some Rico correspondence

[Deanna in OR]

Back to the "Assumptions" (sorry, this is going to be a long one)--

 

There are a couple of assumptions that are not stated by Dr. Bloom that are implicit in his statement

"if Rico is really learning sound-meaning relations, as Kaminski et al. maintain, it should not matter who the speaker is."

Unspoken Assumption 1. "(Word) sounds are identical, independent of speaker."

An electronic analysis of the same, identical word said by 2 different people, even native speakers of the same language, would show that this is not a true statement. The electronic spectrum would be similar, but not identical. That's why we type these words instead of talk to our computers--word and language recognition by computers is a far more difficult challenge than science fiction authors thought it would be 50 years ago. I've tried using a language-recognition software with my computer and unless you have a disability that requires you use it, it is much slower and more clumsy than keyboarding. AND this kind of software isn't easily generalized--you have to teach the software how YOU say the words in its vocabulary, for each person who uses it.

 

(Caveat on this--there are limited-vocabulary, generalized voice-recognition applications used for phone menus now, but if you've ever had to work your way through one of these and say "two" or "yes" 3 times before the system understood you, you will know one meaning of "limited").

 

Certainly humans and dogs (and many other animals) learn to generalize the meaning of a sound over the subtle variations from one voice to the next. But I know from my business conversations with colleagues from other countries that these variations can challenge even the human brain (especially on the phone!).

 

Unspoken Assumption 2. "If the test subject has really learned the sound-meaning relations under investigation, the subject will always make the choice to demonstrate this with the associated behavior".

 

As we've discussed in detail here before, many higher-order organisms will make choices about their behavior based on many factors besides a voice command from a human. Even if the voice command is "understood" by the dog (or human), the context--the location, the dog's view of the relationship with that human voicing the command, as well as other motivators or distractors in the environment--can affect whether a dog responds to one person (or even the same person) in the same way, EVERY time (or even most of the time).

 

Context matters--Even a well-trained working dog would probably look at its human strangely if he said "Come by" in the living room.

 

Designing an experiment that also tests these assumptions would create a better understanding of how dogs learn and communicate with humans. For example, would a dog respond the same way to a voice recording of their primary handler, if the human weren't in the immediate vicinity? (Would a 7-year old kid?)

 

It's a fascinating topic, with lots of room to learn more.

 

Deanna

[Lunar]

Deanna, I completely agree with your summation of the assumptions here. The first one being something that seems to be overlooked by the scientists, which I find very odd. As someone else pointed out in this thread, hearing the same thing spoken in a language you don't know by two different people can sound so different you wouldn't know it's the same. Wouldn't a dog view our commands the same way? I know HOW I say a word matters to Zeeke just as much as which word it is.

 

I realize the limits that go into scientific testing - that in order to prove something they actually need to see it happen (thus if the dog doesn't respond to the commands, they can't prove the dog knows it, nor can they prove the dog doesn't). And I'm not saying the researcher can do anything about that. It's the people who take the research at face value and don't understand that there are a lot of variables that go into it - the ones who say "Oh my god! That's amazing, that's the smartest dog ever!", those are the ones who annoy me. No, that's not the smartest dog ever - it's a smart dog who is thus far the best at meeting the scientific rules for testing. 200 words doesn't seem to me like a huge number. Most of us here know we have very intelligent dogs - some of us brilliant dogs. But we know just as well that it probably wouldn't be able to be tested in any straightforward yes/no way.

 

Dogs just aren't so straightforward, and I think the assumptions about words and the way dogs understand them is too simplified. We assume dogs learn similar to us, same as we assume dogs feel similar to us. I was thinking about this earlier today - when we say a dog grieves for the loss of his/her sibling dog, we are able to understand how the dog feels, this gives us greater understanding and abilities to handle it. Just as we explain "barrier frustration" "seperation anxiety" "boredom" - all these are human emotions that we project on the dog. I'm not saying they're incorrect - they're probably very similar - but we don't really know. Same with learning and words. People who don't really think about it, who don't read books and don't really study their dogs, tend to get very frustrated when they try teaching their dogs, because dogs just don't learn the way humans might expect them to. Someone mentioned generalization - that some dogs will learn sit and know that sit means sit no matter where you are. Other dogs need to learn sit in the house, in the yard, in a parking lot, in a strange building... That doesn't mean either dog is more intelligent than the other, they just have different learning styles. Same as people, schools are finally recognizing that some people learn visually, others aurally, others by experience. Any means of standardized testing will end up making erroneous assuptions about learning, and will be inherently flawed. And yet, with dogs and schools, it's the best way of testing and comparison available at this time.

 

Sorry for being long-winded.

[SoloRiver]

Hi Eileen,

 

Sorry, I was sloppy and imprecise. I wasn't responding directly to what you wrote (though I quoted you) but more generally. Just pointing out that science proceeds by hypothesis refutation.

