Communication and hypothetical thinking

Keith Frankish has an interesting article at Psyche pondering what ability separates modern humanity from archaic humans (such as homo erectus).  His vote is hypothetical thinking.  From the article:

The ability I mean is that of hypothetical thinking – the ability to detach one’s mind from the here and now, and consciously think about other possibilities. This is the key to sustained innovation and creativity, and to the development of art, science and technology. Archaic humans, in all probability, didn’t possess it. The static nature of their lifestyle suggests that they lived in the present, their attention locked on to the world, and their behaviour driven by habit and environmental stimuli. In the course of their daily activities, they might accidentally hit on a better way of doing something, and so gradually acquire new habits and skills, but they didn’t actively think up innovations for themselves.

This description of hypothetical thinking seems a lot like imagination in general.  However, I think the description of archaic human behavior vastly underestimates their abilities, or even that of mammals in general.  Being habit and stimuli driven seems more like a description of fish, amphibians, or arthropods.  It seems like we have enough accounts of imaginative forethought in mammals, birds, and cephalopods to conclude that this kind of thinking starts much earlier than modern humans.

Anyway, Frankish goes on to posit that it was the development of language which ultimately led to the development of hypothetical thought.  Citing Daniel Dor and Daniel Dennett, he describes a sequence where humans first learned to call each other’s attention to aspects of the environment, then to describe those aspects to each other in the absence of immediate stimuli.  Finally, we began talking with ourselves, instructing our own imagination, leading to the feeling of a private inner experience.

The entire hypothesis is interesting and deserves to be read in full.

However, aside from the issue I noted above, it feels anthropocentric to me, and I think it gets the sequence of developments wrong.  There are compelling reasons to think that episodic memory goes back to reptiles, and includes mammals and birds.  And episodic memory is imagination, just of the past.  When we remember a past episode, we’re essentially imagining, that is simulating, a past sequence of events.  The same machinery enables us to simulate a possible future sequence.

Personally, I think if we’re looking for the special sauce of modern humanity, it’s symbolic thought itself, along with the underlying deep recursive metacognitive framework.  Symbolic thought enables language, art, mathematics, and many other capabilities.  It’s arguably what allows a primate from the African savanna to ponder concepts far beyond its ecological niche.

It also supercharges imagination.  It is accurate to say that human hypothetical thought is magnified far beyond that of other species.  So in that sense, I think Frankish is right.  Most animals only seem able to plan a few seconds or minutes ahead.  Humans can do it days, weeks, months, or years into the future.  Doing that is extremely difficult if you don’t have symbolic concepts like “day”, “week”, “month”, etc.

So it seems like Frankish is on the right track, but takes a wrong turn with the language first aspect.  It’s worth noting that he’s far from alone in this view.  He cites Daniel Dennett, who does express similar views.  And other writers such as Joseph LeDoux and Frank Amthor also see language as foundational for human consciousness.

This view doesn’t seem right to me, but maybe I’m missing something?  Are there compelling reasons I’m overlooking for the language first hypothesis?

32 thoughts on “Communication and hypothetical thinking

  1. Re “Frankish goes on to posit that it was the development of language which ultimately led to the development of hypothetical thought.”

    This doesn’t explain the hunting patterns of some predators. These hunters hunt in groups with one group trying to herd prey to run into the other group. How do the predators in the waiting group understand they are to stand off and do nothing and let the others herd the prey their way?

    I can see how the technique is taught with older, wiser hunters taking young predators with them in that group and indicating they are to lie down and shut up. But how did this pattern get started? How was it communicated? Was it initially a chance occurrence and the predators learned from that? There seems to be at least some imagining involved here. And it doesn’t seem to be a manifestation of language … or instinct.

    Liked by 1 person

    1. Good point. I think that’s part of the compelling reasons I mentioned in the post. For wolves, lions, or hyenas to work together as they do on a hunt, definitely seems to require some level of forethought. It actually seems like the lack of language increases the amount of prediction needed since the other animals can’t clarify each other’s intentions.


  2. Evolution is your friend. The eye looks amazing, and figuring out how it evolves is difficult until you understand the iterative process of selection over many generations.

    Predators that could coordinate survived to reproduce, their behaviours self selecting from basic responses. Hard to get your head around, but think small. What’s the least behavioural change that sets the foundation for building on more complex behaviours.

