A core concepts of postmodern philosophy is relativity - the idea that there is no absolute truth, method to establish truth, or generally valid ethics, but that all these are only meaningful relative to a particular perspective. In other words, what is a valid standard e.g. to find out whether a statement is true depends on who asks the question and his social and cultural context.
In contrast, science in general aims to establish objectively true statements about reality. For postmodern philosophers in turn, science is just the discourse of a particular group of people and no more or less true than other discourses - that is to say, what science holds true is true within the scientific community, but can't be claimed to be equally true outside. Naturally, scientists vehemently disagree with this position.
In order to get an idea where these different ideas come from and why their proponents think they are valid, it is useful to take a more detailed look at what we commonly call 'reality', how aspects of that are relative to perspective and what we really mean by 'truth'.
RealityReality is a concept that features rather prominently in Postmodern Philosophy, for example in statements like 'Reality is created in social discourse.' However, the words 'reality' and 'real' can be applied to quite a number of different things, so what does a sentence like the above actually mean?
For example, one would probably not view last night's dream as something real when contrasted with a cup of coffee which is ready to grab on the table. However, the same dream could be called real in comparison to another dream which one imagines one could have had, but did not actually have, or to yet another dream which is so strange one cannot even imagine it, but only formulate the abstract thought that such a dream could occur. After all, last night's dream is something one actually experienced, to be contrasted with something which one did in fact not experience but merely imagine.
Another example is the value of money. In some sense, this value is something very real, as it causes observable experiences (as anyone can discover by trying not to pay his bills). Yet, in a different sense, the value of money is not real - it depends on a shared belief. Someone who does not know the concept of money will not see a 100 US$ bill, but only a small sheet of paper, and it would not occur to him at all that he should trade a cow for this paper - after all, a cow is useful, the paper isn't. Trying to pay such a person with money will quickly demonstrate that there is no intrinsic value in money, but that it is only valuable as long as the other person believes it is. As it happens, the same is true for the value of gold. Gold is valuable because people believe so. Part of the reason for this is that gold is a rare substance, but there are plenty of rare things that are not valuable - my early childhood drawings for example are unique, and yet nobody would consider them valuable.
It is thus reasonable to make a few distinctions. The most useful starting point is the conscious self: We know that we are conscious, and we know who that conscious 'I' is. There appears to be something like an inner world, in which thoughts, dreams, feelings and such like are experienced in consciousness, but there also appears to be an outer world, which is reflected in the conscious mind through perception by the senses - sight, hearing, touch, smell, taste, balance, acceleration and kinestetic sense.
It is possible to formulate a philosophy which does not assume the existence of an outer world - this leads to a position known as Solipsism which will be dealt with elsewhere, here we follow the line of thought which assumes that there is indeed an outer world. If so, one can formulate four different categories of possible experiences. The first class of experiences is dominated by properties of the inner world. The second class is mainly driven by things which exist in the outer world. The third class arises from both the state of the inner world and the conditions of the outer world, and the fourth class of experiences is a special subset of the third and deals with experiences made in the interaction between people - these depend on the state of the inner worlds of two (or more) persons.
To provide an example to illustrate these classes: Suppose you are preparing for a trip to Paris and you have never been there before and have not studied a travel guide. The way you imagine Paris to be before you go there has much to do with your inner world, but close to nothing with how Paris actually is (since you don't know that). Thus, this experience belongs to the first class. On the other hand, the number of churches in Paris is something that has nothing to do with your inner world, but much with how Paris itself is, therefore this is a question associated with the second class of experiences. The third class is illustrated by a question like what the flair of Paris is like. Clearly, that has to do with Paris and its scenery, but it also has to do with your state of mind when you observe it. The same scenery will lead to a different response when you are depressed or when you are eager and open for new experiences. Finally, the question if Parisians are friendly people depends on both your frame of mind and on the mental state of the people you meet. If you insist in speaking English, then your experience will probably be very different than if you are willing to try French.