 

Bloom wasn't saying that dogs can't learn words. What he was saying was that so far the evidence has not shown that dogs understand words the same way humans do. More on that in a moment.

 

Kaminski et al., for their part, were looking at the limited hypothesis of whether or not Rico, this particular dog, can learn words by fast-mapping, a facility previously thought to be limited to humans. I think they did show that Rico can do this. They then go on to offer that if dogs can in fact do this thing which was thought to be unique to humans, perhaps they also have other human-like communication capacities, but they don't say that they think dogs have language in a human sense. This further question is what Bloom was talking about.

 

Animal language studies are usually conducted in order to understand how human language evolved. Language doesn't fossilize, so the only way we can look at its evolution is to study our animal relatives. The fact that dogs, who are relatively distantly related to us, seem to have some important foundation capacities that we built upon on our way to full-fledged language capability suggests that these capacities are very ancient. That's what's most interesting about Kaminski et al.

 

Human language is (we think) unique in that it is symbolic, so the question is always, "When did symbolic language evolve?" For animals to have human-like language abilities, they must understand words as symbols. This is what Bloom was talking about when he has his little discussion about socks. Rico may know the word "sock" but not understand it the way a human understands "sock."

 

Words are signs, "stimulus patterns that have meaning" or stand for something. There are three kinds of signs: icons, indexes, and symbols.

 

http://www.cs.indiana.edu/~port/teach/103/...mbol.short.html

 

An icon stands for something by physically resembling it: a picture of a dog. I think words for sounds can be iconic ("meow," "boom"), but I can't remember if onomatopoeia technically counts as iconic. The word "oh" is sort of iconic if it refers to a facial expression.

 

An index stands for something because it is associated with the presence of what it refers to or with a certain context. Smoke is an index of fire. Picking up the leash is an index for going for a walk. Cues used in dog training are most often indexical.

 

A symbol stands for something for no reason at all. It is totally arbitrary and does not depend on a correlation in space or time to be learned or to have meaning. There is no particular reason that the word "sheep" should stand for a category of animals with particular characteristics, and you can teach a child what a "sheep" is without the kid ever seeing one. You can discuss "sheep" in an abstract sense. Human language has words that are iconic and indexical, but is made up mostly of symbols.

 

Bloom was pointing out that, so far, it has not been shown that dogs understand words in a symbolic sense, the way that humans do. He was saying that while dogs may learn word meanings, they may not be the same kinds of meanings that humans learn. Not that dogs cannot respond to words or understand human desires communicated through words.

 

I know that my dogs understand words in an indexical sense. I think they probably do not understand them in a symbolic sense. That doesn't mean they don't understand me or we don't communicate with each other, just that we don't communicate the same way I would with another human (which is fine with me -- their non-humanity is one of the things I like best about dogs). When I say "ball" to Solo, he gets excited because the word "ball" is normally an index indicating that a game of ball is about to commence. He will look in my hand, or search for the ball. He probably does not sit around thinking in an abstract sense about how much he likes balls. We cannot have a conversation about that green ball he played with yesterday or the red ball I bought to play with tomorrow.

 

The fact that dogs are so dependent on context to understand words or commands is an argument against human-like language capacities (it suggests that words are understood as index, not symbol), not for it. Symbols have meaning when divorced from context. The fact that "come bye" has no meaning except around sheep (unless you're of the ball and lawn chairs school ) shows that dogs are brilliant contextual learners who nevertheless may not have human language.

 

When Bloom says that it shouldn't matter who is talking, I think he was talking about this. He wasn't saying that dogs can't learn words or understand them in some way, but that without showing that context does not matter one cannot support eliminate the hypothesis that they are simply indexes, and therefore cannot claim that dogs understand words the way humans do.

[PennyT]

I think Deanna and North49 and others who made the point about context are

right. Who the speaker is matters for people, too, and Paul Bloom does not dispute that exactly.

 

I

suspect that Paul Bloom's statement about knowing the word "milk" was an

intentional oversimplification. As Melanie said in her first post on the subject, the

oversimplification is not being extended to Rico because the issue is

whether dogs learn words the same way children do. I suppose (but am not

positive)

that in studies with children movies or pictures are used to eliminate the

big scary guy

or wimpy babysitter or Cree accent

factors.

 

Movies probably won't work for most border collies because only a small

percentage pay attention to what is on a television screen. I have seen a

border collie, who belonged to a friend, run around the back of a television

set to lift the sheep or

other animals in a video or t.v. program, then paw the screen (the owner had

to attach Plexiglas in front eventually to avoid more scratches) and finally

settle down to watch. I had a border collie who seemed to understand

that the animals weren't really in the box and watched with quiet, intense

interest. Most people I know who let their border collies inside

the house have a story or two like these.