    Liked by 2 people

    1. Hi Ashley,
      I think that’s definitely true. I guess the question is, are we seeing innate behavior hammered out over innumerable generations, or flexible behavior adapted by the organism itself? Is there learning involved? With mammals, I think we can say that a portion of the behavior is the learned flexible type.

      Admittedly, this is very species specific. But social species seems to have to navigate far more complex dynamics, dynamics which can change rapidly, requiring more intelligence than a lot of solitary behavior.

      Liked by 1 person

  3. There is so many things in this Frankish piece that I don’t like it is hard to know where to begin. I pretty much agree with a lot of your critique.

    He talks about Dennett’s idea of consciousness as a “virtual” system but then adds in language into that mix. Does that come from Dennett or is that Frankish?

    Whoever it comes from would limit the virtual system pretty much to humans who can talk to themselves. And the whole notion of people talking with themselves seems right out of Julian Jaynes. I do think people hearing voices in their head was important to societal control of behavior since it might have played a role in allowing humans outside of contact with other humans in authority to still obey that authority. It was important in the development of superego in Freudian terms. But I think it is ridiculous as a requirement for control of consciousness or consciousness itself.

    Also, the notion of Homo erectus living a “static” life probably is mostly wrong. During those million+ years of evolution the brain grew to nearly human size and Homo erectus spread as far as Indonesia (Java Man). So there must have something more going than eking out a living.

    Liked by 1 person

    1. Dennett does describe that concept all the way back in his 1991 book, Consciousness Explained. As Frankish notes though, it’s not original to him. Dennett presents it as speculation, not a firm theory. (He even admits it’s a “just so” story.)

      It does seem similar to Jaynes, or at least what I’ve heard about Jaynes’ theory. (I’ve never read his book, so I’m basing that on the brief summaries I’ve read about it.) Although this is seen as happening deep in prehistory, as opposed to Jaynes’ idea of it starting around Homer.

      The way Dennett describes it is to make language the medium through which the mind broadcasts content around, through which it “autostimulates” itself using the medium of internal speech. The idea is that the visual stuff follows after that, all through the Baldwin effect. This is usually where I get off, because the idea of the sensory and other primal broadcasts following the language ones seems….unlikely to me.

      I have read that Homo erectus culture, at least as revealed by the artifacts they left behind, was pretty static for long stretches of time. But I agree that they had to be more sophisticated than Frankish implies.

      Liked by 1 person

      1. It’s hard to tell about Homo erectus culture since they may have had many artifacts composed of wood or other perishable material. Something was going on with the brain growth but it is clear that major advances in weapons and tools didn’t come until the end of the period as far as we know now. New discoveries always seem to pop up to push back timelines and upset theories when somebody gets lucky and finds something new that survived.

        I think there is something in the Jaynes’ argument if you use his narrow human-centric view of consciousness but, like you, I think it happened earlier than Homer.


        1. Like many things, I think Jaynes’ argument is more plausible than it initially looks. It does focus in on a real phenomenon, that of people mistaking their own inner voice for something else communicating with them. That obviously still happens all the time today.

          I reviewed a book a while back by Robert Bellah, who identifies the development of what he calls “theorectic culture” with the Axial Age. Theoretic culture is thinking about thinking. I think this is what Jaynes was zeroing in on, but mistook it for development of a new type of mentality. In reality, we may have first become conscious that our minds weren’t indivisible things.

          The problem is the Axial Age is the first evidence we have of that kind of thinking. But it’s also a time when writing first proliferated. It’s hard to separate new concepts from ones that may have existed prior to then, but just never surfaced in the very limited writing that took place in pre-Axial societies.

          In any case, I agree that it’s a pretty narrow conception of consciousness. It’s not like people in 2000 BC, or even 100,000 BC, didn’t feel pain, and it seems very unlikely they didn’t know they were in pain when they were.

          Liked by 1 person

  4. I’m inclined to agree with your take, Mike. This paper from Malcolm MacIver discusses how planning may have started peaking as soon as animals got to the savannah, where there is lots of space, but also lots of places to hide.

    I also think that language was a consequent, as opposed to a cause, of what makes the difference.

    Question: [in the hopes of not having to do the research myself] what do you know about “working memory”. Is that a prefrontal thing? Do monkeys have it?


