Lecture 06

Press the right key for the next slide (or swipe left)

also ...

Press the left key to go backwards (or swipe right)

Press n to toggle whether notes are shown (no equivalent if you don't have a keyboard)

Press m or double tap to see a menu of slides

core knowledge and social interaction

Our overall project is to understand something about the emergence of knowledge of minds, objects, colours and the rest in human development.

My proposal is that we have to take two factors into account. One is core knowledge, the other is social interaction.

Our current problem is to understand how abilities to communicate emerge in development and, hopefully, also to understand what role they might play in explaining how humans come to know things.

Here is my crude picture. (This doesn't do more that say that social interaction and core knowledge both matter.) You can see that communication with words plays a special role. Last time we tried to understand how humans first come to communicate with words, with the aim of identifying possible roles for language in explaining the developmental emergence of knowledge.

Yet Another Problem

We ended up with Yet Another Problem. The problem is, in short, that we don't know how children might first come to communicate with words.

To make progress with this problem it may be useful to switch from thinking about communication with words to thinking about non-verbal communication ... (But in the appendix I try to sketch a rough idea.)

This fits my overall plan which, as I mentioned before, is to work backwards from communication by language through non-verbal communication and so to understanding action.

non-verbal communication

Non-linguistic communication is special because it links to two of our ambitions. First, we want to better understand how humans acquire abilities to communicate by language, and these seem to be built on their non-linguistic abilities. Second, non-linguistic communication appears to be a paradigm manifestation of social intelligence, and perhaps one that doesn't already require any knowledge knowledge at all. So non-linguistic communication may be an aspect of social intelligence that can explain the origins of knowledge in development.

Pointing

Adults can use pointing to ...

request
inform
initiate joint engagement (‘Wow! That!’)

Tomasello calls the third kind 'declarative'.

Infants point from around 11 or 12 months of age, and point to all of these ends ...

From around 11 or 12 months of age infants spontaneously point to request, inform and initiate joint engagement \citep{Liszkowski:2007mm}.

Infants spontaneously point from around 12 months.

Here is a situation in which a child is playing at her table. Then something appears from behind a sheet. The infant spontaneously points at it.

Liszkowski et al 2004, figure 2

Infants point to intiate joint engagement Liszkowski et al 2006.

Why do infants point?

Liszkowski’s idea: find out by seeing what satisfies them.

hypotheses:

for themselves
(prediction: ignoring them makes no difference)
to draw attention to themselves
(prediction: looking at them is sufficient to satisfy them)
to direct attention only
(prediction: looking at the referent is sufficient)
to initiate joint engagement
(prediction: looking at them and the reference is sufficient)

‘Four hypotheses about what infants want when they point were tested. First, on the hypothesis that infants pointed for themselves (see above), E neither attended to the infant nor to the event (Ignore condition). Second, on Moore and D’Entremont’s (2001) hypothesis that infants do not want to direct attention and just want to obtain attention to themselves, E never looked at the event and instead attended to the infant’s face and emoted positively to it (Face condition). Third, on the hypothesis that infants just wanted to direct attention and nothing else, E only attended to the events (Event condition). Fourth, on our hypothesis that infants want to share attention and interest, E responded to an infant’s point by alternating gaze between the event and the infant, emoting positively about it (Joint Attention condition).’ \citep{Liszkowski:2007mm}

‘When interacting with an adult who always reacted consistently in one of four ways, 12-month-olds pointed most often across trials if the adult actively shared her attention and interest in the event (Joint Attention condition)’ \citep[p.\ 305]{liszkowski:2004_twelve}

‘Analyses of infants’ points within each event revealed a complementary set of results. In the conditions not involving joint attention, infants repeated their point more often. This repeating behavior presumably indic- ates that they were dissatisfied with the adult’s response, and so they were persisting in their pointing behavior hoping eventually to obtain the desired response (which was presumably joint attention, since children did not repeat themselves very often in this condition).’ \citep{Liszkowski:2007mm}

Now imagine an experiment with four conditions.