 

However, for experimental

purposes, using videos or movies or recorded voices is highly unlikely to be

at all successful because most

dogs haven't a clue what's on t.v. (thank goodness) and need a working

relationship with a

live person in order to do their best when what is required fairly complex

or not typical

doggy behavior.

 

Amy summed up the matter brilliantly. I wish I knew the source of

her quote. Amy, do you remember?

 

It's safe to say that Rico would not have learned to identify all those

objects from a verbal command and to demonstrate some kind of learning by

process of elimination completely on his own. A human helped him along from

the time he was under a year old.

 

Over

how ever many thousands of years of

domestication,

a large number (not 100% but enough to have an impact) of the dogs living in

some association with people survived

to reproduce by doing

something or not doing something at the direction, whether spoken or tacit,

of familiar people. For example: Guard and help move the livestock = live to

reproduce. Kill the livestock and bite the human pack = don't live to

reproduce. Guard the human pack and warn that strangers approach = live to

reproduce. Fail to guard from strangers and fail to warn = don't live to

reproduce. For less complicated behaviors, like figuring out where the human

put the food on the basis of social cues or adjusting to a new human or

humans as the pack, survival (living long enough to produce young) would be

enhanced by responding to anyone. And then again, maybe it was all a happy

coincidence.

 

If Rico must show that it does not matter to him who the speaker is to show

normal word learning, then he

probably has to cooperate at object identification with a stranger. If he

will not do

so at something close to the same level as in the study, then I have the

uneasy feeling some variation of the standard textbook version of Clever

Hans

will be exhumed as the

explanation.

 

What other explanations are there besides elevating the definition of

language use as

Chomksy did to require syntax back when those first studies of teaching apes

sign language were done? I doubt that would fly with the Rico

experiment because one of problems with the experiment seems to be that

Rico may only understand words in terms of a particular syntax. (I have no

idea if the

no-syntax-means-no-language objection to concluding apes learned some sign

language is still considered viable. It just came to mind.)

 

There may be ways to show Rico (or

another dog because Rico is already 9 and these things take time)

understands sound-meaning relationships that would avoid trying to order

someone else's dog around and

at the same time meet the standard of not it mattering to the dog who the

speaker is.

 

Dr. Bloom does acknowledge that who the speaker is can matter to both

dogs and humans. He also said to me: "I entirely agree with

your main point, which is that attention-to-who-the-speaker-is

can be a sign of intelligence, in both dogs and humans."

 

That was not my main point. I didn't quote that initially because it seemed

unkind to have to clarify by saying that Dr. Bloom seemed not to have

understood my main point or chosen not to acknowledge it.

 

My main point was what most of the posters on this board have stated: who

the speaker is can and often does matter to a dog or human

without the dog's or human's skill at a task ever being called into

question. Dogs who are very good at something can vary in how and whether

they perform based on how

seriously they take a person and on how well dog and person know each other

and on how good the person is at getting dogs to do things.

 

Dr. Bloom did say "if Rico

really is learning sound-meaning relations...it should not matter who the

speaker is." He also said specifically to me, "Once

you know that 'milk' refers to milk,

you understand the word even if its spoken by a stranger.

This is true of any normal 2-year-old child. If it is not true

for Rico, then he is not learning words in any normal sense of the

term."

 

So it is very clear that Dr. Bloom thinks Rico has to show understanding of

a word "even if it is spoken by a stranger" as one of the things necessary

to

prove the learning of a word in a normal sense.

 

Suppose Ricotta (I'm naming a hypothetical daughter of Rico's) learns to

identify 200 objects for her owner but has not been in the kind of

experimental setting Rico was. The owner then takes Ricotta to the place

where the experiment is going to be done, leaves Ricotta, and goes home.

 

Researchers Ricotta doesn't know come in. The set-up for the Rico experiment

is otherwise duplicated.

Ricotta only brings her favorite toys or none at all or various objects from

10% to 60% or even 70% of the time but doesn't do at all well with

unfamiliar items. Would it be reasonable to conclude from any of those

results that the dog has not learned words in any normal sense of the term?

 

How about if Ricotta's performance for the new researcher starts improving

over a series of trials or even toward the end of the first session?

Wouldn't that be discounted because familiarity with the speaker had entered

into the situation? If Ricotta's performance for a stranger improves as dog

and person get to know each other (and it probably would), would it still be

reasonable to conclude the dog hadn't learned words in any normal sense of

the term?

 

How about if Ricotta identifies objects pretty well for a stranger when her

owner leaves but

not well or maybe not at all with the owner present? That's fairly likely but strictly speaking why should it matter?