Of course, making a distinction between these classes is to some degree arbitrary. One cannot really identify phenomena which are exclusively limited to the inner world (after all, dreams reflect observations of the outer world or your imagination of Paris reflects things you have may have heard about Paris), and neither can one be absolutely sure that a phenomenon exists exclusively in the outer world, as the senses are imperfect and there is cross-talk between the state of the mind and the awareness of sensual information. Thus, conceptually there is a continuum of experiences which are all more or less driven by both the inner and the outer world. The classes then are a matter of degree - for example, there are experiences for which there is every good reason to believe they are almost exclusively driven by the outer world, and for practical purposes this is good enough to warrant a class in which the role of the inner world can be safely neglected (if you do not accept this argument and insist that for every experience in principle and in practice both inner and outer world are relevant, please refer to my counter-argument here).'reality as it is' rather than 'reality as it appears' below. Thus, in the following, we will frequently refer to the observer - a self-conscious mind making an experience. The four different classes of experiences could then be labelled as follows:
Subjective experiences are those which (almost) completely depend on the mental state of the observer. One example would be the sensation of pain. Pain is obvious to the observer, but there is no way another outside observer could conclusively verify that a person feels pain or not, and also to judge and compare the level of pain among two people is quite impossible - except by asking the two to describe and rate their experience themselves! Similarly, if you would feel pain, you would not find any amount of Neurologists taking scans of your brain and concluding that it is quite impossible that you feel pain because the pain centers of your brain are inactive as a proof that you don't really feel pain - presumably you'd insist that you feel what you feel regardless of what the scan says. Other examples for subjective experiences are dreams, visions, fantasies or emotions. We tend to think of subjective experiences as 'not real' - however they are quite real in the sense that we actually experience them.
Objective experiences on the other hand are those who do not depend on the mental state or any attitude of the observer. This implies that a single, unaided observer has no possibility to distinguish between what is subjective and what is objective. Rather, what is objective has to be deduced from observations which are common to many observers and from agreement of observations aided by detectors - tools to overcome the limitations of the senses. The scientific method is a framework designed to remove the observer dependence from observations (as far as this is possible) and hence gives access to the objective. An example for an objective experience is gravity: Any observer who drops a stone in the gravitational field of Earth will observe that the stone falls down (as long as there are no other forces acting on the stone - if you attach the stone to a rubber band, it will of course not fall down), and this observation is independent of what the observer believes or expects will happen: Even for an observer who expects the stone to float upwards, the experience will still be that the stone falls down. Other examples include the solidity of rock, or in general the physical properties of the outer world.
Interactional experiences depend on both the mental state of the observer and the properties of the outer world and arise between them. They are more general than purely subjective, but do not generalize enough to be called objective. As an example, consider the experience of visiting a graveyard at night. For most people, the experience is probably scary. This has to do with the outer world - they would not find the same place scary when visiting at noon, or they would not find visiting downtown at night a scary experience, so the experience is linked to something objective like a place or a time. Moreover, the experience is shared among people - it is not difficult to find others who will, given the same circumstances, make the same experience. And yet it is not universal - there are people who will (for whatever reason) not be scared when visiting a graveyard at night. The term intersubjectivity is used to describe such shared non-objective experiences.A good sign of an interactional experience is that the experience is only made if the observer approaches it with a given frame of mind, but as long as this is the case, it is repeatable - but if the observer has a different frame of mind, he will not make the experience. Other interactional experiences are finding meaning in an event, perceiving purpose in a development, or having a peak experience (an intense, religious-like awareness of the universe).
Communicational experiences arise from the interaction between two (or more) persons and depend on the mental state of both. As an example, consider the regulations for a public forest. If they are experienced depends on the interaction between the observer and someone who formulates and enforces them (a 'lawmaker'). If the observer walks into the forest and does not know the regulations, he will never experience them as long as he does not meet anyone. Even if he knows them, he may voluntarily follow them or ignore them, but that is his choice alone and a question to his state of mind as long as he does not meet anyone. If he meets a lawmaker, but the lawmaker is unable or unwilling to enforce the regulations, the observer may still ignore the law without experiencing it. To turn the regulations into an experience requires the observer to know them, and the lawmaker to be willing to enforce them, i.e. the validity of the law arises from the communication between two people, and ultimately is linked to the power to enforce one idea over the other. In a similar example, the value of a 100 US$ bill requires both the buyer and the seller to agree that it is not a piece of paper, but represents value. Communicational experiences are often not openly recognized but implicitly learned by growing up in a society in which everyone accepts given belief sets. For the medieval people, the divine mandate of the king was as real as the value of money is for us - a belief the whole society agreed on.
