Or;How do Things remember ?
Stanislaw Lem,the great Polish Science Fiction writer,has written a highly readable book on his childhood,which is called "The High Castle" or in Polish "Wysoki Zamek" - a book full of the most remarkable observations and unusual experiences.One passage which has caught my attention ,goes something like this in my English rendering;
"I believed,without confiding the secret to anybody,that dead objects like man might suffer from shortcomings and inadequacies,and even that they might be forgetful.If one had enough patience ,one might be able to force them to multiply by taking them by surprise.Thus,if one assumed that a pocket-knife ,which hade been left in a drawer ,did forget where it was supposed to be ,one might later find it in a completely different place,e.g.among the books in a shelf.As it would lack possibilities to get out of the drawer,in this impossible situation it would simply duplicate itself and become two.Thus in my view ,the things were suubject to a certain logic,they had to obey some rules,and for someone who knew them well,it might be possible to force the dead matter to some desirable results."
What Lem describes in a negative way,is what it means to live in a well ordered world. Of course things as we know them in everyday lifte are not like Lem expected.They are flawless.They do not forget and we can never force them to multiply by surprise.The things are so clever;they remember where they are,even when we have forgotten where we put them. How do the things remember ?
And is not this flawless memory of things an important part
of what we mean when we say that a certain segment of our experience was "real" as opposed to "unreal" ?
Two objections of course immediately present themselves to this way of reasoning.
The first is that this form of order is a local phenomenon.The physical world contains much more than "things" and it is of course
unsatisfactory to limit the realm of reality to that relatively limited sector of the physical world or what we take to be the physical world which contains "things" organised in a wellordered pattern of time and space coordinates.The realm of things extends from somewhere close to our body and outwards to a limit which might be quite difficult to define exactly.The biscuit which I take with my afternoon coffee definitely stops to be a thing when it disappears into my mouth and not unlike a musical tone dissolves into a,hopefully pleasant sensation.At the other end there are e.g. galaxies.It seems very weird to say that galaxies are things as the different stars and clouds of heated matter which make them up are not all present at the same time.
In a plasma there is hardly any well memorized order of the sort we expect from a campus building or from the drawers where we keep our underwear.On the subatomic level memory really seems to start to fail in different ways depending on what interpretation we want to give to subatomic phenomena..So the flawless memory of
things cannot reasonably be extended to the material world in its entirety. The sort of order which can be expressed by saying that the things have such a good memory , rules,but not everywhere and on all levels.But it seems to actually rule wherever we are.
So,to identify this sort of memory with reality is obviously to define "reality" in a more narrow way than is ordinarily done. We should not be surprised.It has been shown by many philosophers in many contexts that "unreal" can mean many different things.
I leave it there,for the moment.
The second objection is that things can hardly have any memory at all,because things do not have any minds. Not everybody might agree on that point,but I shall assume that such things as we normally call "dead" do not have any minds. And memory is normally thought of as something which can only exist in the context of a mind.
When I say that the drawer is so clever,because it is able to remember exactly what I put into it,this could pass as a metaphorical way of expression.When I say that the dollar is falling in its relation to Deutsche Mark I do of course not mean it literally,and when I say that the book is very intelligent I do not expect to be misunderstood as an animist either.We actually ascribe qualities to things all the time which are really mental qualities in some way or another related to their internal representations.
Of course,e.g.in Henri Bergson´s philosophy the world itself tends to take on mind-like characteristics,because he introduces a concept "pure memory" or "mémoire pure" by making a rather difficult distinction between what could become the meaning of a memory image ,the raw material of memory as if it were,and the memory as represented in the memory image in an act of actual recollection. A less abstract name for memoire pure might be potential memory.
Potential memory cannot recall itself but it is the sort of subjectmatter which can be recalled.Of course everything does not lend itself with the same ease to become represented in memory.A certain degree of order is a prerequisite.
An example which might help to clarify this point is the sort of organisation which antique writers recommend in the so called
Ars commemorativa .The latin writers of rhetoric manuals e.g. the
anonymous author of the book Ad Herennium (falsely ascribed to Cicero) recommend us to visualize and memorize a building,with archways,colonnades,stairs and big rooms organized in a definite way.Statues and herms help us to identify the places between the different pillars.Everything has to be clearly lighted and to have such an architecture that things do not stand in the way of each other.This is called the locus .In the locus we then place the different representations of what we want to remember.They are called imagi. If ,e.g. we are preparing a political speech where we want to speak,first of naval defense and then of territorial defense,we might place an anchor in one archway and a sword in the next one.
The trick of artifical memory is to map topologically a memory representation on top of another,where the first order representation supports the second order one.It seems to have worked fairly well for the antique rhetors who used it.With natural memory the difference is that the representation is supposed to map something which is not a representation itself.
