Comment by tsimionescu | Hacker News Reader

Comment by tsimionescu | Hacker News Reader

Parent Full thread

Language has pretty notable differences from animal communication that are actually very well studied.

One aspect is that language can string together small particles to produce arbitrarily complex meanings. In contrast, most animal communication is extremely simplistic - typically one sound or gesture corresponds with one piece of meaning, and the order or number of repetitions has no additional meanings. There may be some exceptions or at least grey areas (whale song, dolphin sound+dance, bee dances, some apes and birds), but the vast majority of animals have an extremely simplistic communication scheme.

Another huge difference between animal communication and human language is that animals communicate in response to direct internal or external stimuli (hungry - > chirp, see dangerous animal -> distress call). In contrast, humans communicate about much more complex feelings, and can choose when to exteriorize their language use.

Note that humans also communicate in ways that aren't language. Our "body language", our cries of pain, baby's cries, grimaces, winking, etc are the equivalent of animal communication and are not language. Sign langauge is absolutely langue though, and you'll notice it fully ticks the two boxes I was mentioning earlier.

Prairie dogs are another example that likely goes against your understanding. They also serve as an example of how we might miss complex communication in other species: researchers had to slow down their chirps and calls in order to find the patterns that made up their syntax and grammar. I imagine that LLMs will be useful in the future in helping us to find patterns in behavior or vocalization that might serve as language or language-like communication.

Also, in response to your second point: most people with domesticated pets could give examples of deception or signalling intent.

Perhaps indeed we are missing a lot of more complex communications. We will maybe learn more with LLMs, we'll see.

Related to deception or intent - those are still at best second order reactions to internal/external stimuli. The fact remains that dogs or even chimps don't, for example, tell stories to their friends about what a fun ball they saw yesterday or how sad they've been last week.

Nobody claimed that they did. The argument is that we have no reason to think these are categorical differences so much as differences of degree.

For example, there are clear relationships between a baby’s gesturing and their language development. These appear to be different enactments of the same systems (also people born blind will gesture while they speak).

Even the “at best second order reactions to internal/external stimuli.” Every human thought and action is at best an Nth order reaction to internal/external stimuli. Differences of degree.

But there is a vast gulf between humans, even those with severe developmental disabilities, and the smartest animals. Only a handful of animals can even pass the simplest knowledge to other, or learn by seeing others of their kind do something. Humans can't even function if we don't use language to communicate with others about things which aren't pressing concerns. There is plenty of room in that gulf to say THIS side is conscious and THAT side is not. Just like an analog-to-digital converter can get clean 0s and 1s out of a continuously varying noisy signal if it is sufficiently clean, but we don't claim that digital electronics doesn't exist just because there is some spectrum.

I do agree that there is some continuum of conscious experience between certain animals up to humans. I even believe there is an extremely remote but real possibility that other organisms such as plants, fungi, even bacteria, have some sliver of internal life that could resemble our consciousness.

But panpsychism goes WAY beyond this into a mush that makes no sense. Rocks and atoms very obviously are not conscious, and any non-religious theory of consciousness that posits that they have to be, even a tiny little bit, is in my opinion self-evdiently wrong (e.g. IIT).

Of course, if we want to posit that some form of god exists beyond physical phenomena, panpsychism is valid again.

>But there is a vast gulf between humans, even those with severe developmental disabilities, and the smartest animals. Only a handful of animals can even pass the simplest knowledge to other, or learn by seeing others of their kind do something.

You're not doing your argument any favors with this kind of hyperbole. Many animals mourn and communicate that mourning. Many animals have been shown to pass along information to others about helpful or hurtful humans, good places to hunt or forage, even elements of culture like behavior or personal calls. African wild dogs will die of a syndrome akin to heartbreak if separated from and unable to communicate with their group.

It's not too much to say that humans and other animals think and communicate differently. Most do not have the physiological capacity for certain types of advanced and abstract thought that are central to our own experience of reality. But that's a far cry from having no semblance of conscious cognition. Too many animals exhibit behavior indicating an awareness of the self as a separation from the other and the environment for your characterization to be unequivocally correct.

Sure there’s space for one side being conscious and one not being: tell me where the line is. Tell me what does it. Then tell me where the same line is between a cluster of cells and an embryo and a baby and an adult. When does consciousness “happen,” and where does it come from?

Reply on news.ycombinator.com