Metaphysics - Levels of Reality

hello youtube today i want to uh explore a notion that is somewhat popular in philosophy and perhaps in the sciences as well and that's the idea of levels of reality um so uh i should say i have a minor cold so um you know just bear with me on on that one if i'm coughing and spluttering a bit through this video uh but anyway the basic thought here is that reality is structured it structured into a hierarchy of levels and we have sort of higher level entities and lower level entities where higher level entities in some sense uh emerge from the lower level ones on the lowest level that we know of we have the elementary particles quarks electrons muons neutrinos bosons photons and so on i mean perhaps that's not the lowest level many physicists think that there's actually a lower level of strings but the the the quarks and electrons and so on that's that's the lowest level that we understand well at a higher level there are atoms and molecules and then if you go higher still there are cells and proteins and so on and then higher still organisms and then species and then ecosystems and i don't know maybe something beyond that but that's that's the basic picture it's a fairly intuitive one i think um now that the concept of uh levels plays an important role in philosophy many philosophical problems can be understood as really being about the question of what is the relation between two different levels for instance in the philosophy of mind if we are materialists we will presumably as to take it that minds arise out of processes going on in the brain and perhaps the wider body mental states are at a higher level than neural states and the big question is what is the relation between minds and brains what is the relation between these higher level psychological properties and the lower level neural biological properties so today we're going to ask how should the notion of levels be explicated and what is it that makes one entity a denizen of a higher level than another and i'm just going to briefly look through some some different options and explain some of the basic problems that they face so first thought is that levels are distinguished by the sizes of the things that exist in them with higher levels containing larger things elementary particles are very very small molecules are larger cells are larger again and and so on and now this view faces a number of problems first of all presumably the fundamental forces would all count as lower level properties but gravity occurs over a very great range the force of gravity shapes whole clusters of galaxies and you can contrast this with something like the strong nuclear force which is the force that binds uh protons and neutrons together into atomic nuclei this operates over the range of less than an atom so here we have uh two forces that are fundamental so these would seem to be lower level things and yet there's a very great divergence in the size of them or in the the sort of range over which they operate um but then you know i mean that there are responses here uh for one thing forces are transmitted by particles so uh the electromagnetic force is carried by photons and uh it is somewhat speculative at this point but perhaps gravity is carried by uh fundamental particles called gravitons um so it you know the the defender of this view might say well actually we can situate gravity comfortably at the lower level because even though it in in a sense it occurs over a very great range the force is actually carried by very very very small particles um i think that's not a very plausible response for a couple of reasons first of all because as i said gravitons are still a speculative idea i think that in metaphysics and so on we really should be appealing to current science we can't make assumptions about open scientific questions based on what might best accord with one's philosophical views um you know if it turns out the gravitons exist that's fine but maybe it won't maybe science will go in some in some other for some other theory uh in which case you know the levels picture would not be vindicated but second even if there are gravitons arguably gravitons should be understood as simply being at a lower level than gravity um you know gravity gravitons would explain how the force of gravity is realized you still have uh gravity as a force that occurs over a great range and and this would still be at a fairly low level if not the lowest you know i mean it remains the case that gravity operates on on me uh on a you know i'm a relatively small thing in the scheme of the universe and it also operates on on entire clusters or super clusters of galaxies so arguably this is still a problem there even even if there are gravitons um furthermore another uh as another example of of some of the difficulties with size um some structures that can be seen with the naked eye are whole molecules i believe that pure diamond for instance consists of uh billions of covalently bonded carbon atoms so it would in in some sense i think it arguably counts as a whole molecule uh similarly a salt crystal is simply a repeating pattern of sodium and chlorine in the ionic lattice i'm not sure that would count as a whole molecule i'm not an expert on chemistry but rather smaller scales dna strands are considered single molecules and they're sometimes many inches long although they're tightly coiled up but there's an example of where the chemical level of molecules um sort of bleeds into the biological level the second and more serious problem with size is that size just doesn't seem to be applicable to all levels consider electrons traditionally we think of electrons as being like tiny billiard balls or perhaps like tiny planets orbiting the atomic nucleus that's actually just wrong in fact electrons don't have specific sizes they are rather more like clouds of probabilities where the cloud is densest the probability of detecting the electron there is higher where the cloud is less dense the probability of detecting the electron is lower but there isn't any fact of the matter where the electron really is as it were it exists as a kind of smudge extending over an indefinite range in this image of a hydrogen atom we have the nucleus here this little uh pinkish dot and then this whole uh sort of black cloud is the electron