In each condition, there are several trials involving something appearing and, hopefully, the infant pointing at it.

So how do the conditions differ?

In one condition, the experimenter ignores the infant when she points.

In another condition, the experimenter looks at the infant only.

What predictions should we make?

If infants point to draw attention to themselves, what can we predict?

They should be more satisfied in these conditions.

They should be less satisfied in these conditions.

But how can we measure satisfaction?

Within a trial: less satisfied with response should lead to more pointing.

Across all trials: more satisfied with responses should make it more likely that pointing will occur in a trial (at least once).

If infants point to initiate joint engagement, what should we expect then?

Satisfied in this condition and not in any other.

So here's the setup again (but schematically this time).

Liszkowski et al 2004, figure 1

And here is the first key finding: more pointing overall (across trials) when there's joint attention.

[*todo: redraw Ulf's figures and put this & next on one slide]

Liszkowski et al 2004, figure 3

And here is the second key finding: less pointing within a trials when there's joint attention.

Liszkowski et al 2004, figure 4

12-month-old infants point not only to request but also to initiate joint engagement.

(And to inform.)

Why is this significant?

Because it implies two things:

First, it implies that infants' pointing is referential communication; that is, communication about an object.

(Contrast sharing a smile; we're communicating, but not necessarily referring.)

Second, it implies that infants have some understanding of joint engagement.

I'll come back to this in later.

A Puzzle about Pointing

So far it's been all very straightforward, but we're about to run into a puzzle.

The puzzle will help us to understand why someone might think that ‘infant pointing is best understood … is best understood … as depending on … shared intentionality’

a puzzle about pointing

\subsection{Why don’t ape’s point?}

Contrast apes with humans ...

‘there is not a single reliable observation, by any scientist anywhere, of one ape pointing for another’.

Tomasello 2006, p. 507

\citep[p.\ 507]{Tomasello:2010dy}

footnote: ‘There is actually one reported incident of a bonobo pointing for conspecifics in the wild (Veà and Sabater-Pi 1998)’

Tomasello 2006, footnote 1

‘Although some apes, especially those with extensive human contact, sometimes point imperatively for humans […],

‘no apes point declaratively ever.’

\citep[p.\ 510]{Tomasello:2010dy}

Tomasello 2006, p. 510

What does declaratively mean? Liszkowski and Tomasello call pointing declarative when its done to initiate joint engagement.

Why don’t apes point? (Tomasello’s question)

motor issues?

But they do gesture

understanding action?

But they are sensitive to facts about
the goals of others' actions.

So the discrepancy is not easily explained.

Comprehension is also missing ...

This question is the puzzle. Or, rather, it's half of the puzzle. (The other half is about why infants don't point until they're around 11 months old.)

A clue: apes don't comprehend declarative pointing ...

In this experiment, we contrast failed reaches with pointing ...

Hare and Call (\citeyear{hare_chimpanzees_2004}) contrast pointing with a failed reach as two ways of indicating which of two closed containers a reward is in. Chimps can easily interpret a failed reach but are stumped by the point to a closed container.

You are the subjects. This is what you saw (two conditions). Your task was to choose the container with the reward.

Infants can do this sort of task, it's really easy for them \citep{Behne:2005qh}. (And, incidentally, they distinguish communicative points from similar but non-communicative bodily configurations.)

The pictures in the figure stand for what participants, who were chimpanzees, saw.

The question was whether participants would be able to work out which of two containers concealed a reward.

In the condition depicted in the left panel, participants saw a chimpanzee trying but failing to reach for the correct container.

Participants had no problem getting the reward in this case, suggesting that they understood the goal of the failed reach.

In the condition depicted in the right panel, a human pointed at the correct container.

Participants did not get the reward in this case as often as in the failed reach case, suggesting that they failed to understand the goal of the pointing action.

(Actually the apes were above chance in using the point, just better in the failed reach condition. Hare et al comment ;chimpanzees can learn to exploit a pointing cue with some experience, as established by previous research (Povinelli et al. 1997; Call et al. 1998, 2000), and so by the time they engaged in this condition they had learned to use arm extension as a discriminative cue to the food’s location' \citep[p.\ 578]{hare_chimpanzees_2004}.)

