Post by KoNeko on Nov 6, 2004 21:11:50 GMT -5
Memory as a web
As mentioned above, there is some sort of psychological connectedness between mental states. Part of the reason mental states have such connectedness is because there is some sort of causal sequence in which the states arise, and knowledge of the causal sequence is relevant in order to understand the context of the experience. For example, suppose two friends are having a conversation on the subjects of A, B and C (in that particular order). The conversation about A is what gives rise to the conversation about B and so on. So the experience of talking about subject B only arises in the context of having talked about A, and accordingly, the content of a memory of talking about B (including its causal history) counterfactually depends on sort of memory of having talked about A.15 So suppose the experience of having part of that conversation (say, the part when subject C is about to be discussed) is transplanted into another person. Even if that person knew about the causal origin of ¡§we are about to discuss subject C¡¨, it would still be pretty strange to have such a mental state. The relevance of ¡§we are about to discuss subject C¡¨ can only be understood in light of having talked about A and B, and that C leads on from B, and so on. So the two friends who converse about A, B and C are in a position to fully understand what is being said and why, because the sentence ¡§we are about to discuss C¡¨ occurs at a specific time in the sequence of sentences making up the conversation, and the person who had the transplant would not be able to access the full conversation.
I return to the point I made earlier about mental states being with contextual webs here. In order for someone to fully understand the context of an implanted Q-memory, that person must understand even the very basic concepts embedded in the experience that gave rise to the memory. For example, if I have the Q-memory of walking your children to school, I must understand what a school is, and that children go to school, and that the small people walking in front of me are children, and so on. This may be rather obvious. However, the contextual background of any such mental state would be richly diverse and far reaching, which would be relevant when we consider the content of our experiences when we Q-remember something.
Here is a (more extreme) example illustrating the vastness of the contextual web required for memory and other mental states. I am typing this essay on my computer. This gives rise to my memory of typing the essay. I know that essays are a form of assessment at my university, and I have the desire to do well in writing this essay. Perhaps I even have the mental image of receiving a high mark in this essay from my assessors. Now, there is a set of mental states in my mind which are all interrelated. Suppose this entire set of mental states is transplanted into a tribesman from a remote part of Africa who has never been in contact with western civilisation. (Alternatively, we could always transplant the mental states into an alien, but I do not think that this is necessary.)
What would the tribesman experience? He has never seen a computer or a university before. The Q-memory he has of sitting at a desk, typing an essay on a computer and having mental images of receiving a high mark from people he has never seen (and for a subject that he has never studied) would be impossible for him to even begin to understand. Even if he was somehow informed of the mental states¡¦ causal origin, it is doubtful that he would still be able to comprehend what he Q-remembers. We would be inclined to say that does not have the requisite contextual web to buttress the Q-memory.
So, if at all, how could a Q-memory be fully coherent? Given the different degrees of contextual comprehension in the above examples, I would argue that either it is not comprehensible at all, or it is only comprehensible if the Q-memory is transplanted along with every other memory, belief, desire, thought etc. that the original person who had that experience had (or, in other words, a complete brain transplant), and all of my own memories, beliefs, desires etc. were erased to avoid any conflicts (as in the example of walking your children to school). Any Q-memory case that does not fall into one extreme or the other could be cast as one in which the person who Q-remembers does not have the requisite contextual web for him or her to have the same experience as the person who had the original experience, and on that ground it would contradict conditions (1) and (2) in the requirements for Q-memory.
The web of other mental states
What I have shown so far is that there are memories that cannot be coherently made into Q-memories. Parfit and Shoemaker¡¦s account of Q-memory does not accommodate for these abnormal cases involving brain transplants, implanted memories, etc. It may meet the memory requirement in ordinary cases because it solves Locke¡¦s circularity problem, but it is clearly not sufficient in all cases. In extreme cases, a "memory transplant" under Parfit¡¦s account does not provide the requisite contextual background to the transplanted Q-memory, resulting in sometimes bizarre and unintelligible outcomes in the subject. Thus, it seems utterly convincing that we need some sort of contextual web connecting a Q-memory to other memories, beliefs, desires and thoughts in order for a Q-memory to fully make sense.
I admit that this requirement of a contextual web also extends to ordinary memories as well.16 Every memory (Q-memory or not) exists in at least one context ¡V the history of the rememberer. Memories (and other mental states for that matter) do not exist successively in a vacuum. Part of the memory¡¦s content is the relation that it has to other memories in the network, including any subsequent memories. For example, the teenager whose parents are going through a bitter and painful divorce may recall family holidays in his childhood as exercises in forced civility, even if he was not aware of it at the time. However, the need to examine the contextual background is more apparent in transplanted Q-memories than ordinary memories, so we may not have turned our minds to it previously. Even so, I argue that memories of all sorts require some type of link to other memories (and other mental states) in order to give them adequate context.