that's what you know it's sort of it's it's just kind of smudge it's a cloud of probability similar problems with size arise in other cases what is the size of a species is a species the same size as its constituent organisms taken as an aggregate well if so then consider a species on the edge of extinction containing just one member in that case the species and the organism which are two apparently distinct levels actually have the same size on the other hand if we reject this uh approach if we reject the idea that a species is simply the size of its constituent organisms how how is the size of a species determined um i mean it would seem that on this view species just don't really have any particular size size isn't isn't applicable to species i mean for a more everyday example take um weight weight is a relational property defined as the force on an object due to gravity what is the size of an object's weight that question just doesn't seem to make much sense weight is not the sort of thing to which size is applicable so we can't distinguish levels by sizes if some entities or properties can't be said to have sizes the second approach is complexity as we move up levels entities and properties become increasingly complex molecules are more complex than atoms cells more complex than molecules societies are more complex than cells now this view faces uh again some some rather obvious problems first of all we would need to provide a clear definition of complexity there are measures of complexity but how we would apply these universally across the board is is questionable but even if there is a universal complexity measure it looks like it's just not going to be able to do the job um so consider water many of the apparently higher level properties of water are actually quite simple the surface tension which causes it to assume the least surface area or its viscosity which is a measure of its resistance to flow those are perfectly simple properties they're very straightforward yeah simple simple properties surface tension and viscosity nothing nothing complex there but on the lower level these properties are realized by extremely complex uh interactions there are billions of uh h2o molecules all constantly forming and breaking hydrogen bonds plus there are um dispersion forces which are a weaker attractive forces caused when the movements of an electron in a molecule induce a temporary a temporary dipole you know there's all these things going on on the molecular level very very very complex interactions which realize these simple higher level properties so in in that case the the lower level looks more complex than the higher level similarly consider mental states such as beliefs desires emotions pains compare that with the incredible complexity of all of the neural connections in the nervous system you know so far from higher levels being more complex often when we move up a level we gain enormously in simplicity now one response to this is to say well a higher level is attained when lower level phenomena become sufficiently complex that it's more useful to treat them as a whole you know to look at them as sort of being one thing rather than treating them as an aggregate so you can you can kind of treat um water you don't when we interact with water on the higher level we don't treat it as something made of billions of little molecules um you know we just sort of treat it as a continuous fluid um and that's so we might suggest look that's because uh the the lower level phenomena have become sufficiently complex that it's just more useful to treat it as a whole thing rather than as a kind of aggregate so levels are defined by complexity but the complexity that matters is on the lower level the problem with this view is that it looks to me like this just means that levels become a matter of what we can understand uh they they are epistemologically important but they don't they don't have any kind of metaphysical existence there are not really any levels of reality levels are just kind of useful pragmatic fictions um you know we we uh it's easier for us to understand water or to sort of see water as being a continuous fluid rather than as being composed of billions of little molecules but but that's just a sort of fact about us it's not it's not a fact about reality in any sense we'll return to this point a bit later a third option is composition we might suggest that higher level entities are in some sense composed of lower level entities molecules are composed of atoms cells are composed of molecules organisms are composed of cells now if we're going to treat levels as being defined by composition relations the obvious question is what kind of composition is required presumably not just any composition will do a computer is composed of the screen keyboard the mouse pad and so on but we wouldn't consider a computer to be a higher level entity than a screen or a keyboard of course some people might try to bite the bullet here and say that actually the computer is a higher level entity than the screen so that if if entity a is composed of entities b c d etc then a is a higher level entity than b c d etcetera i i would say that that view just trivializes talk of levels we might as well just drop the concept of levels and talk simply of composition in that case you know if any kind of composition um creates a new level then what's what's this concept of levels even doing just talk about composition uh that's a much much more simple concept and that's all all you would really be talking about there so perhaps we could suggest that uh levels are characterized by types of things certain types of things are composed by certain types of other things biological things are composed of chemical things but it's difficult to see how to make that precise in a sort of non-question begging way um after all there can be different types of things on the same level uh like the different types of chemical elements the most serious difficulty with this approach is that composition relations between different types of things are often vague and context dependent so consider for example what is a gene composed of genes were postulated to explain patterns of inheritance of phenotypic traits and these days they're often identified with sequences of dna that encode proteins however complications on