Parts of the world we know easily lend themself to be represented in memory. Things are connected in particular ways with each other.These connections are not always spatio-temporal,which obviously raises the need for a more generic concept than connected in space and time.For this more generic concept I shall from now on use the word "connectedness".
The following might serve as an illustration:I sometimes teach in a room on the first floor of the Hall.To get from there to my office ,which is on the fourth floor, I have a choice between a few alternative paths,of which some include an elevator,others not.
Whatever path I chose I have to start in a (particular) corridor on the first floor and to end in a (particular) corridor on the fourth floor. Of course we take topological order in the surrounding space for granted.This is always the same.So the lecture hall remembers where it is and the office remembers its place in the order.Furthermore,there is a reversibility.If I go back I expect the rooms to appear in an order which is the mirror image of the first one.
These are topological properties of experienced space which we take for granted .These are the properties of the antique memory palace,the loci on which artificial memory was built.
So far so good.What I have said up to this point does not seem very controversial to me. Let us assume now,,that,having at last arrived in my office on the fourth floor,passing all the rooms of one of the ordinary paths which lead there ,I fall asleep in my office.And in a dream I enter another room which is thus a room of my dream.It might be a Greek Temple Hall,or may be an agora.As we need some name for that room ,let´s call it "the last room".
The last room clearly has different properties than the other ones.It does not relate spatially.It has no place of its own in the topological order of the rest of the rooms.It does not fit topologically.It does not make any sense to say that it is included in the next last room either.There are many things represented in different ways in my office - by pictures and books - which are not spatially included even if the physical vehicles of the representations are included in it.It does not belong to a path in the sense that there is a determinate sequency of passages which leads to it.Like the prompter on the screen of my computer,when it passes from the extreme right margin to the left my attention seemed to get there without any mediation.(It is of course possible to construct an abstract topology which can take care of the prompter,e.g. by representing the screen as the two-dimensional projection of a cylinder;for the case with the last room I am less certain.) I am not certain that the fact that the last room is a room in a dream is relevant either.It is of course a "mental image" in the sense that in some way its vehicle of representation is in my brain.But would we have had a much different problem if the last room had been produced by means of cyberspace and virtual reality ? Clearly in that case too we would have to make a distinction between the room represented and the means of representation.And furthermore,"real"rooms are not less unreal in the sense that they are of course represented to us as well. In the standard case of being in my office I am in the room and the room is in me,in the sense that the vehicle of representation of the office is in my brain and that it represents me as being in the room.
I do not know,for certain,either whether it is right to call the last room the last one."Last" suggests a place in a temporal order but is it the room or its mental representation which takes place in an order ?
So the last room is connected to my total experience but not in quite the same way as the other rooms in the actual sequence.There are differences.One is that it does not have the same excellent memory as the other rooms.It does not recur with the same willingness.If I leave it,I am not certain that I will ever find my way back to it.
So the last room is connected but obviously connected in such a way that it does not recur in the same pattern as the other rooms.It does not,figuratively spoken,have the same excellent memory.
What sort of connectedness is ,then, this one ?
Lacking other glue lets try then,with causal connection !
The problem is of course,that I do not have more reason to believe that the office,or the sequence of rooms which led to the last one, have more to do with the last one,than a suddenly occuring
tooth-ache has to do with the room where I got it.
Temporal and causal connectedness without spatial connectedness seem weird to us.Causes are events . Effects are events.
Causes are supposed always to have precede effects.We often reason as if external events could cause dreams and dreams can cause events.But this is of course a metaphorical way of speaking,like when we say that an antique stone speaks about the battle of Thermopyle.Of course no stone can speak.There is not even any text which can speak.Somebody performs a speech-act by means of an (inscribed) stone and we emulate it through its medium until it reaches us.
By Analogy; to dream is an event,which takes place in time,also in space,I dream in a particular hotel room etc. but the dream room or dream locality are not in space .Can something occupy a place in time without occupying a place in space ? It seems weird,but if we do not want to assign to the dream room a place in time which is not a place in space it seems impossible to claim that it is the sort of entity which can causally connected to the rest of our experience.
So,even if it makes perfect sense to say that I dreamed a certain dream on Monday last week and a second dream two days later,if I may have my way,this does not entitle us to say that the contents of the two dreams display a temporal order.They are simply not related in time,and thus not causally related either.I am aware that this conclusion might cause trouble for some types of psycho-analytical theory of the more mechanistic type,but I am willing to take the consequence.A dream is related to the rest of our life not in the way an effect is related to its cause,but rather in the way an interpretation of a text is related to the text .
So what sort of connection is the connection between the last room and the rooms that led up to it ?
