Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

A fine arcticle, articulated aptly. I do take issue with this one sentence though:

> We are the only animal that can perform the miraculous feat of making the link between a specific vocal sound and an object that exists in the world.

I assume the author intended a more specific meaning (perhaps the assumption was that other animals communicate via a discernable "expressed emotional state" as opposed to actual words), but those words at face value are demonstrably false. All you have to do to prove it is look at any family of birds. Penguins are able to recognize their offspring's cries in a sea of cries on a crowded beach. Birds (and all creatures to an extent, who haven't yet gone extinct) have an amazing ability to recognize the sounds that their offspring produce, even when the final audial profile of that very specific sound is tainted by so many other competing profiles. If you've ever worked in sound design then you understand what an amazing accomplishment this is.

From a physical perspective, when two equal waveforms come together in a controlled space the resulting sound is doubled. By "resulting sound" I mean the final waveform produced by all sound producers in a space (a sound environment, in which each individual's perception of the environment is entirely dependent upon their position within that environment). And so imagine many birds sharing the same frequency as your child (we're birds now, for the rest of this argument), and those frequencies are overlapping - multiplying those particular portions of the resulting sound wave, and yet you are still able to find your kid. Animals can definitely make "the link between a specific vocal sound and an object that exists in the world."



Also see the 'smartest dog', who memorized the names of 1,022 individual toys and could make connections like 'ball' meaning any unfamiliar round object.

https://en.wikipedia.org/wiki/Chaser_(dog)

https://www.usatoday.com/story/news/nation/2013/11/24/smart-...


i have seen some videos of dogs pushing on voice activated buttons to 'talk'.

There was always possibility that dog was thought to press sequence of buttons to indicate something. A pavlovan reaction trained to look like dog is building a crude sentence.

But the Chaser studies are quite convincing that dogs are capable of linking speech to abstract general ideas. Its absolutely amazing.


My dog will grab my hand and put it on things he wants me to do “hands things” with: open a can of dog food, open the door, throw a ball, get stuff out of cracks/under things, etc. Although, most of the time he just wants pets.


My favourite (and it seems universal) is when they do something they know is wrong/forbidden, say destroy something.

You show them the thing asking 'whats this?' 'did you do it?'. They just look to the side pretending that they are not paying attention to you. Like you are not there, until they break and just try to hide :D

Its priceless.


What if, by "object", we mean not a specific instance of an object, but rather the whole category of objects? Do penguins or other animals link a specific vocal sound to "shell" or "children" or "nest"?


Dogs can definitely do that: teach a dog the meaning of 'ball' and some will then apply that to any round object that behaves more or less like the balls they're used to.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: