Does intelligence necessarily lead to an abstract language?

Question

Pretty much what it says in the question title.

Suppose a non-human, social species of Earth animal found itself on an evolutionary path that favored increased intelligence of a kind not completely dissimilar from that of humans.

These creatures, from the beginning, are able to plan ahead and at least to some degree anticipate the consequences of their actions. (It is not beyond them to reason like "If I do X, then do Y, then Z happens, which is good".)

Would such evolutionary pressure necessarily also lead to said species developing an abstract language of a similar kind as humans have in abundance?

For the purposes of this question, let's ignore what this species' language would be like; such a language, if it develops, could be based on any method of communication including vocalizations (sound), body language, or scent (or even telepathy or magic for all we care here). "Language" here pretty much just means "a method of communication between individuals".

Let's also ignore how humans would react to such a species and its evolution in such a direction. (In fact, if that makes you feel better, you can suppose that humans aren't around in this world.)

Seeking well-reasoned answers describing why an abstract language would or would not necessarily follow from a species achieving or evolving toward intelligence. Bonus points for answers that discuss how such a language might also evolve over time as the species evolves. Bonus points also for specific citations, but citations are not a substitute for answers being well-reasoned in their own right.

Answer 1

Human language evolved for the interaction between humans. That may sound like a trivial fact, but it isn't: If there had not been social interaction between humans, humans would not have evolved language.

Therefore the question basically reduces to: Does a species need to be social in order to develop an intelligence of similar levels as humans? Of course, given that we have very little experimental basis (as far as we know, th4e number if species with human-level intelligence on earth is, well, quite limited), it's hard to say for sure, but I would risk to say: Yes.

Why? Because natural interactions are relatively simple, including the interaction with possible prey if that prey is itself not that intelligent. Human intelligence certainly did not evolve in order to hunt; the main hunting strategy of the early humans was to outrun the prey (which is why humans are the best species when it comes to extended runs; other animals may run faster on short distances, but none would beat us at a marathon). A bit of intelligence is, of course, useful in hunting, but I doubt too much would be needed.

It is IMHO exactly social interactions that drove the intelligence; that is the place where you have to be able to think several levels deep, exactly because the other one does, too. If you are facing a non-intelligent being, all you need to know is: What will be the reaction. A bit of intelligence is sufficient for that. But to interact with your just-as-intelligent peers, it is important to be able to think one step further: "What does he think about me?" Also social interaction includes coordination as well as passing on knowledge. Both also profit from higher intelligence, as you need to understand the motives and plans of the other person in order to either coordinate with or learn from that person.

Note that coordination and teaching are also the main two areas where in language is useful, therefore I conclude the evolutionary pressure of developing intelligence and of developing language come together.

Also note that the structures needed for both have a very high overlap, as evidenced by the fact that a lot of our inner thoughts are a sort of inner monologue; indeed Ludwig Wittgenstein famously declared that the limits of language are also the limits of thought.

Answer 2

Abstract language will develop in that species, but only after they have crossed a certain level of self consciousness and inter-dependence is really high.

This has been decided on the basis of human history. It is impossible to state exactly when detailed language originated in humans, but it was certainly during a time when primitive societies were forming and humans had quit arboreal life and started living on ground level. It was a time when they were learning to work together for defense against terrible sabertooths of Africa (particularly Machairodus).

Such a stage is much more likely to arise in group-living animals living in the wild than lavishly living and well-cared-for pets and other domesticated animals. Almost all of evolution is driven by necessity and language is no exception.

Answer 3

Intelligence alone? No. Agriculture alone? No. Intelligence + Tool use + Cooperative Hunting + present-tense language? Yes. We can find many examples of intelligent life or tool users or cooperative hunting but we don't see any examples where all those traits come together except in humans.

So the question becomes, what long standing situation would force the ability to comprehend future-tense concepts? I think it's agriculture. Humans had already developed a strong cooperative hunting trait and extended that cooperation to other aspects of social life.

There are many animals that cooperatively hunt (the above link has an extensive list) but none that operate on a time scale beyond what can be encoded in an instinctual memory. For example hungry->get with group->go hunting->take prey->eat->wait->hungry. Or a longer term cycle, cold days->follow herds/move south->warm days->follow herds/move north. Animal instincts are sufficient to encode that kind of "future" information. Further, the animals don't need to convey that their juveniles. Going north and then south is just how life is.

Agriculture is different though. It requires knowing "if I put [seed in ground] and [water the ground] then I get more food and more food is good." Note that there are some animals that practice agriculture such as ants with aphids or fungus so developing agriculture alone isn't sufficient to develop an abstract language. The farmer ants and their fungus co-evolved together.

As pack hunters, humans already had a primitive present-tense only language for coordinating hunting attacks. Extending that language to conveying concepts about the future state of a crop isn't too large of a stretch.

Evolution of capabilities:

(format: initial state-[evolutionary pressure]->post-pressure state) Some capabilities will evolve in parallel. Generally, the more basic traits come first earlier in the list.

• solitary hunter gatherers-[group hunters do better than solitary hunters]->small group hunter gatherers (requires that cheaters don't profit from hunting)
• small group hunter gatherers-[prevent cheaters]->primitive social systems
• no communication system-[coordinated hunts fare better]->primitive communication systems (probably on the level of "go there" "throw spear!")
• wood tools-[stones hurt more than wood]->stone tool creation.

An observation that primitive grains tend to grow in a certain area might induce a band to stay in an area longer than they ordinarily might. Over a period of years, they might notice the effects of rain and drought on grain growth. A simple idea to divert water from a local stream to provide water to a wild wheat field might be all it takes to initiate agriculture.

Once they start planting the remaining seeds, that idea of planting a seed equals food later has to be conveyed somehow and that is an instance of a primitive future-tense idea. Dry and wet are also abstract ideas that are essential to successful farming.

Answer 4

It is essential for an intelligent species of the type you describe to be capable of abstract thought.

Why? Because they must form models of reality that allows them to form and evaluate hypotheses, and encode learning by modifying existing hypotheses which exist in a suitably compressed form, based on their actual experiences. The possible permutations of events an intelligent animal must face are so vast that compression via abstraction into probabilistic models is the only sensible approach.

Given that abstraction is a necessity, do they need abstract language as well? Assuming that the species is social, and part of the learning process involves transfer of existing abstract models from one member of the species to another, then the answer must be yes - our species needs a way of communicating abstract concepts between them.

In humans, since we are social from day one, possibly without abstract language development we could not actually develop and hold onto abstract models in our minds. In that sense, abstract language and thought probably co-evolved. However, if the species is not social from day one but can develop hypotheses about the world before they can communicate at all, then we would say that abstract thought precedes abstract language.

Here are the attributes of the species which I think need to apply to have abstract thought but not abstract language:

• The creatures learn independently from birth. No parental coaching.
• There is no communication of learned concepts between members. Immediate wants are communicated using abstraction symbols which are baked into the mental architecture at birth. So here the mechanics of communication are inherent to their phenotype rather than developed, I would not call this a language, more like a 'fixed protocol'.

Just because learned concepts are communicated does not mean they cannot be passed on - possibly they have some kind of lamarckian evolution which allows the biologically determined communication protocol to be passed onto offspring.