|Experiment: artificial bringing about of an event or artificial creation of a state for testing a hypothesis. Experiments can lead to the reformulation of the initial hypotheses and the reformulation of theories. See also theories, measuring, science, hypotheses, Bayesianism, confirmation, events, paradigm change, reference systems.|
Books on Amazon
|Experiments||John Dupré Gespräche mit Affen Reflexionen über die wissenschaftliche Erforschung der Sprache in D. Perler/M. Wild (Hg.) Der Geist der Tiere, Frankfurt 2005
Perler I 298
Experiment/Research/Language/Animal/Interpretation/Dupré: it is disputed whether the monkeys
A) are only simply conditioned a la skinner, so that they produce desired results. That is, the monkey does essentially the same as the rat which is pushing a lever. ((s): This could just as well be interpreted as symbol use?).
That is, it is claimed that the monkey does not know that a character (used by him) means e.g. "Banana". Then the sign "Give me a banana" would be the same as shaking the tree, so that some fall down.
B) Actual use of symbols. This does not have to mean yet that the monkeys have all the abilities that we have in this context.
E.g. monkeys can learn to produce utterances that are suitable for different foods. But they do not seem to be able to answer when they are to sort out these objects!
A similar separation of production and reception has been established in children.
However, this is just the exception, most monkeys can very well generalize.
For example, a gorilla from Patterson can supposedly report from the past and utter threats as well as insults: e.g. death: "Pleasant, Heia, Nest" "When gorillas die?" "Problem old".
Dupré: these are certainly more than conditioned reflexes.
A) sign language: Gardners, Terrace, Patterson, Fouts: Level of communication as high as possible.
Communicative intention exceeds "literal" interpretation.
B) Artificial Languages: Premack, Rumbaugh: Possibility to gain clean data has priority.
Possible behavior is narrowly limited. Therefore there are only a few characters.
Dupré: this dichotomy is actually a caricature: for the representatives of b), on the one hand, criticize the affective connections, and grant particular importance to spontaneous utterances.
On the other hand, the representatives of a) are strict with the tests.
Double blind test: even the observer does not know the correct answer.
Dupré: but also here the desired answer can be suggested, if the blind observer wants to figure out what the animal "means".
Method/Ockham's razor/animal/experiment/Dupré: commitment to economical explanation: curtails the allegations of the monkey researchers allegedly in two directions:
1. if there is a channel through which the observer might have given the monkey a correct answer,...
...then we assume that the assumption that exactly this happened is more economical than the one that the monkey has shown an ability.
2. Ockham's razor is mostly directed against the assumption of creative linguistic usage in monkeys. Adoption of an error is more economical.
DupréVsOckham: Economy is hardly an objective, theory-independent concept. Why is it more economical to adopt a complex and hidden communication channel than to assume that the monkey knows what it is doing?
"Kluger Hans"/Dupré: also shows that there is a high degree of communication between humans and animals. (s) This can also be called "sign use", if the animal obviously understands it correctly.
Typically occurs in activities where there are correct or incorrect answers.
N.B.: these are, ironically, precisely the experiments that are supposed to provide unique data.
But if the monkey produces new, unexpected utterances, then the criticism cannot work.
Ironically, it is precisely the stereotypical reactions that are suspected of being manipulated.
Method/primate research/Dupré: perhaps these standards of animal research are just not appropriate.
D. Perler/M. Wild (Hg)
Der Geist der Tiere Frankfurt 2005