As mentioned above, there is some sort of psychological connectedness between mental states. Part of the reason mental states have such connectedness is because there is some sort of causal sequence in which the states arise, and knowledge of the causal sequence is relevant in order to understand the context of the experience. For example, suppose two friends are having a conversation on the subjects of A, B and C (in that particular order). The conversation about A is what gives rise to the conversation about B and so on. So the experience of talking about subject B only arises in the context of having talked about A, and accordingly, the content of a memory of talking about B (including its causal history) counterfactually depends on sort of memory of having talked about A.15 So suppose the experience of having part of that conversation (say, the part when subject C is about to be discussed) is transplanted into another person. Even if that person knew about the causal origin of ¡§we are about to discuss subject C¡¨, it would still be pretty strange to have such a mental state. The relevance of ¡§we are about to discuss subject C¡¨ can only be understood in light of having talked about A and B, and that C leads on from B, and so on. So the two friends who converse about A, B and C are in a position to fully understand what is being said and why, because the sentence ¡§we are about to discuss C¡¨ occurs at a specific time in the sequence of sentences making up the conversation, and the person who had the transplant would not be able to access the full conversation.
I return to the point I made earlier about mental states being with contextual webs here. In order for someone to fully understand the context of an implanted Q-memory, that person must understand even the very basic concepts embedded in the experience that gave rise to the memory. For example, if I have the Q-memory of walking your children to school, I must understand what a school is, and that children go to school, and that the small people walking in front of me are children, and so on. This may be rather obvious. However, the contextual background of any such mental state would be richly diverse and far reaching, which would be relevant when we consider the content of our experiences when we Q-remember something.
Here is a (more extreme) example illustrating the vastness of the contextual web required for memory and other mental states. I am typing this essay on my computer. This gives rise to my memory of typing the essay. I know that essays are a form of assessment at my university, and I have the desire to do well in writing this essay. Perhaps I even have the mental image of receiving a high mark in this essay from my assessors. Now, there is a set of mental states in my mind which are all interrelated. Suppose this entire set of mental states is transplanted into a tribesman from a remote part of Africa who has never been in contact with western civilisation. (Alternatively, we could always transplant the mental states into an alien, but I do not think that this is necessary.)
What would the tribesman experience? He has never seen a computer or a university before. The Q-memory he has of sitting at a desk, typing an essay on a computer and having mental images of receiving a high mark from people he has never seen (and for a subject that he has never studied) would be impossible for him to even begin to understand. Even if he was somehow informed of the mental states¡¦ causal origin, it is doubtful that he would still be able to comprehend what he Q-remembers. We would be inclined to say that does not have the requisite contextual web to buttress the Q-memory.
So, if at all, how could a Q-memory be fully coherent? Given the different degrees of contextual comprehension in the above examples, I would argue that either it is not comprehensible at all, or it is only comprehensible if the Q-memory is transplanted along with every other memory, belief, desire, thought etc. that the original person who had that experience had (or, in other words, a complete brain transplant), and all of my own memories, beliefs, desires etc. were erased to avoid any conflicts (as in the example of walking your children to school). Any Q-memory case that does not fall into one extreme or the other could be cast as one in which the person who Q-remembers does not have the requisite contextual web for him or her to have the same experience as the person who had the original experience, and on that ground it would contradict conditions (1) and (2) in the requirements for Q-memory.
The web of other mental states
What I have shown so far is that there are memories that cannot be coherently made into Q-memories. Parfit and Shoemaker¡¦s account of Q-memory does not accommodate for these abnormal cases involving brain transplants, implanted memories, etc. It may meet the memory requirement in ordinary cases because it solves Locke¡¦s circularity problem, but it is clearly not sufficient in all cases. In extreme cases, a "memory transplant" under Parfit¡¦s account does not provide the requisite contextual background to the transplanted Q-memory, resulting in sometimes bizarre and unintelligible outcomes in the subject. Thus, it seems utterly convincing that we need some sort of contextual web connecting a Q-memory to other memories, beliefs, desires and thoughts in order for a Q-memory to fully make sense.
I admit that this requirement of a contextual web also extends to ordinary memories as well.16 Every memory (Q-memory or not) exists in at least one context ¡V the history of the rememberer. Memories (and other mental states for that matter) do not exist successively in a vacuum. Part of the memory¡¦s content is the relation that it has to other memories in the network, including any subsequent memories. For example, the teenager whose parents are going through a bitter and painful divorce may recall family holidays in his childhood as exercises in forced civility, even if he was not aware of it at the time. However, the need to examine the contextual background is more apparent in transplanted Q-memories than ordinary memories, so we may not have turned our minds to it previously. Even so, I argue that memories of all sorts require some type of link to other memories (and other mental states) in order to give them adequate context.