\footnote{ The contrast between the two conditions is not due merely to the fact that one involves a human and the other a chimpanzee. Participants were also successful when the failed reach was executed by a human rather than another chimpanzee \citep[][experiment 1]{hare_chimpanzees_2004}. }

\textbf{Note that} chimpanzees do follow the point to a container \citep[see][p.\ 6]{Moll:2007gu}.

Hare & Tomasello 2004

‘to understand pointing, the subject needs to understand more than the individual goal-directed behaviour. She needs to understand that by pointing towards a location, the other attempts to communicate to her where a desired object is located; that the other tries to inform her about something that is relevant for her’

\citep[p.\ 6]{Moll:2007gu}.

Moll & Tomasello 2007

Why don’t apes point comprehend pointing gestures?

What we've said is about comprehending pointing.

But our question was, Why don't apes point?

I think we can answer both questions, the one about production and comprehension together.

I've taken the detour via comprehension only because I it's easier to see the answer in the case of comprehension.

Here's what we already have about comprehending pointing gestures.

This explains why apes don't comprehend pointing gestures --- they don't know (2) or (3) or both.

But what can we say about why they don't point? Think about what would be involved in producing a pointing gesture.

Informative pointing

To comprehend:

know that this person is pointing to location L;
know that by so pointing she is attempting to communicate; and
know that what she is attempting to communicate is that object X it at L.

Why don’t 3-month-olds point?

Tomasello also asks this question.

‘the specific behavioral form — distinctive hand shape with extended index finger — actually emerges reliably in infants as young as 3 months of age (Hannan & Fogel, 1987). […] why do infants not learn to use the extended index finger for these social functions at 3 – 6 months of age, but only at 12 months of age?’ \citep[p.\ 716]{Tomasello:2007fi}

(Again Tomasello's answer involves shared intentionality: it's because they don't understand shared intentionality until around their first birthdays.)

We can answer this in the same way --- they don't understand communication.

This makes sense of why chimpanzees don't point --- they don't understand communicative intention.

Why don’t apes point?

(And why don’t they understand declarative pointing?)

Because they fail to ‘understand that by pointing towards a location, the other attempts to communicate to her where a desired object is located’ (Moll & Tomasello 2007)

Note that this is an explanation which doesn't mention shared intentionality.

That's deliberate; sometimes Tomasello and colleagues answer this question by appeal to shared intentionality; I wanted to consider a simpler answer first and postpone thinking about shared intentionality for as long as possible.

Compare \citep[p.\ 516]{Tomasello:2010dy}: ‘they do not understand communicative intentions’

But

What is it to understand this?

Do 12-month-old humans understand it?

But is this consistent with the findings that 12-month-old infants do point?

What is it to understand that by this pointing action another intends to communicate that?

\subsection{pointing vs linguistic communication}

‘the most fundamental aspects of language that make it such a uniquely powerful form of human cognition and communication---joint attention, reference via perspectives, reference to absent entities, cooperative motives to help and to share, and other embodiments of shared intentionality---are already present in the humble act of infant pointing.’ \citep[p.\ 719]{Tomasello:2007fi}

‘cooperative communication does not depend on language, […] language depends on it.’ \citep[p.\ 720]{Tomasello:2007fi}

‘Pointing may […] represent a key transition, both phylogenetically and ontogenetically, from nonlinguistic to linguistic forms of human communication.’ \citep[p.\ 720]{Tomasello:2007fi}

appendix

I want to say a tiny bit more on what is involved in understanding a pointing gesture.

Suppose that we are doing puzzle. Then if I point to a piece, I probably intend you to do something with it in the context of our activity.

By contrast, if we are tidying up, a point to the same object might mean something different.

So:

Comprehending pointing is not just a matter of locking onto the thing pointed to; it also involves some sensitivity to context \citep[see][]{Liebal:2010lr}.