 

Or suppose Ricotta's owner accustoms her to fetching objects or identifying

them in some other way for strangers, and Ricotta then performs up to her

father's standard for a new stranger in an experiment. Wouldn't the

conclusion

then have to be that the only dogs who learn words normally are dogs

familiar with working for strangers and thus to not caring who the speaker

is? This scenario seems to me to be the likeliest way to get a dog not to

care much about who the speaker is. It seems odd, however, to weld the idea

of normal word learning to habituation to cooperation with strangers, or

maybe that is the basis of domestication and in a broader sense of the

social

contract and civilization.

 

Regardless of my expansive leaps in defense of Dr. Bloom's position, he

either does

not know much about dogs or is intentionally setting up at least one null

hypothesis (I do know what one is; I just don't think that stance matters

much in terms of commentary as opposed to research.) that, unlike the

who-is-the-speaker question, any high school student needing a science fair

project could do.

 

The last sentence in the piece in Science is that

until several issues are resolved "it is

too early to give up on the view that babies learn words and dogs do not."

 

One of the issues Dr. Bloom poses is, can Rico "learn a word for something other than a

small, fetchable object?"

 

I realize that people in the sciences think some of my notions are

laughable. But how can anyone who owns and interacts with their dogs not

think that question is amusing?

 

Is there anyone who has responded to this thread and who works their border

collies on sheep who genuinely believes that their dogs do not know the word

"sheep?"

 

Sheep are fetchable (although not usually fetched in the sense Dr. Bloom

uses the

word) but hardly small.

 

Most of my currently working, not retired dogs know the difference between

"Where are the sheep? Show me

sheep." and various ways of saying "Bring the sheep to me."

 

Or how about the word "car" or the phrase "go for a ride?"

 

I hope it's obvious that I do not think a dog and two or three year-old

child are comparable in

intelligence. I would have to be convinced by a lot more than what Rico can

do. Then again, I don't have children so maybe I give them too much credit.

 

Edit: Melanie, in the piece in Perspectives, Dr. Bloom never once suggests what you do about meaning as necessary in your most recent post.

 

Penny

[Bill Fosher]

When we talk about comparing intelligence across species, we start to become like a fabled invention of the US Naval Air Corps called the Clang Bird. This contraption flew in circles of ever-diminishing radius (radii?) until eventually it flew into its own tailpipe and crashed on the ground.

 

Show me a two year old child who can gather up sheep and read and respond to their intentions well enough to put them in a pen. Even setting aside questions of physical ability and coordination, I don't think a toddler could do that most mundane task for a Border collie. Does that mean that the Border collie is more intelligent than a two year old? No. Because the two year old has intellectual capacities that the Border collie will never have, ranging from abstract logic to empathy and visualization.

 

But at the same time, I think it's a mistake to maintain that humans are more intelligent than other animals. We are certainly better at establishing measurements by which we can win intelligence contests, to be sure, but there are lots of things that we can't do that animals can. We write these abilities off to instinct or survival, but to be perfectly candid about it, a dog's ability to anticipate approaching storms could be learned just as surely as we learn what time of day the mail is usually delivered. We simply don't understand what's going on because we are insensitive to some of the information the dog is picking up.

 

Human beings have enormous intelligence, to be sure. We are the only species so far that has been able to engineer the means to put itself out of existence. And to have come up with so many options for how to do it! What brains we have!

 

Different is surely not lesser.

 

-------------

 

On another topic, I once heard a collection of carefully edited sound bites taken out of spoken sentences. The bites were exactly one word, extracted from normal conversation, recorded in different parts of the US and Canada. Without the context clues, it was very difficult, if not impossible, to tell what the words were, and even whether some of them were the same words.

 

"Towel" was one that struck me the most. I think they had five clips of men and women uttering this word as part of a sentence. They were all reading the same sentence, something like "After his shower, he used a fresh towel to dry himself."

 

There were pronunciations ranging from "tile" and "tall" to "tow-ell" to "towl" (rhyming with foul). Most were single syllables. So suppose Rico had been taught to seek a "tile" and been praised when he came back with a piece of terrycloth. Then I come along and tell him to get a "tow-ell" and he comes up empty or selects the new thing rather than the familiar terrycloth becuase he's never heard of a "tow-ell" before, so I obviously don't mean the tile.

 

Yet put those words in context -- "After his shower, he used a fresh towel to dry himself" -- and no speaker of the English language would have a problem understanding what was being said.

 

I'm sure that the study could be set up to somehow control for dialect, but this is only the grossest example. When we understand so little about how dogs hear and understand our words, I think it would be pretty hard to deduce anything one way or the other from lack of performance.

 

What's needed is a way for a dog to ask a question back. When I say "tow-ell" and Rico doesn't understand, he should be able to ask for help. "Do you mean 'tile?' "

 

Of course, we all know the "what are you talking about" look that our dogs give when they are asking for help or thinking we must have flipped our lids. Do you suppose you could get a behavioral scientist to recognize that?