One can now use these four categories to classify actual experiences. Here, actual experiences are those which can not be simply altered at will. For example, although pain is subjective, it is not possible to end it by deciding not to feel it any more. Gravitation cannot be altered by an act of will, neither it is possible to decide not to be scared on a graveyard or to escape punishment for theft just because one decided that the law doesn't really exist. This is to be contrasted with imaginary experiences in which one visualizes or remembers an experience under conscious control. It is quite possible to imagine how it feels to suddenly ignore gravity and float away, or how the world would be if one could ignore an inconvenient law. While the act of imagination itself is something that in a sense 'really' happens, it is generally acknowledged that such imaginary experiences have a lesser degree of reality than actual experiences. An even less real class of experiences are potential experiences - they correspond to the abstract thought that it is possible to visualize a world without gravity without actually doing so - so these are experiences in which not even the imagination 'really' takes place.
'Reality as it is' and the aim of knowledgeWe tend to have the notion that what we experience should be described in terms of a reality, which is somehow filtered through the act of perception. One can advocate the position that there is really nothing beyond the perception which is filtered, this again corresponds to Solipsism. But once one moves away from Solipsism, then one has to acknowledge the existence of something which is perceived.
The idea of reality is it is is now the separation of the thing which is observed from the act of observation. Note that this is very different from the scientific method - the scientific method aims at removing the dependence on the mental state of the observer from the observation, but the observer and the observation are still present in the scientific method. Indeed, modern physics has a clearly defined separation between observables and non-observable quantities, and only the observables have physical relevance. The question of what things are when we do not observe them is, at least within modern physics, not considered meaningful and not really addressed.
The question of reality as it is is then difficult to answer directly. Our primary connection to anything is experience, and the question how things are when not accessed through experience or observation them is by definition beyond direct experience. The question may then be raised if knowledge of reality as it is is of any use at all, if such knowledge cannot, even in principle, be linked with any experience.
Consider someone with a glass of a clear liquid who asks 'Is this water?' A possible answer would certainly be 'According to our best knowledge, there is no fundamental substance called 'water'. Rather, what you're holding is a very complicated quantum state of fundamental matter and gauge fields with a particular large-scale ordering structure.' This answer is almost certainly not what such a person is asking for. When someone asks for knowledge, most of the time it is knowledge relating to her (actual or potential) experience - in the particular instance, she might like to know if she can drink the liquid safely, or if she is a chemist, if it has the chemical properties of water and can be used in an experiment which requires water as solvent - but knowledge on a fundamental level far from her experience is probably not what she is after.
To see how the point relates to reality as it is, assume for a moment our world would be just as depicted in the movie 'Matrix', i.e. we'd all be plugged into a supercomputing grid which artificially generates all our perception by directly interfacing with the nervous system. All the physical laws of the universe which we perceive would in this case really be parameters and instructions within the computer programs generating our perception - they would be reality as we experience it, whereas the supercomputers would correspond to reality as it is.
In the movie, the protagonist Neo is taken out of the Matrix, experiences reality as it is, and when he returns, he experiences that he can bend the laws of physics within the Matrix (since they are not real), for example he can move with superhuman speed or defy gravity. But in this case, reality as it is influences what can be experienced purely within the Matrix - people can watch Neo fly, they can observe him bending the physical laws, and from this they can deduce that there is something beyond what they've taken as physical laws before and deduce at least some of the laws of that something, although they might never be able to determine that that something is a supercomputing grid. It is even conceiveable that without ever stepping out of the Matrix (like Neo did), a hardcore skeptic who does not trust his perception discovers that he can circumvent the apparent laws of physics and from this deduce the existence of something beyond the apparent.