This is nicely brought out in a study by Christina Liebel and others.

\subsection{Pointing: referent and context}

pointing: referent and context

Liebal et al 2009, figure 1

Liebal et al 2009, figure 2

18-month-olds can do this, but 14-month-olds can't. (Don't infer anything from null result.)

Liebal et al 2009, figure 3

Liebal et al 2009, figure 4

‘Already by age 14 months, then, infants interpret communication cooperatively, from a shared rather than an egocentric perspective’ \citep[p.\ 269]{Liebal:2010lr}.

‘The fact that infants rely on shared experience even to interpret others’ nonverbal pointing gestures suggests that this ability is not specific to language but rather reflects a more general social-cognitive, pragmatic understanding of human cooperative communication’ \citep[p.\ 270]{Liebal:2010lr}.

What is a communicative action?

Why are we asking this question?

Let me try to explain it like this ...

Recall what we said about comprehending and producing pointing gestures ...

Informative pointing

To comprehend:

know that this person is pointing to location L;
know that by so pointing she is attempting to communicate; and
know that what she is attempting to communicate is that object X it at L.

To produce:

know how to point to location L;
know that by pointing to location L you can communicate with this audience;
know that what you can communicate is that object X is at L.

Both comprehending and producing require knowing things about communication.

To know something about communication you have to understand something about what it is, I suppose.

So to know what is involved in being able to produce and comprehend pointing gestures, we have to know something about what it is to communicate.

(This will also tell us what apes 6-month-old humans lack that prevents them from communicating with pointing gestures.)

Reminder of the question.

What is a communicative action?

Let’s start with some simple examples of non-communicative actions.

Ayesha hits Ben intending to bruise him.

Purely physical interaction.

Ayesha fakes a yawn intending to cause Ben to yawn.
Ayesha lays a trail of false footprints intended to deceive Ben.
Ayesha waves at Ben with the intention that he will recognise that she intends him to come over.

Why are these actions non-communicative? Ayesha intends her fake yawn to have an effect on Ben, but the effect is a physiological one. The response she wants from him is mechanical.

But there are also non-communicative actions which require a rational response from the people they’re directed at. For example:

Note here that although Ayesha intends to provide Ben with misinformation, her action isn’t communicative.

Intuitively, there’s a difference between deliberately providing information or misinformation to someone and communicating with her (Grice 1989: 218).

So what makes an action communicative?

Paul Grice has a neat answer to this question.

He notes that we sometimes achieve things merely by letting other people know that we intend to achieve them.

Waving is one of the simplest examples:

In this example, Ayesha’s goal is to get Ben to come over. Her means of achieving this is to get Ben to recognise that this is what she intends. So when she waves, her intention is that waving will let Ben know that she intends him to come over.

Goal: get Ben to come over

Means: get Ben to recognise that I intend to get Ben to come over

Intention: to get Ben to come over by means of getting Ben to recognise that I intend to get Ben to come over.

You can achieve some things just by letting people know that you intend to achieve them. To achieve things in this way is to perform an act of communication.

Note that, on this Gricean view, communicating involves having intentions about intentions.m

\subsection{A Gricean view}

First approximation: To communicate is to provide someone with evidence of an intention with the further intention of thereby fulfilling that intention.

(compare Grice 1989: chapter 14)

\citep[compare][chapter 14]{Grice:1989ha}.

To communicate, then, is to attempt to fulfil an intention by making it manifest to someone else that you have this intention. If you’ve studied Grice you’ll know that his analysis of meaning led to a long and boring series of counterexamples and refinements, most of which shed no light on the nature of linguistic communication (Schiffer 1987). But what’s really important about Grice isn’t the attempt to analyse meaning: it’s his insight

Recall the comprehension of pointing case; what is the confederate doing if she's pointing to inform?

The confederate means something in pointing at the left box if she intends:

\begin{enumerate}

\item that you open the left box;
\item that you recognize that she intends (1), that you open the left box; and
\item that your recognition that she intends (1) will be among your reasons for opening the left box.