the path from dna to rna to protein make uh the sort of straightforward identification of genes with dna sequences quite difficult um so dna sequences contain coding sequences known as exons and non-coding sequences known as introns after the transcription into rna introns are removed from the rna transcript and the exons are spliced together this isn't a rigid process um through through mechanisms known as alternative splicing exons can be removed and they can you know different exons can be removed in different ways so one dna sequence can give rise to various rna transcripts as you can see in this diagram you can splice exons together in in different ways for the drosophila d scam gene alternative splicing can produce 38 016 different rnas which allows the the same dna sequence to play various different roles in different tissues uh that's discussed in the article alternative splicing of the dresser villa d scam prm and pre-mrna uh is both temporally and spatially regulated by collotto and gravely another intriguing process is dna methylation in which methyl groups can be added to dna nucleotides usually repressing gene expression so a methylated dna sequence will play a very different role in the genome from an unmethylated sequence now it's important to note that these are not just um accidental processes as it were processes like this are absolutely essential for the proper expression of dna patterns of dna methylation are heritable from parent to child and without it the organism simply wouldn't be able to grow properly alternative splicing is an essential component in the operation of the immune system for instance it's it's basically it's one of the things that allows the immune system to produce potentially millions of different kinds of antibodies and so tailor itself to specific infections you can check out nessa carey's book the epigenetics revolution if you're interested in this topic the point for us is this when we talk about genes as higher level entities we're generally referring to those mendelian entities that explain patterns of inheritance of phenotypic traits these are the entities we're thinking about when we draw punnett squares like this here we're interested in the probabilities of the inheritance of certain traits from parents to offspring and um you know we look at the the sort of laws that govern that inheritance and so on but genes in this sense may be best conceived not simply as composed out of dna sequences but as being composed of dna sequences plus various other processes involved in the expression of the dna which is of course dependent on various other processes going on in in the cell on this view there is no single answer as to sort of what composes a gene it's it's context dependent um you know in in different circumstances genes may be composed of different things so there may be um some problems appealing to uh composition a fourth view is that levels are distinguished by causal relations uh william wimzat in his book re-engineering philosophy for limited beings suggests that we can distinguish levels because entities of different levels interact primarily with one another we can draw uh causal relations between various entities and where causal chains are dense we have a level so one problem with this is that causality like composition is often you know it's context dependent it's often a matter of perspective we we assume that humans normally interact with other humans and not with human cells because we easily see humans whereas we can't normally see cells if we saw cells as easily as we saw humans we would probably see all kinds of causal interactions you know sort of the human as an organism and the cell you know so you know we see sort of the huma a human organism causally interact with the cells of another human or with another organism as another example suppose an asteroid impact occurs well what entities does the asteroid interact with does it interact with cells or individual organisms or with whole species um does it interact with individual trees or with the whole forest i mean there's simply no way to tell without first assuming the level's picture i suppose prima facie we would say that it interacts with entities on all levels you know interacts with molecules with organisms with species and so on but you wouldn't want to say that the asteroid exists on all levels um you know what would we say that there's some like a a different asteroid for each level or yeah i mean it's that that doesn't seem to make so much sense so a different form of the causal view is proposed by craver and bechdel in their article top-down causation without top-down causes unlike whimsic craver and bechdel see levels not as large scale structures of reality but as arising only in the context of mechanisms so they say an item x is at a lower level than an item s if and only if x is a component in the mechanism for some activity psi of s they define mechanism as follows they say mechanisms are entities and activities organized such that they exhibit a phenomenon as an example consider the eye the eye transduces light into a pattern of neural activity that's the highest level understanding of what the eye does now we can compose this process decompose this process into various lower level activities light enters the eye where it is focused by the lens onto the retina where it is converted into a pattern of activity in the optic nerve here we have entities and activities organized so that they exhibit the phenomenon of the transduction of light into a pattern of neural activity and we can decompose this further focusing on the retina we can go to a lower level to explain its functioning we can consider the activation of rods and cones in the retina how there are different types of cones which enables color vision and how these send signals to ganglion cells and so on so that the higher level phenomenon of activation of the retina is explained by appealing to lower level rods cones and ganglion cells and of course we can move to a lower level still to explain the functioning of those cells now interlevel causality would amount to causation between parts or components and holes of mechanisms it would it would be a causation between a part of a mechanism and the mechanism itself which craver and bechtel take to be impossible the cones are a