However, now consider a perfect Matrix - one cannot step out of it, one cannot bend the apparent laws, one cannot experience anything beyond the apparent laws. For someone within this perfect Matrix - what use would be the knowledge that in reality everything is just a simulation? He would still die when jumping from a cliff, all his experiences would be governed by the apparent laws - so in order to structure his life, to make decisions which influence his experiences or to understand what he sees, knowing the apparent laws, even if they are not reality as it is but just reality as it appears, would be far more useful.
Thus, usually we ask for something which at least potentially relates to our experience. If something is an illusion is only relevant if we have some way of seeing through the illusion, if that knowledge makes any difference. But if there is even in principle no way to relate some knowledge to any possible experience, it's not quite clear what the meaning of such knowledge would be.
As access to reality as it is remain then (at least) two possible paths: The first is the trust that a theory that can successfully predict from one observation what we will see at the next observation must contain at least some properties of whatever happens inbetween. This doesn't need to be true (as the Matrix example shows) - a conspiracy of some sort might deceive us. The second path utilizes the observation that the conscious mind has a double role: In one sense, the mind is the observer, but the mind itself must in some form also be part of reality as it is. Thus, the mind observing itself and only itself could be expected to lead to an experience of reality as it is - simultaneously being the observer and the observed. The first path leads to fundamental physics, the second to mysticism - and it is interesting to note that both paths lead to a non-dualistic picture of the world.
Principles of measurementIn science, a measurement should be an observation of nature which, as much as this is possible, eliminates the role of the observer and the particulars of the observation. This leads to a principle like minimum bias (the observation should not include a selection effect - for instance counting the number of sick people in a hospital will not lead to reliable estimates for how sick the general population is because people go into hospital because they feel sick). Another important idea is repeatability - a different scientist should be able to redo a measurement and get the same answer. There is consistency, i.e. the same quantity measured in two different ways should be the same value.
These however are principles, not laws - there are cases in which any of them is dropped. For instance, cosmology has only one universe to study, there is no possibility to repeat the experiment. Particle physics experiments often cannot be done selection-bias free as some particle species are much harder to detect than others and the biases need to be modeled.
Finally, it is worth mentioning that in Quantum Theory, the observable can even in principle not completely separated from the act of observation, i.e. observing the same physical state (a photon field) in different ways might lead to different outcomes (light as wave or light as particle).
Principles of deductionIn going from measurements to tentative laws of nature, various principles of deduction come into play. Partially they are formal logic - for instance transitivity: If A implies B and B implies C, then A implies C. They are rooted in everyday experiences in that we learn that nature around us actually works that way (they're not mathematical necessities and alternative consistant logical deduction systems can be formulated - they don't describe nature however). Standing on a weaker foundation, selection principles such as Occam's razor can be found which states that of two competing ideas, the one requiring less assumptions should be preferred. Unlike logical deduction rules, such principles have less to do with mathematics and more with efficiency. However, just like logical reasoning principles, deduction rules are rooted in everyday experiences. Seeing dirty footprints in the house, a woman will probably suspect her husband rather than visitors from Mars faking footprints because usually the simpler explanation turns out to be true.
Another tool for deduction is probability calculus in which the likelihood of truth given some information is inferred. The philosophical underpinnings of probability are rather complicated and we will not discuss them here. However, it can be stated that practically all deduction principles of science are formalizations of what we usually call 'common sense' (even if that may be difficult to see once things are formulated in abstract math). There may however situations in which common sense fails and the applicability of the scientific method is questionable - see here for a deeper discussion.
Principles of modelingModeling is the way a theory about nature is formulated and applied to make predictions which can then be verified or falsified by new measurements. It is important to realize that scientists are usually unable to work any but the most trivial consequences of a fundamental theory even where it exists (e.g. in physics) and often do not even have a fundamental theory of anything (e.g. in psychology). Thus, more often than not, simplifications are used, the models. One thing that is confusing for non-scientists is that often a model can be disproven, but this doesn't imply a failure of the underlying theory, it might just imply that the assumptions under which the model was made were not valid.
