully probe your own responses, but instead will merely throw at you a bunch of words they see as proving their point (without wondering whether you two are merely talking past each other, if they are actually right, etc). If all you’ve ever experienced is this confrontational style, then you might not even be aware that there’s another style of conversation.

Concluding thoughts

Genuine engagement is necessary in a number of important conversational contexts, but it is also a skill which most people have not cultivated or learned. A dismissive partner, in relationship counselling, being told they’re “not listening”, or a flat-Earther repeating a point already refuted, are vivid examples of people failing to genuinely engage—to the detriment of their ability to maintain a relationship, or understand basic facts about the physical world (respectively). The many ways of failing to engage (e.g., failing to listen to others, failing to respond to other’s reasoning, etc) are often individually presented as failures of critical thinking or empathy. But it’s useful, I’m suggesting here, to think of them not in isolation, but instead as instances of a broader failure: the failure to engage others in conversation.

Engaging in conversation