Knowledge and Discourse
August 30, 2019
Stuart W. Mirsky

I was having a discussion the other day with a friend about whether animals without language can see themselves in a world that is beyond the present moment and environment in which they are. After I had suggested that they cannot because they lack a capacity for discursive thought, which requires a language, my friend demurred and pointed out that we can never know what animals know or what they are thinking because we can't get into their heads, read their minds. Well, I said, we can't get into one another's heads either, can we, and yet we know in the only important sense of knowing that they have minds rather like our own. After all, the mere fact of our discussion made it plain that I considered my friend to be a thinking subject, rather like myself, with ideas I could grasp and respond to. And he thought the same concerning me.

But you really can't know, he insisted, and perhaps at some point animal psychologists would be able to give us enough information so we could definitively state, one way or another, whether animals (like dogs and cats, rats and squirrels, primates and porpoises, etc.) had a sense of a wider world, one that extends beyond the moment in time and the locus in space in which they stand. He rejected out of hand, it seemed, my suggestion, that it is language that makes all that possible and that because of this we can do what dogs and cats can't, e.g., think about the days of the week, see ourselves as having a history and project the historical self we recognize as us into the future. Maybe they can, he said. You can't know. But maybe scientists can eventually tell us. And he promised to do further research on the matter and get back to me.

So how do we know what we know (or think we know when we think we know it)? Could there be any other way we could know such things besides the capacity to use language and thereby form concepts which reflect or depict the world as we conceive it? There is, of course, the history of this in philosophy to consider. Early modern philosophers in the rationalist school held that knowing was a matter of having certain ideas that are innate in us. We take the inputs we get from the world, our sensory inputs, and sort and organize them according to these innate concepts until they make sense to us. If we want to know something in a way that is indisputable we must look to the innate ideas through which we enstructure the raw data of our senses. While the sensory inputs we get can change, the innate ideas cannot because they are part of our nature, of who and what we are.

Against this view, of course, the English empiricists argued that sensory inputs alone were enough to enable us to construct relationships between what we receive and that, as we do that, through increasing contact with the world of sensation, we form concepts. Thus our concepts are not prior to our knowledge of the world but a function of it, produced by the "writing" of our experiences on the tabula rasa, the blank slate, of our minds. On this view, knowledge is entirely contingent and, had we different experiences we would have had different concepts. Nothing is given but the inputs themselves and they are given only in the sense that they are where everything starts. That they are as we experience them is contingent. But that they precede everything else is given.

There is, of course, another perspective, another possibility here, the pragmatic one, first formulated by C. S. Peirce, William James and John Dewey, the founders of American pragmatism but further explored by the later Wittgenstein who saw that what we do matters as much or more as what we say and that saying, at bottom, is just another form of doing. On this view, as Wittgenstein put it, meaning (the semantic dimension of our words) is in the use we put words to. Knowing that something is the case, knowing about it, rests on knowing how to deal with it, knowing how to act towards it. Here knowledge becomes activity, in this case recognizing what works and what doesn't. Knowledge on this view is successful interaction with the world.

But something is still left out. In our own experience we are aware of something else when we know something, or think we do. We are aware of the thoughts that occur in the process of knowing something, whether as prompts to act in particular cases or as accompanying phenomena, of a mental sort, when we do act. We aren't zombies, mindless automatons, even if knowledge is being able to act effectively. If I say I know what a word means I mean that I know how to use it to be sure. But don't I also have a variety of mental pictures, images called up in my mind which I associate with the sound of the words when spoken or the configuration of images on a written page? We may each have different images, to be sure, because our life histories will differ. But still we understand each other. How, if I hear the word "elephant" and my picture of it differs from yours? Well who says we must have the same picture? But must we not have similar ones?

That seems at least necessary in part. There must be sufficient similarity to connect the sounds I hear with the sounds you utter if we are to understand each other. But there is no reason why my elephant should look like yours. What we have to address then is the nature of such mental pictures. When we think about other subjects like ourselves we treat them as having a mental life as we do. Indeed, it is that shared quality that enables us to treat them as people rather than objects like rocks and trees and roads and bridges. We grant them a teleological dimension, that they reflect a purpose, and we do so because it is that dimension which enables us to make the most sense of the observed behaviors of those others.

As Daniel Dennett puts it, we see intentions and thus a mental life in others because this is the most effective way to deal with entities of that type. We don't see anything in particular though. We just see them in a certain way. We adopt what he calls the "intentional stance" towards them and treat them accordingly. We act toward them as though they have what we have, purposes, intentions, and the ideas of a world that underlie these.

Intentionality here is a matter of something going on in the system but what is going on is not something we see but something we impute because of the behaviors we see. Such behaviors, to be understood well enough to enable us to respond appropriately (which is to say effectively), enable us to construct a picture in our mind of their subjectness, their mental lives. a picture which helps us better understand them. Here our knowledge is plainly a function of our interaction with our world. We may explain this by reference to minds in others or to souls but it doesn't matter because the picture is not of what is actually there. It's just a useful means of talking, of imagining the world, or that part of it which they occupy, which we then rely on to get by.

And here knowledge can be seen as behavioral no less than activity can. But it is not divorced from the mental. It doesn't imply the non-existence of the mind, of the intentional. It only tells us these things aren't the same as the sensed phenomena we receive and compile neurologically into images of a world. But, again, this activity in the brains of creatures like us is "not," as Wittgenstein put it, "nothing either." The mental is real enough for without it we would not have creatures like ourselves which have needs and wants, desires and objectives. And we are here, aren't we? We do have us.

But it is easy to forget about the mental when we are explaining knowledge as practice, knowing that being underwritten by knowing how, as in knowing how to do the things the one said to be knowledgeable will do. The mental is not separable from the active even if we don't have to have the same images from person to person when we are engaged in discourse and seeking to understand the other. What is required, on the pragmatic concept of knowledge, is only a threshold of similarity which when crossed will prompt expected and familiar behaviors in the other. If we are to understand one another we need not have all our mental pictures in common but only enough to create enough associated similarities to generate the expected behaviors.

So do animals know their world in the way we do without the benefit of language? Are mental pictures enough? Well can animals without language have the sorts of pictures we have? Can they have the kind of linked images which constitute ideas of a broader domain than their immediate environment, of a road that leads somewhere else, of a town or city in which it runs, of a state or society in which the town is situated? Can animals have a world like we have, which extends beyond this time and place if they lack the linguistic tools needed to construct such pictures? Can my dog see me in his mind's eye sitting somewhere far away at a desk he has never seen, or even one he has? Granted dogs can miss us and pine when we are gone and perhaps, as my friend reminded me, greet me on my return with increased excitement if I have been gone a long time. But is that because he has been thinking about me somewhere else? Can he know there is a somewhere else or only a "not here now"?

Can we do any of these things in mental pictures without benefit of a language which can enable us to organize our inputs in ways that kick up pictures (based, of course, on things we have actually seen!)? Can dogs have a world like ours, one that is part of a larger landscape which we can think about even when we aren't there? Or imagine if we've never seen it? What can we know about the domain beyond our immediate environment in space and time without the ability to formulate such images and what could give us that ability besides language?

Indeed, can we even have concepts like space and time and knowledge at all without it?

Article originally appeared on Ludwig (
See website for complete article licensing information.