A good model makes controlled approximation, i.e. it can be mathematically demonstrated that the model is different from the real theory by no more than a known amount. Likewise, a model should be widely applicable, i.e. not only explain one set of measurements but account for a large body of data. Any model should also incorporate core physics principles like the conservation of energy or causality - this in essence is just another consistency requirement.
In systematically testing models, both against data and against consistency with other models, and in using formalized deduction rules, the scientific method aims at preventing the observer from fooling himself and at establishing observer-indepent results as much as this is possible. The question raised by postmodern philosophy is whether this does not in itself constitute a very particular perspective, and whether results obtained by science only make sense relative to that perspective. To understand this better, let us proceed to investigate relativism in philosophy and what philosophers have argued about science from this perspective.
RelativismRelativism is the general frame for various postmodern philosophical ideas, which can be formulated in the way
'X' is relative to 'Y'
where different philosophers and thinkers have inserted various things for 'X' and 'Y'. Here, 'X' is usually something which is commonly (naively) believed to be something universal and objective (ethical norms, concepts of thought, epistemic standards, perception, truth, reality...) and 'Y' a frame or point of view from which things are observed (society, language, personal choice, neural architecture of the brain, social status...). Relativism is then the statement that 'X' is not an objective category but rather depends on the observer, in particular on properties of the observer with regard to 'Y'. Some combinations of 'X' and 'Y' are completely uncontroversial ('perception is relative to neural architecture'), others are quite provocative ('truth is relative to personal choice').
Every possible relativist statement has a descriptive form which contains the observation that there are people who disagree on 'X', for example 'People from different societies have different ideas about what truth is.' The descriptive form of the statement does not imply the validity of its content - it could well be that people from different societies have indeed different ideas about truth, but some of them are simply mistaken. The normative version of a relativist statement is stronger, as it claims that its content itself is valid: 'Truth is relative to society.' is a claim that there is in principle no way to establish truth outside the context of a particular society.
There are also weak and strong versions of relativist ideas. This distinction has to do with the amount of variation found in 'X' given a change in 'Y'. Strong versions claim that the variation of 'X' can be arbitrarily large, weak versions are more careful and state that while the general form of 'X' is fixed, variations may still occur in detail. For example, a weak version of 'Truth is relative to society' would maintain that there are societies in which 'grey' statements like Bill Clinton's 'I did not have sexual relations with that woman' would be accepted as true, but there are also others in which they would be seen as false, but there would in all societies be sentences like 'I was born from my mother.' where no disagreement about truth or falsehood is possible. A strong version would claim that even the truth of the latter example is not universal, that a disagreement about its truth is possible and that its truth cannot established outside the context of a society.
The two fronts of relativismAny relativist position amounts to a tension between two rather different fronts. Let us consider the example of 'truth' being relative to some 'Y'. The first front could be called the anti-realist front. Here, the argument is that there are no objective, framework-independent criteria to establish truth, in other words the claim is that there is no objective reality corresponding to the meaning of truth (claims in philosophy that there is no objective reality behind X are often alled 'anti-realist').
On the second front, however, the argument is rather different. The problem to be addressed is that no one's experience is that thinking about something makes it always true or factually correct already. In other words, we experience that our expectations can be mistaken, that not everything we think turns out to be correct. But that implies that within a framework, i.e. once a definite 'Y' is chosen, the relativist position must argue that there are facts to be established, that one can distinguish between something being true or false. If that cannot be done within the framework, then it follows that there is no criterion to tell if what one thinks is true or not, and hence that thinking something to be true is the only remaining standard. From here, Solipsism follows. Thus, within any framework 'Y', the position to be argued must be a realist position.