component of the mechanism of activity in the retina so it seems to make no sense to say that cones cause activity in the retina the the cones are part of that activity they're part of what that activity consists in so we can distinguish levels by tracking causal relations ebola causes frank's death so the ebola virus and frank occupy the same level that's perhaps a surprising consequence we would usually think of viruses as lower level entities than people but remember that we're not at this point trying to distinguish levels by appealing to size and you know the point is causal relations and small things can be causally related to big things note that this is quite a limited view of levels on this view levels arise only in particular contexts we can't organize all entities into a clear hierarchy of levels we find levels where we have some complex phenomenon that can be analyzed into parts the phenomenon is on the higher level the parts are on the lower level and those parts might be analyzed into further parts that are on lower levels still however there are problems even with this limited view so we've just seen that ebola causes frank's death so ebola and frank must occupy the same level but now suppose we wish to know why frank is maintaining a fever of say 39 degrees centigrade well in this case it looks more like the virus should be treated as part of the mechanism producing the effect it certainly seems to count per craver and bechtel's definition of mechanism remember they say that mechanisms are entities and activities organized such that they exhibit a phenomenon the virus is among the entities and activities organized such that they maintain frank's body temperature above the norm the virus combined with various molecular processes occurring in frank's body is what realizes the phenomenon of his higher body temperature so now it appears that the virus is on a lower level in frank given the question of frank's high body temperature the ebola virus must be treated as being on a lower level than frank but given the question of frank's death the ebola virus must be treated as being on the same level as frank so i mean one response here is um you know that okay one yeah the ebola virus isn't part of the mechanism producing frank's high temperature um you know perhaps we could say that the mechanism here uh just involves processes of frank's own body cells uh you know that that end up creating the fever and the virus is the cause of the the one of the causes of the fever rather than a component of any mechanism producing it but i mean that's sort of my point right it seems perfectly legitimate to treat it either way uh we can the virus it doesn't really seem to make much difference whether we treat the virus as a component of the mechanism producing the fever or as a cause of the fever that just you know just seems to sort of depend on our perspective and these kinds of cases are quite easy to proliferate is the gut microbiota part of the mechanism of digestion or is it merely something that has a causal influence on digestion uh it seems to me that this is you know just just i mean it doesn't really matter what we say here both views are equally legitimate uh without the gut microbiota we wouldn't be able to digest food or at least we wouldn't be able to digest it nearly nearly as well um you know whether whether it's part of the mechanism that you know realizes the phenomenon of digestion or whether it's a causal influence on that mechanism what what difference does this make um so the point is there are various cases where the distinction between constitution and causality really just seems to be up to us and you know so we can't then appeal to causal relations as uh as as defining levels of of reality um you know metaphysically so those are a few um a few ways we might try to define levels of reality and as we've seen they they have some problems um i've been quite brief with each of them so you know the various ways we can make them more sophisticated but that's something to think about i should note that um as a fifth option we might try some combination of these previous four options perhaps levels of reality are determined by uh size complexity composition causality and maybe other features and different features apply to varying degrees in different levels that's certainly a possible view um although i think it would probably take quite a lot of effort to make it precise and to develop it in a way that avoids the problems with each feature considered on its own so um yeah i'm not going to talk about that too much here but but that's an option uh now i should also note that anybody who holds that there are levels of reality will need to explain why this view is to be preferred to let's say a deflationary account according to which levels are simply constructed by us levels are epistemological they are um we group certain phenomena together because we find it useful to do so as a purely pragmatic matter we can't treat everything as simply being fermions and bosons that would be far too complex so we understand the world in terms of various so-called higher level entities but metaphysically there is nothing beyond the fundamental level ultimately there are no higher level entities there are just fermions and bosons or perhaps strings you know whatever physics ultimately decides is that the fundamental level and everything else beyond that is is constructed by us it's it's a sort of useful tool for us you know so so the the kind of reductionist view the the sort of hardcore reductionism which holds that there's nothing but fermions and bosons why should we prefer the levels picture to that hardcore reductionist picture before we end let me note an important argument against the notion of levels of reality this is jaguar kim's exclusion argument suppose we have some higher level properties or entities m emerging from lower level properties or entities p m might be mental states and p might be brain states if we're going to say that m and p are literally different things that occupy different levels of reality it looks like we need to say two things first we have to say that m is supervenient on p to say that m is supervening on p is to say that you can't change m without changing p so if mental states are supervenient on brain states this means that any