[PennyT]

I have just finished mowing the dog yard and under trees in my front field

where the regular tractor can't go because it will damage tree roots. Hay mowing goes clockwise; getting rid of the grass mowing goes counterclockwise in circles of ever

decreasing size like Bill's Clanger. Before that, I was helping someone

learn to work a border collie on sheep. More circles, only these were

ever-widening.

 

There is nothing I feel more like

doing right now than comparing the intelligence of various species in

concentric mental circles with my sweaty body inside in

the air conditioning and a tall glass of iced tea in easy reach.

 

Saying that the intelligence of one species is neither better nor worse but

only different from that of another species begs the all the questions. It's

true but so what? Questions about which species can do what and at what

level are good, useful, and interesting.

 

I think (make that I hope) it has been quite a few decades since children

were taught a sort of textbook mantra that the abilities of nonhuman animals

"can be written off as instinct or survival" while human consciousness and

reasoning are something wholly other. I know my father was taught to and did

write off abilities in nonhuman animals in the way Bill stated. My father

was born in 1905 and died in 1999. He was a brilliant man, one of the giants

in his field, but he never gave up his belief in that claptrap.

 

We used to

argue about it regularly. One of my favorite queries was, when a bear

stalked by a hunter or photographer turns the tables and does the stalking,

which without question does happen, why is the human thought of as thinking

but the bear thought of as acting on instinct? Why aren't both or neither or

either part of the time thinking or acting instinctually? My father's answer

was that bears only act on instinct. He did eventually come to

understand that survival of the fittest means surviving to produce young.

Convincing him was not easy. Yet, he never gave in and agreed that conscious

choices can or might be made by nonhuman animals. His faith in this was not

religious in any "normal" sense of the word since was an atheist, but it

always struck me as religious.

 

I was not taught as he was in school. By the time I was in college whether

nonhuman animals only acted instinctually was up for grabs or maybe past

that because I wasn't paying enough attention.

 

Bill also noted: "Of course, we all know the "what are you talking about"

look that our dogs give when they are asking for help or thinking we must

have flipped our lids. Do you suppose you could get a behavioral scientist

to recognize that?"

 

That's an interesting point because when they test human infants they sure import meaning into a startle response or looking longer at something .

[Bill Fosher]

Well, I didn't take too many psychology courses that delved into animal behavior when I was in school, but there was no doubt in my mind that the professors that did touch on the subject were very much old school. Animals have no free will -- they simply obey instincts. Get food, find a mate, etc. To them, the interesting question was how abstract a behavior could be for them to learn to associate it with obeying an instinct. For instance, could a rat learn that when it presses a button with a triangle on it, it gets a peanut and when it presses a button with a circle on it, it gets nothing.

 

Similarly, when primates were being taught to sign it was generally with food rewards. There was little signing for its own sake, and the scientists were amazed to discover that some apes would "talk to themselves" -- or sign when there was no one around to give them treats for doing so.

 

It's all based on the underlying presmise that these animals are performing a behavior only because they have been rewarded for it in the past, and are anticipating a reward for performing it now -- or at least have learned to associate it with a reward. (Or, alternatively, are performing or abstaining from a behavior to avoid correction.) And the reward is always presumed to be following an instinct.

 

We see this in the sheepdog training on both sides of the divide between bizarro world and normalcy. In bizarro world, treats are chucked at the dog when it's where the handler wants it to be, clickers are clicked, and the dog learns a behavior. In the normal world, the dog is allowed to work away when he's doing right.

 

In both cases, the presumption is that dogs are serving an instinct: in bizarro world, it's to get a treat. In the normal world, it's to work sheep.

 

The great trainers of sheepdogs understand that there's some nuance to this whole process: that dogs are choosing all the time, and part of what we are doing in the training process is helping guide them toward a mutually beneficial set of choices. Dogs are not slaves to their instincts.

[Tuck's BCBuddy]

Bill,

You wouldn't, by chance, be a superman fan, would you? LOL

[sea4th]

I wonder how those who feel that animals function solely on instinct explain the necessity for decisionmaking capabilities in a guide dog for example? That's probably something more within the realm of understanding that the average person can connect with, rather than the fine nuances of a working sheepdog, which even many "dog" people cannot understand. I've posed this question for years to different people who thought dogs functioned on one level up from an amoeba. Those who seemed to have all the answers for everything else, couldn't come up with an answer for that. I never got the chance to pose this question to anyone from the scientific community, but it sure left some food for thought for anyone else I spoke with. Would you trust the life of a blind person to something that functions solely on instinct? Conversations of "well, my dog did this", or "I watched my dog do this", are pretty much flippantly written off as something from a slightly "left-of-center" pet owner, and so don't really hold up. And the nuances of working livestock with a dog like a border collie is even beyond the understanding of most "dog" people. I took an obedience class one time with one of my border collies given by an instructor that used food as a motivator. She had goldens. I refused to use food and tried to explain to her that I knew my dog, that a job well done whas the dog's reward. I was told that "it's her way or the highway". I hit the highway.