    Liked by 1 person

    1. For the MacIver paper link, it looks like WordPress may have ate it. But I’ve covered his views before (and largely agree with them):

      On working memory, that’s a complex (and controversial) question. The PFC is definitely involved in the executive aspects of it, but my sense is it’s a coordinating role, not necessarily one of storage, except the parts directly related to planning. The parietal regions also appear to be heavily involved. But there are a lot of those semantic pointers to other locations in the cortex. Which is to say, I think the contents of working memory are distributed throughout the cortex. I think it’s more about what is active at the moment and dominating activity.


      1. For the MacIver paper, google “Spatial planning with long visual range benefits escape from visual predators in complex naturalistic environments”. It’s an addition to his theme, but it hits the topic of your OP squarely, and just out last week or so.

        As for working memory, I can see/understand the *content* being all over, but I could also see the use of temporary localized storage of combinations of said content. Such combinations would allow for episodic storage, I think. I need to learn more about episodic memory as well.



        1. Just for bookmarking on the MacIver paper (and for anyone else interested):

          Episodic storage is interesting. The hippocampus is heavily involved in forming episodic memories, so much so that some people have proposed it as the seat of consciousness before. But once the episode is formed, it’s role seems to recede. It’s like it records a linked list of pointers, uses them to repeatedly fire all the assemblies involved, which causes them to solidify and associate enough to fire in sequence, then gradually retreats as a causal force for that episode.

          The question is whether there’s any one place that the sequence is permanently stored. I’m not sure there is. It seems like each episode is a cascade of representations, each triggering the next. Although it does seem like the concept of the individual episode itself needs to be stored somewhere, with alternate linkages between the individual representations.


  5. I will be boring and totally agree with you, Mike. You hit every nail on its head IMO. Language, and even more so writing and diagramming, supercharges imagination and reflection, but it’s not the horse. It’s the cart, and all mammals and birds (at the least) have the horse, which is imagination, aka hypothetical thinking. I love my fellow humans, but we don’t need to belittle our evolutionary cousins to feel good about ourselves.

    Liked by 2 people

    1. Hey, it’s okay for us to agree once in a while. 🙂

      I do think symbolic thinking makes a major difference, but we should be clear what that difference is. It enhances, dramatically, but it’s working with ancient material.


  6. One more thing on Homo erectus and language.

    I know some (Chomsky among them) think language emerged out of the blue about 70-100K years ago but I think that unlikely. If that is not right, that would mean Homo erectus and likely Neanderthals had some form of language. It may not have fully developed until near the end of the Homo erectus era. I would think it would have been some sort of intermediary between signalling found in other animals and fully expressive modern languages. It might have combined sounds, signals, gestures with a rudimentary pidgin like structure. It may have developed in part to instruct on tool making which allowed speech to get tied neurologically to tool making.

    I haven’t found a paper reference but here is an article about an anthropologist who believes the same.


    1. I agree that language, or at least proto-language, is most likely ancient. It may be that full syntactical language as we use it today doesn’t begin until the time Chomsky points to, but the idea that there was no language at all until then has always seemed implausible to me. I definitely think it came in stages.

      Still, something dramatic seems to have happened in the last 100,000 years. Most of the evidence we have for symbolic thought before then is scant and controversial, but in the last 100k years it becomes clear and gradually more pervasive.

      Although John Hawks once pointed out that may be due to decay of the relevant evidence.

      Liked by 1 person

  7. “And episodic memory is imagination, just of the past. When we remember a past episode, we’re essentially imagining, that is simulating, a past sequence of events.”

    I’m not sure I agree with that. I know people who can remember things just fine, but who have little or no imagination.

    Memories are (poorly) recorded facts, whereas imagination is unbounded. Memory is a matter of recall, whereas imagination is a matter of new construction.

    “Are there compelling reasons I’m overlooking for the language first hypothesis?”

    I don’t think so. I agree that symbolic thought seems key.

    One might ask why no other animals developed language akin to ours. If language came first, it seems like it should have.

    Liked by 1 person

    1. Is it a coincidence that the only animal that makes sophisticated tools also is the only animal that has a sophisticated communication system?

      We are unique in both regards and the tools came first.


      1. But sophisticated tools surely came after language. Many animals use tools of some kind, so I’m afraid I can’t see tools as the source of our intellectual development, but as effect.


    2. There’s actually pretty substantial evidence that episodic memory is more of a reconstruction mechanism than a retrieval one. Of course, it depends on semantic memory for what it reconstructs, but remembering a scene is reconstructing it. It’s why memories are so easy to contaminate just by the power of suggestion, and why older memories are particularly unreliable as a result.