One possible answer is of course to say that there is no connection at all.But that does not make any sense.At least no common sense.My dream is mine.It is not yours .It is not nobody`s.In some way it belongs to a complex which we might provisionally call my life experience and my life experience is not yours.And observe,that in order to use the possessive pronomen here,we do not have to dogmatically announce that there is such a thing as a central substance which is me and which is the base of the use of the possessive .Films do not have ego-substances and still it makes sense to say that one sequence belongs to one film and not to another.The last room belongs to this life experience and not to another.
After all I arrived there.
The standard answer to the question what makes a dream my dream is of course that it is connected to me as a subject in a unique way. Only I can dream my dreams.But is not this a rather unfounded assumption ? Why should I not be able to dream somebody else´s dream ? Is there really any intrinsic quality of my dreams that make them mine ? (And by intrinsic quality I think I mean what G.E.Moore would have meant with an intrinsic quality;one which could not be detached from it without making it qualitatively different.) Honestly,I do not know.
But if we deny that my dreams have an intrinsic quality which makes them mine,their uniqueness must consist in the unique way they are connected to the rest of a field of experiences.And thus,to try to define the uniqueness that way is to beg the question as we do not know that it is connected but not in what the connectedness consists.
At this point it might make sense to say a few words about immaterialism.An important point in immaterialism is that it deprives us of the possibility to sort objects,such as towers; into two distinct categories such which exist outside us as real towers and inside us, represented as being outside us,and such which only exist inside us,represented as being outside us but not really being outside us.For immaterialists everything which exists exists as representations.If we only look at representations it is hard to tell the difference.
So the connection between the representation and other representations becomes the only thing we have to go after. It is very important indeed to Berkeley´s immaterialism that there is no logical tie between the tactile representation of roundness and the visual representation of roundness. Because if there were ,we could compare as if it were from a point outside the world.(This might be what Nicolas Malebranche is out to find in his celebrated seeing in God. )
In the case of dreams it seems to us as if we might be able to compare,because there is the awakening.In the case of virtual reality there might be such criteria as the fact that a certain pattern,after long experience,say the reflections of sunlight in a lake , seem to repeat themselves,or something like that.
Here,actually the illusory shows a better memory than the real (we might think of a sequence of virtual reality which at the end
reveals itself by starting to repeat itself) which shows that repeatability is a fairly shaky test of reality. We see it as the test of the reality of a room that we can reach it again,following one or more established paths.But if we arrive at the same place,whatever path we chose to take, we are not content at all. A labyrint where all roads lead to the center makes an unreal impression.(Jerome K.Jerome made a grandiose joke of it in the passage about the Hampton Court garden labyrint in "Three Men in a Boat") So what we tend to regard as a test of reality is not organisation and connectedness in abstracto but some particular types of organised connections between experiences.They have to be organised in such a way that they give something like a decision procedure.Different operations,like taking one path out of many possible ,has to give a different outcome than taking another one.(It might be the case that future developements in information technology,virtual experiences in a cyberspace might weaken or change our entire system of sensitivities on this point.Already air travel ,as practised in our days,gives us a strong feeling,but alas,illusory of being able to move from a place to another without passing the places in between.)
What would it mean to say that the world as a whole is illusory ?
Would it imply immaterialism ?
Or maybe something entirely different ?
In order to maintain that the world as a whole is illusory,do we need something less illusory to compare with ? There seems to be a good point in Gilbert Ryle´s observation  that in a country without coinage there can be no counterfeits.
If by "reality" we mean a set of experiences which are orderly connected in the sense described,i.e. that they form a suitable subject matter for memory, obviously our total experience,the "world we found" - to use Wittgenstein´s handy expression - contains real and unreal elements.Of course dreams ,experiences of virtual reality,hallucinations etc. - in other words all sorts of virtual experiences also belong to the world we find.But they are obviously differently connected.
As the dream argument of Descartes seems to show there are local decision procedures but hardly any global one.
What would it mean then to claim that orderly connections are only local phenomena which do not extend over the whole context of experience ? That we cannot trust our senses ? But certainly we trust our senses! Even the most sceptical philosopher turns into a dogmatist when he hits road traffic after hours. He does not expect some of the meeting vehicles to be virtual and others non-virtual.
The local properties of the system do not cause us much trouble.It is when we try to imagine it as indefinitely extended that we run into philosophical problems.
A general problem of orientation,or navigation,if you prefer that,might serve as an illustration.
Assume that we are travelling in a universe which we ,for the sake of simplicity assume to be two dimensional and to be populated by only three sorts of objects ,hyphens,dots and the blanks between them.Like in the Turing-machine then movement is only possible forward or backward.We assume that the hyphens and dots and blanks are randomly distributed but form some recognizable constellations.Travelling from a place in such a system to another another particular place the traveller has to identify the goal by the constellation in which it is inbedded.(He has of course,first to decide whether to go forward or backward and then where to stop.) To identify the area where he wants to go,he can always study it in a more extended context.