 

I have a friend who had an "alternate" herding breed. She worked this dog on sheep. It took her 7 yrs. to get this dog to the point to do what her first working-bred border collie was doing at 5 months of age, and the foundation of the other dogs training was food rewards. For her border collie, food was a distraction from the task at hand. She marvaled at that.

 

One time, we rode out together to work our dogs. I told her to take Tam out and work him, that he could teach her a lot by her taking him out there.

Before she stepped out onto the pasture with him, she gave him a treat. The expression on his face was one of "What the h---!". He glared at her and refused the food. She didn't understand, but I could sure read what was going through my dog's head.

 

I've also gotten into discussions at family gatherings (I'm the dog "nut" in the family--that automatically makes me weird). The so called intellectuals in the family would often remind me how much more intelligent apes were than dogs---knowing these people, there was the implication with this statement that dogs were nothing more than eating and "waste-producing" machines and it was based on this attitude of theirs that I would reply to them, true, and perhaps we'll know more in the future, but how widespread of a benefit has primate intelligence been to mankind. How many economies have been built on primate intelligence? Not that it makes a darned bit of difference, but rather than allow the ignorant denigration of dogs, this is yet another aspect to the argument I've had to learn to use, depending on who insists on perpetuating their beliefs with me.

 

IMO, between the scientific community and Judeo-Christian beliefs of the lowliness of anything but mankind (well let's face it. Those beliefs still exist within our species as well), we still have a long way to go in recognizing other beings as something to be respected in their own right, that maybe we don't exist in a hierarchy together, but that we travel on different planes in our time here on earth.

 

Sorry for rambling.

 

Vicki

[Deanna in OR]

Just a quick note--this has been a really interesting, stimulating and learning discussion for me! I have learned something from every single input and have had my own thinking stimulated.

 

thanks everyone!

Deanna in OR

[PennyT]

Guide dogs: Vis a vis guide dogs and the who-is-the-speaker (or who is to be

cooperated with) question, I

wondered earlier if a guide dog school, managed by someone whose scientific

curiosity was animated by a dark and

puckish sense of humor, ever decided to train half the people who were

receiving guide dogs with the dogs and the school's trainers before sending

the teams home and not to

train the other person/dog teams together first.

 

Who is smarter questions: The reason I think it's a good idea to say up

front that I don't regard even very smart dogs as as smart as children is

in part political and similar to what Vicki brought up about her opinions

being

discounted by her family because she is the family dog nut. I do not wish to

be mistaken

for a PETA nut or for someone who overestimates dog intelligence because

they interact with dogs so much, making me a meat-eating, most animal

experiment condoning dog nut. At the same time, I don't think

that discounting animal intelligence and emotions is limited to Judeo-Christian tradition at all though. If I were going to make

one of those 19th Century hierarchical rankings, I think I would give the

Judeo-Christian tradition fairly high marks.

 

Charles Darwin thought well of the intellect of dogs and other nonhuman

animals. Sixty or so years before studies of what animals

(including humans) can do became dominated by operant conditioning, he wrote THE DESCENT OF MAN. Here are some quotes from chapter three:

 

If no organic being

excepting man had possessed any mental power, or if his powers had been of

a wholly different nature from those of the lower animals, then we should

never have been able to convince ourselves that our high faculties had been

gradually developed. But it can be shewn that there is no fundamental

difference of this kind....

 

That which distinguishes man from the

lower animals is not the understanding of articulate sounds, for, as every

one knows, dogs understand many words and sentences. In this respect they

are at the same stage of development as infants, between the ages of ten

and twelve months, who understand many words and short sentences, but

cannot yet utter a single word. It is not the mere articulation which is

our distinguishing character, for parrots and other birds possess this

power. Nor is it the mere capacity of connecting definite sounds with

definite ideas; for it is certain that some parrots, which have been taught

to speak, connect unerringly words with things, and persons with events....

The lower animals differ from man solely in his almost infinitely

larger power of associating together the most diversified sounds and ideas;

and this obviously depends on the high development of his mental powers....

 

Several writers, more especially Prof. Max Muller,* have lately

insisted that the use of language implies the power of forming general

concepts; and that as no animals are supposed to possess this power,

an impassable barrier is formed between them and man.*(2) With respect

to animals, I have already endeavoured to shew that they have this

power, at least in a rude and incipient degree. As far as concerns

infants of from ten to eleven months old, and deaf-mutes, it seems

to me incredible, that they should be able to connect certain sounds

with certain general ideas as quickly as they do, unless such ideas

were already formed in their minds. The same remark may be extended to

the more intelligent animals; as Mr. Leslie Stephen observes,*(3) "A

dog frames a general concept of cats or sheep, and knows the

corresponding words as well as a philosopher. And the capacity to

understand is as good a proof of vocal intelligence, though in an

inferior degree, as the capacity to speak."

