Koko the Gorilla seems to lie intentionally.

'She lies when it suits her. "Who broke the kitchen sink?" one of the staff asks her. Koko indicates another staff member, whirls about and starts to laugh.'

http://findarticles.com/p/articles/mi_m1252/is_12_131/ai_n85...

Even Alex the Parrot seemed to intentionally give wrong answers to questions when he was bored with a task.

Dogs and cats can and do lie, by trying to cover up things they've done (literally, usually), and by trying to direct your attention away from where it's covered up.

How do you define lying, though? Does the animal have intent to deceive, or does he just have a cause/effect memory that makes it behave in ways we anthropomorphize as lying?

I've seen some dogs do some pretty clever things, and I think they have more brains than most people give them credit for, but I don't think they intentionally deceive.

If you're asking if I think the dog is consciously lying, then no, because that presupposes consciousness (that is, that the dog is a person). Do I believe that the dog is engaging in the same behavior that in a person we would call "lying"? Yup. Intent only matters if there's someone to do the intending. Deception, however, is an activity that needs no intention -- it evolved before consciousness.

If it doesn't require intent, would you consider Batesian mimicry (http://en.wikipedia.org/wiki/Batesian_mimicry ) "lying"? Or does it have to be in the form of behaviour to be a lie?

(I'm not trying to argue a point here, just being curious. I'm not sure how I'd answer these myself, but I find this interesting to think about. Thanks for making me think :) )

I dunno. I'm not sure there's a bright line between Batesian mimicry and behavioral lying, but I haven't thought about it much.

I don't think it's obvious that something having "intent" is distinct from something having a sufficiently complex cause/effect memory engine.

Good point. That's partly what I'm trying to figure out.

My internal argument goes something like this: on one hand, if the behaviour is the same how can you claim the intent is different? But on the other hand, how do you define deceptive behaviour without intent?

Here's a thought experiment: suppose someone had no short-term memory but had a dog who was trained to ask (as dogs do) for his meal each day. If the dog decided one day to ask for an extra meal, is he lying, or is he responding to a reward? And if the answer is that he isn't lying, because the dog doesn't differentiate indicating he hasn't eaten (which would be a lie) and asking for a meal (which wouldn't), isn't this true of any way that a dog could "lie"?

Consider also the example of a cuckoo's chick. These things obviously have no "training". Genetics is the only thing they have to work with. And they "convince" doting mother birds to feed them extremely well (apparently, so well that other birds, just flying past the nest will sometimes give them food they were carrying back the their young). Is this a "lie" ?

Its sounds more like a clever hack. "Multi-mother food source". It'll be fixed in the next patch. Just keep logging on to steam.