difference in mental states entails a difference in brain states this seems reasonable enough and that's true of all high level entities the point is that higher level entities don't simply float free as it were of of lower level ones um in some sense the entities of the higher level must be dependent on the entities of of the lower level uh you know the minds are dependent on neural states or perhaps you know states of the the brain and the body but the point is there are sort of physical states that minds are dependent on and that's the point of um requiring supervise second we need to claim that m is distinct from p or irreducible to p so in some sense mental states are more than just brain states if mental states are just brain states then of course we don't really have two separate things here we don't have two levels we just have brain states um you know and presumably brain states themselves would ultimately be eliminated in favor of fundamental particles in order to sustain the notion of levels we have to give some sense to the idea that m although it's dependent on p doesn't end up just reducing to p um you know that m has to be something more than just p um so so the the the levels picture requires uh supervenience and distinctiveness but kim points out that there's there's a problem with this with this view suppose some higher level event m1 causes um another high level of an m2 so m1 might be the belief that i have won a prize and m2 might be the feeling of pride that's a perfectly standard causal claim about mental states the belief that you have been successful at some difficult endeavor causes the feeling of pride so let's diagram what's going on here we have m1 causing m2 of course m1 and m2 are each dependent on brain states so we also have p1 and p2 presumably p1 causing p2 this is the intuitive picture the red arrows represent causation and the blue arrows represent the supervenience relation but this picture doesn't really make much sense um because the higher level properties are dependent on the lower level properties so you bring about m2 by bringing about p2 right that's you know you don't just get m2 as it were the way to bring about m2 is to bring about the lower level properties that realize it so we're driven to a picture like this where m1 is the cause of p2 and so it exhibits a downward causation now there are some obvious problems with this picture first of all m1 itself arises from the physical properties p1 since p1 determines m1 i mean surely the real cause of p2 is going to be p1 uh p1 is is what's doing the the real work here um i mean more seriously uh you know saying that to say that the physical properties are not uh say that these physical properties are not caused by other physical properties doesn't really seem to be compatible with with modern physics a basic assumption of modern physics is that if any state of physical particles has a cause then that cause must be a prior state of physical particles so it looks like the physical properties p1 preempt m1 as a cause of p2 giving us this picture on this view the higher level properties are mere epiphenometer they have no causal powers and that raises the obvious question uh you know if high-level entities are mere epic phenomena what's the point of postulating them at all if m1 has no causal powers above and beyond p1 why would we think that m1 is something separate from p1 this just seems like a completely unnecessary expansion of our ontology uh yeah occam's razor tellers tells us not to postulate entities without necessity so why what's the point of of this kind of picture why not just kind of cut away m1 and m2 and just have this sort of extreme reductionist view that you know there's nothing but the physical particles so perhaps we could suggest that p2 is over determined it's caused by both p1 and m1 now over determination does seem to occur in some circumstances let's say that both frank and vincent shoot john and both bullets reach john's head at exactly the same time and then it looks like john's death is over determined it was caused by both frank and vincent if frank hadn't been there john would have died from vincent's bullet at exactly the same time and if vincent hadn't been there john would have died from frank's bullet at exactly the same time but this is a special case um you know if we if we if we use over determination to save the levels picture then it looks like we have to assume that over determination always happens it seems rather ad hoc in any case there's a deeper problem with over determination in this context we can't say that if p1 hadn't been there p2 would still have been caused by m1 because if p1 hadn't been there m1 wouldn't have been there either or at least i mean m1 could have been there but only by substituting some very specific physical entities and properties for p1 you know it takes a very specific kind of brain state to realize the belief that i have won a prize the point is that in the shooting case merely removing frank will mean that john still dies because john will die by vincent's bullet similarly merely removing vincent will mean that jon still dies he'll die by frank's bullet but if you merely remove p1 if that's the only change you make then in doing so you also remove m1 and there's nothing left to cause p2 so there just doesn't seem to be any reason for supposing that p2 is over determined it looks like p1 is doing all of the work so that's levels of reality uh the the notion of levels of reality is uh i think it's an intuitively attractive picture uh what i hope to have shown in this video is just that it's it might take some work to make this intuitive intuitively attractive picture philosophically respectable um you know there are clearly a variety of ways of spelling out the notion of levels and however we spell it out it looks like we're going to face some sort of exclusion problem so we would need to answer that so that's all i wanted to talk about today thanks for watching

Channel: Kane B

Views: 4,680

Rating: 4.9266057 out of 5

Keywords: metaphysics, emergence, levels, jaegwon kim, exclusion argument, downward causation, causation, causality, composition, complexity, philosophy, ontology, genes, mechanisms, philosophy of science, science

Id: UitCVVoUiOc

Length: 37min 32sec (2252 seconds)

Published: Fri Sep 23 2016