[PennyT]

Like Deanna, I want to add a last note because the thread seems to have played itself out. I, too, enjoyed and appreciated all the responses and, as Eileen noted, hope Pearse comes back out of lurk mode.

 

Talking dogs digression: My husband and I have often agreed how glad we are our dogs can't talk. We are fairly sure that they would say over and over, "Let's work sheep. Let's work sheep." When inside, "Let's go outside and work sheep," and then when finally the dogs maybe get the message that it's dark out there and the sheep are up for the night, "Ball. Ball. Ball."

 

Sometimes if my husband doesn't want to play ball, the dogs will offer other toys on the theory (we think) that if they choose just the right one, my husband will be lured into playing, and they are often right about this.

 

Imagine having to listen to the verbal canine repetition from multiple dogs of 200 ways to play. I don't want to hear it. It's not that I am convinced the dogs could not communicate other things. I have the feeling they just might not want to do so except to remind me when it's time for dinner, to be let out, or that being petted and sleeping on the bed are big preferences. I am content and relieved that those communications never take the form of spoken words.

 

Penny

[Tuck's BCBuddy]

I'll third that. If anything I had said caused or came across as something to run Pearse off, twas not my intention. If it did sound that way, however, my sincerest apologies.

[Pearse]

To all concerned about my state of lurkiness, let me assure you that no one said anything that offended me and even if you had, it wouldn't have made me go away. I just had said all I had to say and figured so had most other folks.

 

Pearse

[SoloRiver]

Just a couple more notes:

 

This discussion has turned into one about measuring intelligence, not language. The original study isn't looking at "intelligence" (which is almost impossible to define anyway in humans, much less compare across species). It is specifically looking at the nature of communication with the goal of understanding something about the evolution of human language. Bloom wasn't saying that dogs are dumber or lesser or anything of the sort. It is possible to say that we do not know if dogs understand words in the same way humans understand words without making a value judgment.

 

Scientists who are not familiar with dogs do indeed make some very facile assumptions sometimes when dealing with dogs. I have recently been reading up for a project looking at genetic variation in dogs and found some real howlers in terms of assumptions and study design. For example, one study used AKC registrations to estimate breed population size, which is probably inaccurate for most breeds and wildly inaccurate for the Border Collie (which was included in the study). Another found that Aussies formed a separate branch on a phylogenetic tree, which is a result that to me would suggest either (a) my study was seriously flawed and needed a do-over or ( that this intriguing result should be investigated further. Instead, the authors made up a just-so story about breed history (which most Aussie people know to be entirely bogus) to explain this anomaly and dropped it.

 

But I don't see any such mistakes with the Rico study or with Bloom's response. The question is whether dogs can understand words the way people do. The results of the study were interesting but at the same time not enough to refute the opposing hypothesis (dogs understand words but not the same way that people do). Penny asked what Bloom meant by meaning in his comment and she didn't like my answer, but that is how I read his response because that is the perspective of most research I am familiar with dealing with the evolution of human language. I do think dogs are different but I definitely don't think they are lesser. I'm an anthropologist and a cultural relativist and my relativity extends to dogs as well as humans. I like them just the way they are. They don't have to be little people in fur coats to be interesting to me.

[Bill Fosher]

Melanie wrote:

 

The question is whether dogs can understand words the way people do.

 

Which to me is like asking whether fish can ride a bicycle the way people do. Any answer other than "no" indicates a lack of understanding of the other species, and probably an underlying assumption that the way humans do it is right or best.

 

What I've been bumbling around trying to say -- and which Melanie starts to say -- is that understanding words is not the same as having language and that having language is not the same as being intelligent. Unfortunately, those distinctions have been blurred by the popular press accounts of the Rico study.

[PennyT]

Melanie: Penny asked what Bloom meant by meaning in his comment and she

didn't like my answer, but that is how I read his response because that is

the perspective of most research I am familiar with dealing with the

evolution of human language. I do think dogs are different but I definitely

don't think they are lesser. I'm an anthropologist and a cultural relativist

and my relativity extends to dogs as well as humans. I like them just the

way they are. They don't have to be little people in fur coats to be

interesting to me.

 

Penny: Hmm, Melanie, if you want to reopen the subject that is sure a good

way to do it.

 

People in little fur coats? My dogs? You want some kinda' bet on whose dogs

wear da' fur coats and whose dogs work da' sheep?

 

In fact, though, it wasn't a question of liking or not liking your answer. I

do not see requiring symbolic meaning as implicit in Bloom's article. If he

means symbolic in the sense you explained, he did not say or imply so.