(Compare Grice 1967 p. 151; Neale 1992 p. 544)

\end{enumerate}

(compare \citealp[p.\ 151]{Grice:1969pv}; \citealp[p.\ 544]{Neale:1992uw})

So to mean something by pointing, you have to have to have an intention about my recognition of an intention of yours concerning my reasons.

?

Pointing (and non-linguistic communication) involves intentions about recognizing intentions

One consequence of this would be that we can't appeal to non-linguistic communication in explaining the emergence of sophisticated forms of mindreading.

Why not? [Explain.]

Another consequence is an amazing discrepancy between knowledge of the mind and knowledge of the physical ...

2.5-year-olds look longer when experimenter removes the ball from behind the wrong door, but don't reach to the correct door

That barriers stop solid objects is not reflected in children's practical reasoning until they are about two years earlier.

If we suppose that children who can point have an understanding of communication as Grice understands it, then we are saying that they have a fabulously sophisticated model of the mental around two years before they understand the first thing about physical causation. They don't seem able to knowledgably identify causal interactions among unseen physical objects. Can we really be confident that it's easier for them to think knowledgably about intereactions involving mental states? (Maybe it is; we just don't have evidence, I think.)

Now this isn't an argument against the view that children have a Gricean understanding of communication.

But it does motivate looking for alternatives.

?

Pointing (and non-linguistic communication) involves intentions about recognizing intentions

But why are we even considering this idea, the idea that children have a Gricean understanding of communication?

Because it seems to be the view favoured by Tomasello, Carpenter, Liszkowski et al, who are leading experts on pointing ...

Tomasello takes infants' pointing to be based on what he calls shared intentionality.

shared intentionality

‘infant pointing is best understood---on many levels and in many ways---as depending on uniquely human skills and motivations for cooperation and shared intentionality, which enable such things as joint intentions and joint attention in truly collaborative interactions with others (Bratman, 1992; Searle, 1995).’

Tomasello et al (2007, p. 706)

\citep[p.\ 706]{Tomasello:2007fi}

It's hard to argue with the claim about cooperation; this is important.

But Tomasello doesn't stick with the notion of cooperation. Instead ...

There is this additional element, shared intentionality. I don't understand what it is, but Tomasello and his colleagues are extraordinay scientsits so I think it's worth exploring.

This (shared intentionality) is also the notion that gets us into trouble.

Here's what I take to be the view of Tomasello and colleagues.

Theory of communicative action (Tomasello et al?):

Theory of communicative action \citep[compare][]{Tomasello:2007fi}:

\begin{enumerate}

\item

Producing and understanding declarative pointing gestures constitutively involves embodying (?) shared intentionality.

\item

Embodying shared intentionality involves having knowledge about knowledge of your intentions about my intentions.

\end{enumerate}

No, I don't know what shared intentionality is either. I've asked you to find out in the last essay for this course.

Claims about development:

11- or 12-month-old infants produce and understand declarative pointing gestures.
Abilities to communicate play a role in explaning the emergence of knowledge of minds (among other things).

If the theory of communicative action is correct, then the claims about development are incompatible.

(Also, α rules out a conjecture about minimal theory of mind.)

So, apparently, if Tomasello et al are right, I'm wrong about two things:

First, I'm wrong that knowlegde of others' minds first emerges around three or four years of age and that one-year-old infants have only core knowledge of mental states.

And, seccond, I'm probably also wrong to think that abilities to communicate (whether by language or not) could explain the emergence of knowledge of others' minds.

This is one reason for asking, What is a communicative action?

?

Pointing (and non-linguistic communication) involves intentions about recognizing intentions

What are the alternatives?

I want to mention two alternatives ...

first alternative

\subsection{First alternative view}

One alternative is inspired by opponents of the claim, inspired by Grice, that communication by language involves identifying utterer's intentions.

Inspired by Grice, you might think that this is fundamental to linguistic communication.

But philosophers like Dummett and (for different reasons) Millikan reject this view.