      There’s also good evidence indicating that most of the neural circuitry involved in imagination is also involved in episodic memory. But it obviously can’t be a complete equivalence, since we usually know whether we’re planning something vs when remembering. And imagining some future sequence is usually done based on memory of past sequences, with adjustments for different actions we might take, implying that imagination is probably more sophisticated in the executive systems.

      But when you say know people with little or no imagination, I suspect you mean an ability or willingness to imagine novel or fantastic things. I know people like that myself, although in the ones I know, I tend to think it’s either or a combination of limited exposure to novel concepts and ideological rigidity. (In some cases, it’s related to decades of extensive alcohol or other drug use, but that usually effects memory too.)

      Good point on animals developing language. If language isn’t built on top of cognition, but enables it, why doesn’t it occur without that cognition?


      1. I don’t at all doubt some of the same parts of the brain are involved in both memory and imagination, but think of all the parts of our computer that are used in everything our computer does.

        As a highly creative person, I also know memory and imagination are as different as running a game versus a spreadsheet. Certainly lots of commonality at a mechanical level, but entirely different at a semantic level.

        “I know people like that myself, although in the ones I know, I tend to think it’s either or a combination of limited exposure to novel concepts and ideological rigidity.”

        Yet, I’d bet those people were capable of remembering their past. (Excluding those who’ve damaged themselves and, as you say, don’t have much recall, either.) Imagination is certainly, at least in part, a learned skill (although I also suspect there is something of an innate talent to it, as well).

        Another way to look at this is that lots of animals have memory. When I drive up, and my pal Bentley can see me through the window, she recognizes me on sight (and goes bananas 😀 ). But imagination isn’t really a part of animal mentality.


        1. For studies showing how closely related imagination and episodic memory are, google “imagination episodic memory”.

          You seem to be using the word “imagination” in a very specific manner. But if we take it as simply the ability to form mental imagery and related content in the absence of sensory stimuli, then all mammals seem to possess it. I shared a study a while back purporting to show that dogs had metacognition. Whether it actually showed that is debatable, but it’s one of many that definitely imply they can plan their next move, that is, imagine prospective action sequences and select among them.

          As I noted in the post, I think it’s more accurate to say that humans have imagination supercharged with symbolic thought. So animals don’t have the type of imagination you’re thinking of, but mammals and birds do seem to have basic imagination.


          1. As I said, I don’t at all doubt many of the same mechanisms are involved in both.

            And, yes, I am equating “imagination” more with “creativity” than “visualization”. As you say, it’s highly advanced in humans and essentially absent in animals.

            Having known abstract artists, I’m not sure it’s entirely based on symbolic thinking (unless we mean very general and abstract symbols). Consider Jackson Pollock, for instance. Most find his work quite compelling (I know I do; I love it). But no symbolism I can identify is within it or in its construction.

            Music, likewise, doesn’t always feel like symbol use to me. I used to sit in the dark and jam for hours, just making it up as I went along. It was a largely thought-free process (if anything, thought ruined it; it was necessary to be in the flow).

            Liked by 1 person

  8. A selection-pressure on the evolution of natural language competence had almost surely to do with conducing to a more reliable expression and grasp of communicative intent—thus conducing to a more reliable cultural evolution of knowledge and know-how—greater coordinated social inter- action and increase in practical/technological know-how . Yet there’s so much more behind expression and grasp of communicative intent than an evolving grammatical competence can account for. Grice, ofcourse, laid the foundations here. Trivers has something to say here too, Sperber more recently as well. Motives for speaking seem endless, just as “framing-effects” seem in pursuit of understanding.

    Liked by 1 person

    1. I agree that language likely has a number of causes and adaptive benefits.

      Can’t say I’m familiar with the work of the names you mentioned. (Or if I am, I’m unable to retrieve them at the moment.)


  9. A great deal of the exercise of our intelligence has to do with the question, “How did I get it so wrong and how can I do it better?” Enabling conditions abound—not least episodic memory. Some might ask: ” surely, though, there’s a hidden variable somewhere in this mess— whence really the capacity to reflectively explore remedial variations on a theme of past behavior? ” No, ’tis all entangled, and there’s no prospect of disentangling it. (this Comment is motivated by my latest failed attempt to fix a plumbing problem in my house).


Your thoughts?

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.