Gesture as language: why we point with a finger
Pointing at an object… in one sense you might say that this simple gesture doesn't just replace a word, but that it is a word—perhaps the first word. We know that it and other such gestures play a fundamental role in human language, but until now, we have not known where these gestures come from. To find out more, my colleagues and I investigated the hypothesis that pointing originates from touch.
I was interested in understanding why a pointing gesture picks out one object and not another, because I am also interested in how demonstrative words—words like this and that – pick out their objects. Demonstratives and pointing gestures are some of the simplest and earliest ways we have of "referring" to things, so that understanding these words and gestures gets us close to understanding the foundations of linguistic communication in general.
One evening at the Café Waikiki in Paris, Brent Strickland and I hyphothesized that we were pointing at objects not by creating arrows with our fingers, but as if we were "virtually touching" them in the distance. Brent, who works on gestural communication, had thought a lot about the angle the finger makes when it points, and how precisely it should be directed toward the objects to designate them. Another colleague, Gregor Kachel, had also worked on infants' understanding of others' pointing gestures. We decided to put our heads together and come up with some studies investigating the possibility that pointing originates in touch.
In our new paper, we discovered three things. First, that when people point at objects, they are inclined to orient their finger-tip as if they are aiming to touch the object they point at. The angle of their finger does not predict which object they point at—as we might be inclined to assume. Pointing gestures do not work like arrows, as street-signs do. Instead the line that connects the producer's eye and finger-tip is the best predictor of what they are pointing at. This suggests that pointing is somehow rooted in touch.
Second, we discovered that when we point at objects at odd angles, we are inclined to rotate our wrist as if we are trying to touch the object—even if it's far away. Imagine yourself pointing at the label on a bottle of wine when the label is facing to your right—you might find yourself rotating your wrist clockwise, as you would if you were trying to touch the label; or, if the bottle is rotated so that the label is on the left side, now you may find yourself rotating your wrist anti-clockwise again as you would if you were trying to touch it.
We found this effect even with 18-month-old infants—everyone seems to point as through trying to touch the things they point at. Finally we discovered that although adults can interpret pointing gestures as arrows, very young children and infants seem to have a hard time interpreting them as arrows, and instead treat pointing gestures as referring to the object closest to the finger-tip.
The work sheds light on the origin of human language. Something distinctive about human linguistic communication is that we deliberately "tell" each other things—that is, we inform each other in such a way that the other person knows we are trying to tell them something. It is widely thought that the first gesture or communicative act that infants undertake where it is clear that they are deliberately trying to draw their parents attention to something—to "tell them" about something—in this way, is the pointing gesture. I'm inclined to think of the pointing gestures produced by infants as their first "words", for these reasons.
The psychologist Jean Piaget speculated that pointing may come from reaching. But wanting to have something for ourselves (by grabbing it) is very different from telling someone else about it. You might point something out for someone, simply because you want them to know about it or see it. Think of pointing at the moon simply because you think it looks nice and you want someone to get to see it. You don't want them to hand the moon to you, you just want them to look at it.
In fact, infants produce gestures that are similar to pointing gestures—where they stretch out their arm and hold their hand out with all fingers splayed in a "grabbing" shape—when they want something handed to them. But independently, they also produce these distinctive "pointing" gestures—with the index finger extended and other fingers curled into the palm—that is, not in a reaching shape (since you can't pick something up with just an index finger). Our studies show that a much more plausible origin for pointing is in attempts to touch things.
We think that children discover that they can draw their parents' attention to things by touching them. Touch and visual attention are closely linked—we often look at what we touch, and, we think, parents and caretakers are inclined to look at what children touch more than what they grab or reach for. We think that once children discover that they can draw their carers' attention to things by touching, they "aim" to touch things in the distance, for the same purpose of drawing their carers' attention to those things. Once children discover this, a good deal of the time they spend focusing their own attention on nearby objects through touch becomes spent on establishing joint attention to objects further away, through pointing.
The very familiar scenario where two humans jointly attend to an object or event so that they talk about it is at the heart of human communication. The pointing gesture is the first event in child development where this "triangle" of attention between two speakers and an object is established. But until now we have not known where these gestures come from, and hence how this fundamentally important ability of humans to coordinate attention comes about. We think we have solved that riddle.
What is nice about these studies is that the results are scientifically important, but also at least in the case of the first two, can be checked immediately by any reader for herself. Most people we talk to are surprised to learn that they point at objects in the distance as if touching them, or that they rotate their wrists when pointing at objects at odd angles, but are inclined to quickly agree when they check. It is nice to discover something that was hiding in plain sight. The connection between pointing and touch, once identified, becomes hard not to see—something that's very satisfying to us as researchers.