Convention

‘No speaker needs to form any express intention … in order to mean by a word what it means in the language’

Dummett 1986, 473

\citep[p.\ 473]{Dummett:1986mq}

‘Interpreting speech does not require making any inferences or having any beliefs about words, let alone about speaker intentions’

Millikan 1984, p. 62

\citep[p.\ 62]{Millikan:1984ib}

We might try to provide an account of pointing in which it's not fundamentally a matter of intention at all.

This would be a radical departure from the Gricean view about pointing. But there is another alternative, one which is less radical.

second alternative

\subsection{Davidsonian view}

Like Grice:

‘meaning of whatever sort ultimately rests on intention’

Davidson 1992, p. 298

\citep[p.\ 298]{Davidson:1992pl}

We need to distinguish ulterior intentions from semantic intentions.

ulterior intentions

ulterior intentions: ‘intentions which lie as it were beyond the production of words … [such as] the intention of being elected mayor, of amusing a child, of warning a pilot of ice on the wings’ \citep[p.\ 298]{Davidson:1992pl}.

semantic intentions

semantic intentions: intentions concerning the meaning of one’s utterance.

Why does this distinction matter?

Grice’s explicates meaning and communication in terms of ulterior intentions.

His project is to give a reductive analysis of these notions, meaning and communication.

Ulterior intentions are precisely what is needed for such an analysis of meaning.

This is because ulterior intentions ‘do not involve language, in the sense that their description does not have to mention language’ or any semantic concepts like meaning \citep[p.\ 298]{Davidson:1992pl}.

But, Davidson points out, we don’t have to attempt an analysis of meaning and communication.

After all, Grice’s analysis has been subject to plenty of counterexamples and objections \citep{Schiffer:1987zb}.

(Davidson objects that ‘it is not clear that these principles [Grice’s] are designed to handle the gamut of examples we find in literature’ \citep[p.\ 300]{Davidson:1992pl}.

\citet{Davidson:1991ic} discusses one literary example at length. He argues that ‘Joyce takes us back to the foundations and origins of communication; he puts us in the situation of the jungle linguist trying to get the hand of a new language and a novel culture, to assume the perspective of someone who is an alien or an exile’ \citep[p.\ 11]{Davidson:1991ic}.)

‘The intention to be taken to mean what one wants to be taken to mean is, it seems to me, so clearly the only aim that is common to all verbal behaviour that it is hard for me to see how anyone can deny it.’

Davidson 1994, p. 11

\citep[p.\ 11]{Davidson:1994ol}

This aim ‘assumes the notion of meaning’, but it is important because ‘it provides a purpose which any speaker must have in speaking, and thus constitutes a norm against which speakers and others can measure the success of their verbal behavior.’ \citep[p.\ 11]{Davidson:1994ol}

*todo: this is linked to how Davidson distinguishes first meaning from pragmatic bits; see 'meaning is a psychological concept v2' (for Martin Davies)

But how does this idea translate into a claim about what a pointing action is?

First consider the wave from earlier ...

An example contrasting Grice and Davidson on the wave.

Grice

Goal: get Ben to come over

Means: get Ben to recognise that I intend to get Ben to come over

Intention: to get Ben to come over by means of getting Ben to recognise that I intend to get Ben to come over.

Davidson

Goal: get Ben to come over

Semantic Intention: that Ben take this wave to mean that he should come over.

Ulterior intention: that Ben come over.

These intentions have a means-end ordering; the ulterior intention is further down the means-end chain.

Strictly speaking, that Ben should come over might not be the first meaning of the wave (so there are other options here).

As mentioned before, Grice's view involves intentions about recognising intentions.

By contrast, Davidson's view requires an intention about meaning.

What this involves depends, of course, on how we understand meaning.

But maybe there is a way of understanding meaning on which this is not too demanding. (I'm really not sure.)

Now contrast Grice and Davidson on the pointing action from the Hare et al study, where you're supposed to take one of two containers.

Grice

Goal: get Ayesha to select the left container

Means: get Ayesha to recognise that I intend Ayesha to select the left container

Intention: to get Ayesha to selet the left container by means of getting Ayesha to recognisethat I am pointing to the left container with the intention that she select the left container.