Charles Darwin did not think that dogs were incapable or do not demonstrate

lucid understanding of the term "sheep" and neither do I. When I wrote about

how my own dogs understand the term "sheep," I tried to illustrate that it was in

different contexts with different syntax.

 

I am, however, tickled you reopened the thread because the subject has

hardly been touched by people who know and work border collies on livestock.

 

I also want to add that I value your opinions.

 

Okay, most of the time, I value your opinions.

 

Penny

[Denise Wall]

Well, since we're going to go on with this, here's something most people will not believe. Mick can identify sheep in pictures.

 

 

Even more unbelievable, he can identify cartoon looking colored drawings of sheep as sheep. I have a picture of this somewhere also, I just don't know where right now. This requires a great deal of abstraction for him to make this visual transition from a real sheep to a cartoon like representation of a sheep. My genetics friend Mellissa once showed the picture of him focused attentively on the cartoon sheep picture to a canine neurologist (or someone fancy like that) and he said I must've trained Mick to look at the picture. You'll have to take my word for it that he's actually hunting down anything that looks like a sheep to him. You can tell when he considers them sheep.

 

It makes you wonder how much we really do understand about them.

 

Edited to say I don't mean "hunting down" in a prey drive sense

 

Denise

[cgt]

Show off.

[SoloRiver]

Which to me is like asking whether fish can ride a bicycle the way people do. Any answer other than "no" indicates a lack of understanding of the other species, and probably an underlying assumption that the way humans do it is right or best.

 

Yeah, but see, this is the thing. The reason the researchers are looking at word understanding in dogs is to learn something about humans, not to learn something about dogs. It's not that the way humans do it is right or best -- I don't see a value judgment expressed in the study or the comment (although that doesn't mean the researchers don't have one) -- but that the way humans do it is the way we are interested in researching. It's kind of like if you were studying the evolution of blue butterflies -- it doesn't mean that the red ones aren't as good, just that you happen to be looking at the blue ones at this time.

 

The reason the entire field of primatology exists is that folks who were interested in human evolution hoped that studying non-human primates would tell us something about where we came from (because traits shared in common between related species tend to be primitive, or ones that existed in a common ancestor). The point was to learn illuminate facts about humans, not to understand chimps or gibbons or vervets in their own right. The field has grown and changed (and taken on an important conservation component) so this isn't as true anymore. But, Bloom alludes to this traditional role when he says that dogs may be the new chimps.

 

You and I and Penny and everyone else involved in this discussion are interested in dogs and what makes them tick. The researchers are interested in uncovering primitive traits that we and dogs may have inherited from a common ancestor, and therefore understanding something about our own evolutionary trajectory. If we share certain language-related traits with dogs, it means that these foundation characters evolved a very, very long time ago and are not a derived character limited to humans and our extinct relatives. Since we and dogs diverged much longer ago than we and chimps did, the dog studies give us a lot more time depth than chimp language studies do. The reason all this is interesting is that we are always interested in reconstructing the biology and behaviors of our ancestors, and figuring out when "uniquely human" characters evolved, to a much greater degree than when we are studying other kinds of animals.

 

So, the reason the study may not be satisfying to dog people is that the point of the study ultimately isn't to understand dogs.

 

Penny, I didn't mean to imply that you personally consider your dogs to be humans in fur coats. I just didn't see any value judgment anywhere in either Bloom's comment or the original article and I don't think there's anything wrong or unkind about saying that dogs are different from humans. We agree that their differences are some of the coolest things about them.

 

I think that Border Collies understand human communications, especially when working stock, on a very subtle and complex level, but I don't think it is necessary to invoke human-style language capabilities to explain their understanding, and I don't think this makes them something lesser than what I thought they were before I started thinking about this stuff. Bloom did not specifically mention symbolism in his comment but his "sock" discussion touched on some of these distinctions that appear to be unique to human language and I read his comment within that context. You asked for people in the sciences to explain what he was talking about and I might be mistaken, but having studied anthropological linguistics, this is how I read him and so I gave this particular interpretation. I suppose you could always ask him what exactly he meant.

[Denise Wall]

Charles wrote:

 

Show off

 

Actually, it gets to be quite annoying. He also watches TV and while it's funny at first, it gets old real fast having him moving around in front of the TV all the time.

[cgt]

I'll bet he can find his sheep though, eh?

[Denise Wall]

Here's the sheep drawing I was talking about:

 

 

 

 

Maybe this is not strange??

 

Denise

 

PS I know it's an ugly picture. I won it in a raffle.

[Denise Wall]

I'll bet he can find his sheep though, eh?

 

I must admit he's pretty good at that. Is it worth the other hassle 99.999% of the rest of the time -- I'm not sure

 

Denise

[Rebecca, Irena Farm]

Actually, I love the picture - you're welcome to drop it by my house anytime.

 

It would go beautifully in my new kitchen, if the kitchen ever gets done.

 

That's a Texel, isn't it?