If we assume,now,that the universe in which the travel takes place is transfinite ,there is of course no ultimate method to identify the goal area as the unique destination .In other words - in such a system unique orientation is only locally possible.
It is fairly easy to see the relevance of this model to the Dream argument.If the decision of the status - virtual or nonvirtual - of a string of experiences can always be made dependent of a more extended strand of possible experiences the distinction can only be upheld locally.
In his "Reflexions on Science",Diderot has a famous passage where he describes the extension of Science as a vast landscape scattered with places which are in the darkness and other which are in the light. Somebody might want to trade this metafore with the situation in a Swiss cheese.
How are the holes connected to the solid cheese ? One possible answer is of course that they are not connected at all.The holes are also the cheese,but they are not connected in some special way to the rest.
 If somebody wants an object of comparison she might think of films.How are two films ,say "The Dictator" by Chaplin and "Gone with the Wind" temporally related to each other? Of course there are different times for their production.But there is no well-ordered system for their reappearences.I am willing to claim that this absence of order - which is not the same as disorder - is caracteristic of a great deal of what is normally carried to our "inner experience"
 The decision procedure for the dream is,as Descartes observes,
the awakening.And the strength of the dream argument is in the possibility that the awakening can be thought of as infinitely postponed.No internal criteria seem to be able to substitute for the awakening.
The awakening,which does not of course have to be conceived as a mental act can be generally interpreted as a sort of collapse.In an expression like
I swim the truth of the expression is supposed to be changed by the addition of I dream that, and by consequence in
I dream that (I swim) not only the expression as a whole but also the part which I placed within parenthesis changes meaning.It collapses as if it were ,into fiction.The verbal clause turns from proposition to expressing something which might but does not really have to be the case. Dream is of course not the only verb which has this effect.Actually the whole wide field of intentionals make the same job;think,promise,imagine,remember,hope are only a few examples.They deal with the content inside the that-clause as something represented.And they make its propositional caracter collapse.So we could call such propositions which collapse inside intentional contexts collapsibles.
Are there non-collapsibles ? Descarte´s Cogito can (but probably must not) be constructed as containing the premise that
think is a non-collapsible.If it is so the expression I think in
I dream that (I think)
does not really change meaning by being inserted in the dream clause.In what respect could dreaming a thought differ from having the thought in general ? Similar seems to apply to I think (that I think) because here the introduction of the second appearence of I think into a context with the first appearence of it does not seem to make any difference to its meaning. I thought that P (where P is a proposition) but P was only a thought does not always make P collapse. For example if P is taken to be I thought .(This might have been an implicit element of what Descartes tried to say with his Cogito in which case the Ambulo,ergo sum, seems to be taken care of.Whether it takes care of Arnaults circularity argument is more doubtful.
Interesting experiments of thought can be carried out with
other intentional verbs.What about e.g. I remembered A (where A is a particular) and I remembered that (I remembered A) or I hoped that A and I hoped that (I hoped that A) .
Generally it seems that inserting such verbs (call them intentional if you prefer) that make propositions collapse into the propositional contents make them incollapsible.What is from the beginning representation cannot become more of a representation.
 In a metaphore the vehicle is related to the tenor.In a memory-image the image is related to the real event of which it purports to be a memory.
These and other types of connectedness seem more or less closely akin to the connectedness between the dream and the rest of our lives. And thus seem to raise similar difficulties when we try to analyze them.
Let us take a look at the metaphore.A metaphore is the result of juxtapositing two different ideas ,where,if the procedure is successful, the vehicle is supposed to teach us something about the tenor which we did not know,or did know but did not know that we knew before.This is sometimes expressed as that the vehicle throws a new light over the tenor.
The most interesting fact about metaphores is - of course - that only some of them are succesful . If they all were it would be much easier,because then we could claim that everything is able to represent everything and the connection between idea and experience is purely conventional.The same holds - of course - for the relationship between memories and memory-images.
This is actually very close to the problem which Malebranche discusses in his Recherche de la Verité and attempts to solve by his famous "seeing in God" a theory which might to some extent be expressed without theological terms by saying that Malebranche treats the connectedness between idea and experience as a primitive . One might remark that it seems to be a dangerous philosophical habit to leave to God to take care of all jobs which are too difficult for men to handle.But the case of Malebranche is somewhat special.A way to formulate his thought which might complate with the pious as well as with the seculars might be to say that his way of explaining the ties between ideas and experiences demands that clarifying the organisational system of our experience demands a logically stronger system than that system itself.
 "Dilemmas" Cambridge 1954 p.94
 "Je me représente la vaste enceinte des sciences,comme un grand terrain parsemé de
places obscures et de places éclariés."Pensées sur l´interpretation de la Nature" XIV