Davidson

Goal: get Ayesha to select the left container

Semantic Intention: that Ayesha take this pointing gesture to refer to the left container

Ulterior Intention: that Ayesha select the left container

Strictly speaking, that Ben should come over might not be the first meaning of the wave (so there are other options here).

As before, there's a contrast in what must be intended and so what we're committing ourselves to in saying that infants can produce and comprehend informative pointing.

Can we do anything with this?

I haven't shown that we can.

It all depends on what an intention to refer is.

This is a really hard problem, and not one that I'm going to help you with directly (although I will come back to it in discussing action [analogy between principle of rationality: we need a comparable 'principle of reference'])

So I'm suggesting a possible direction but not providing any answers.

?

intention to refer

Conclusion for What is a communicative action?

The question was, Should we accept that pointing (and linguistic communication) involves intentions about intentions?

Should we accept that pointing (and linguistic communication) involves intentions about intentions?

If Grice (and Tomasello et al) are right about communication, then infant pointing involves sophisticated insights into others’ minds.
But there are alternatives to the Gricean story.
If Davidson’s* alternative is right, communication requires understanding intentions about meaning or reference but not necessarily sophisticated insights into others’ minds.

Is this enough to save the view about development I wanted to offer. Maybe, maybe not …

(*Davidson himself thinks all communication involves sophisticated insights into others’ minds …)

We have to say what meaning or reference is such that infants understand it. I want to leave this as an open problem. When we talk about action next time I'll provide a model for dealing with this sort of problem.

Conclusion for Pointing

11- and 12-month-old infants can point to (i) request, (ii) inform and (iii) initiate joint engagement.
... and they can understand these kinds of pointing gesture.
Understanding pointing requires more than associating gestures with their referents and understanding goal-directed action (see: why don’t apes & 6-month-olds point)
But what more is required?
Is it shared intentionality?

And what is that anyway? Is that something inspired by, and adapted from, Grice?

Words and Communicative Actions

Pointing vs. linguistic communication

a point

associate gesture with referent
identify the agent of the gesture
identify the intended recepient of the gesture
infer message (Why is this agent pointing at that thing for her?)

Recall this experiment in which Hare and Call (\citeyear{hare_chimpanzees_2004}) contrast pointing with a failed reach as two ways of indicating which of two closed containers a reward is in. Chimps can easily interpret a failed reach but are stumped by the point to a closed container.

\textbf{Note that} chimpanzees do follow the point to a container \citep[see][p.\ 6]{Moll:2007gu}.

Chimps do follow the point to the container, but they don't get the message.

Hare & Tomasello 2004

Tincoff and Jusczyk showed 6 month old infants two videos (not pictures: what you see here are stills from their videos) simultaneously.

While the videos were playing, the infants heard a word spoken. The word was either 'hand' or 'foot'.

Which video did they look at more?

Tincoff and Jusczyk 2011, figure 1

Here are Tincoff and Jusczyk's results.

They suggest that 6-month-olds can already associate some words with their referents.

But 6-month-old infants don't communicate, neither with words nor by pointing.

Tincoff and Jusczyk 2011, figure 1

Pointing vs. linguistic communication

a point

associate gesture with referent
identify the agent of the gesture
identify the intended recepient of the gesture
infer message (Why is this agent pointing at that thing for her?)

a one-word utterance

associate word with referent
identify the utterer
identify the intended audience
infer message (Why is this utterer talking to her about that thing?)

Recall our simple description of what is involved in communication.

What I'm suggesting is two very simple ideas. First, to be able to associate pointing gestures with their referents is not sufficient for understanding them; second, basic the pointing-referent associations may be grasped by non-communicators.

(Here the 'basic' qualifier is there because in some cases deciding on the referent may involve thinking about the agent and recepient, their knowledge and intentions.)

And likewise for the case of linguistic communication.

Origins of Mind

[email protected]

Pointing

A Puzzle about Pointing

What is a communicative action?

Words and Communicative Actions