Lady Lovelace's Objection By Alan Turing Summary
Alan Turing’s paper “Computing machinery and intelligence” in 1950 is one of the foundational documents from the early days of Cognitive Science. After its publication, critics proposing objections have never stopped. One of the most reasonable objections similar to so called “Lady Lovelace’s Objection” argues that a machine can never do anything really new, which is to say no machine has the capacity for original thought. In this way, it follows that no computer can have human intelligence. From my perspective, this objection is strong enough and can threaten Turing’s idea about intelligent computers. In addition, Turing’s response can only overcome a part of this objection and is still not robust enough.
I would like to begin the demonstration with summarizing what Turing argues. To answer the question “Can machines think?”, Turing comes up with the “Imitation game”. Three participates are involved in such a test, a human being, a computer and an interrogator. They are locked in separate rooms and …show more content…
can only send messages back and forth by teletype. After a short communication, the interrogator will be asked to determine where the human being sits and the other way around based only on their answers to the interrogator’s questions. If the interrogator cannot tell the difference between the human being and the computer, according to 2a3 of Mind, it is reasonable to say that “there is no interesting and importance feature that distinguishes human from computer” when having the conversation. In addition, since, as discussed in 2a4 of Mind, “… conversation is an activity that normally allows us to gauge the intelligence and thinking capacity of other people …”, it is safe to say the computer can think.
The critics who support this variant of “Lady Lovelace’s Objection” argue that although the answers by the computer fool the interrogator, a computer cannot be considered to be a thinker before it has the capacity to create something new. It is a really robust objection at that time because machines seem really basic and certainly cannot do “original work”.
For this objection, Turing mentioned “The Analytical Engine was a universal digital computer, so that, if its storage capacity and speed were adequate, it could by suitable programming be made to mimic the machine in question.” (Turing, 14). This, from my point of view, is a strong argument against the statement “a machine can ‘never do anything really new’”. Computer program nowadays can actually do something fairly “new”. Machine learning algorithms, for example, can derive a statistic model from a prepared data set by itself. Before the result coming out, even the programmer does not know everything about the model. So if the supporters of the variant of “Lady Lovelace’s Objection” talk about “unprecedented”, their objection is simply false.
By contrast, Turing also questions “How do we know that people ever do anything original?” (Turing, 15).
He claims that human’s original work may be also the result of well-known rules and principles. So people themselves may not create something new. This seems to be a good point at the first glance. However, it is not hard to figure out that his statement is quite vague. By common sense, people classify the things that can hardly think of as new things. If human’s creativity is indeed coming from common principles without any “wisdom”, everyone has the knowledge can come up with the same “original idea” which, in that case, is not original at all. Going over the history of invention, it is not hard to tell it is hard to do “original work” even with all the “well-known principles”. In fact, even Turing himself cannot give an example of how a list of well-known principles leads to something “original”. As a result, I think Turing’s response is still not robust
enough.
Turing’s response to another variant of the objection makes perfect sense. If by original the supporters actually mean surprising or unpredictable, the objection is flawed. This, however, still cannot hide Turing’s failure of overcoming the first variant of the “Lady Lovelace’s Objection”.
To fix Turing’s response, it is important to show that creativity is a tough thing to do even for the human being. In the history, only few people are known to be creative. So it should not get added into the criteria on judging whether or not a creature can think.
In summary, the “cannot create Objection” threats Turing’s idea of machine intelligence and Turing’s response is not robust enough to overcomes it. However, it is not hard to find the flaw of the objection and by extending on Turing’s response, it is not reasonable to say that no machine can think because no machine can do something really new.
I would like to begin the demonstration with summarizing what Turing argues. To answer the question “Can machines think?”, Turing comes up with the “Imitation game”. Three participates are involved in such a test, a human being, a computer and an interrogator. They are locked in separate rooms and …show more content…
can only send messages back and forth by teletype. After a short communication, the interrogator will be asked to determine where the human being sits and the other way around based only on their answers to the interrogator’s questions. If the interrogator cannot tell the difference between the human being and the computer, according to 2a3 of Mind, it is reasonable to say that “there is no interesting and importance feature that distinguishes human from computer” when having the conversation. In addition, since, as discussed in 2a4 of Mind, “… conversation is an activity that normally allows us to gauge the intelligence and thinking capacity of other people …”, it is safe to say the computer can think.
The critics who support this variant of “Lady Lovelace’s Objection” argue that although the answers by the computer fool the interrogator, a computer cannot be considered to be a thinker before it has the capacity to create something new. It is a really robust objection at that time because machines seem really basic and certainly cannot do “original work”.
For this objection, Turing mentioned “The Analytical Engine was a universal digital computer, so that, if its storage capacity and speed were adequate, it could by suitable programming be made to mimic the machine in question.” (Turing, 14). This, from my point of view, is a strong argument against the statement “a machine can ‘never do anything really new’”. Computer program nowadays can actually do something fairly “new”. Machine learning algorithms, for example, can derive a statistic model from a prepared data set by itself. Before the result coming out, even the programmer does not know everything about the model. So if the supporters of the variant of “Lady Lovelace’s Objection” talk about “unprecedented”, their objection is simply false.
By contrast, Turing also questions “How do we know that people ever do anything original?” (Turing, 15).
He claims that human’s original work may be also the result of well-known rules and principles. So people themselves may not create something new. This seems to be a good point at the first glance. However, it is not hard to figure out that his statement is quite vague. By common sense, people classify the things that can hardly think of as new things. If human’s creativity is indeed coming from common principles without any “wisdom”, everyone has the knowledge can come up with the same “original idea” which, in that case, is not original at all. Going over the history of invention, it is not hard to tell it is hard to do “original work” even with all the “well-known principles”. In fact, even Turing himself cannot give an example of how a list of well-known principles leads to something “original”. As a result, I think Turing’s response is still not robust
enough.
Turing’s response to another variant of the objection makes perfect sense. If by original the supporters actually mean surprising or unpredictable, the objection is flawed. This, however, still cannot hide Turing’s failure of overcoming the first variant of the “Lady Lovelace’s Objection”.
To fix Turing’s response, it is important to show that creativity is a tough thing to do even for the human being. In the history, only few people are known to be creative. So it should not get added into the criteria on judging whether or not a creature can think.
In summary, the “cannot create Objection” threats Turing’s idea of machine intelligence and Turing’s response is not robust enough to overcomes it. However, it is not hard to find the flaw of the objection and by extending on Turing’s response, it is not reasonable to say that no machine can think because no machine can do something really new.