The philosophical problem is then to define the meaning of truth such that it can be established within a framework and is distinct from just the thinking something to be true, but formulate the definition such that it does not rest on anything objective outside of a framework. This implies that any relativist position is characterized by a mixture of objective constraints and subjective freedoms. This is also clear from the observation that 'Y', the framework relative to which an 'X' changes, must refer to something absolute. If not, the claim undermines itself. In the definitions outlined above, relativist positions are applicable by definition to the the interactional part of reality - experiences which depend on both the state of the observer and the objective outside world.
However, solipsism is neither particularly interesting (basically things just are as they are experienced for no particular reason), nor are the other implications of it usually what people do in their lifes. For instance, if there is no outside world of any reality, there is no ethics - one can not steal something that does not exist, or kill someone who does not exist. There is no need to accumulate possessions since they do not exist.
As an argument in a debate, solipsism therefore doesn't really work - it can end the debate since solipsism can't be refuted, but it is obvious that no speaker who argues the position actually means it.
Relativity of perceptionRelastivist positions with regard to perception argue that we do not observe the world as it is, but rather that our perception is in an essential way influenced by factors like expectation, concepts or beliefs. This can be summarized in perception is theory-laden, i.e. to some degree we see the world as we expect to see it.
Perceptional relativism can be a challenge to the validity of the scientific method. Some postmodern philosophers make claims that the direct evidence from observation is not reliable, with the implication that if not even the evidence is something objective, no interpretation of the evidence such as a scientific theory can be. One finds e.g.
People see what they expect to see, and that the categories of their perception are largely if not wholly determined by their social and cultural background. So members of different cultures may see the world they live in very differently. And it is not just a matter of reaching different conclusions about the world from the same evidence; the very evidence which is given to them as members of different cultures may be different. (John Beattie 1966)
Given appropriate stimuli, but different systems of classification (different 'mental sets'), our perceptual apparatus may produce perceptual objects which cannot be easily compared. (Paul Feyerabend 1993).
Claims with regard to the theory-ladenness of perception typically refer to experiments with ambiguous images or optical illusions. Consider for example the following picture of a woman. It can either be seen as a young woman, turning away her face, or as an old woman facing left.
However, taking a bit of time to actually consider the evidence shows that the problem isn't really there. The scientific method acknoweldges that there is bias due to the observer, but maintains that the steps taken to ensure that an observation meets scientific standards are sufficient to remove the problem. Let us start considering the ambiguous figures first. Clearly, both the woman and the animal above are neither woman nor animal - they are drawings - black lines on a white background which are suggestive of depicting something. It is impossible to find out what they 'really' depict. But consider seeing a woman on the street or an animal in the forest and making a determined effort to find out what we see - is it actually conceiveable that people would be unable to find out if that animal is a rabbit or a duck?
A similar argument holds for illusions like the parallelogramm: The main issue is that we are able to find out that it is an illusion! Although, at first glance, we may say that the left diagonal line is longer, and a Zulu might disagree - imagine we both make a determine effort to find out and use tools like measuring sticks, or would cover the distracting other lines - is it really likely that we would still disagree about the line length? Not really. In everyday life, we trust to everyone perceiving the same situation to a degree that we share the roads with people from different cultures (provided they have a driver's license). If there were substantial differences in perception, such trust would be likely to kill us.
Finally, consider Hanson: His claim is plainly wrong. I am a physicist, and I am convinced that a heliocentric view works better - and yet I see the sun rising and moving across the sky like everyone else. I'm not special, I'm not aware of a single person (physicist or not) who is able to see a static sun and the earth turning. My interpretation of events does not alter my perception, I just know that I am subject to a perceptual illusion. Even after measuring the line length, I still see the left line in the parallelogramm longer than the right one. Nor is it historically correct that Kepler saw a static sun. From his writings, it is quite clear that both Kepler and Brahe were aware that whatever the interpretation, the perception would be the same.
The upshot of this investigation is that there is perceptual relativism, but it is actually comparatively mild, although it tends to be vastly exaggerated by some. It simply does not happen that massive disagreements of perception remain after a determined effort to cross-check perception and to augment it with tools. It is plainly not true that observations are so theory-laden that they merely confirm out prejudice and we see what we expect to see. One cannot invoke perceptual relativism to argue that there can be disagreements if a computer is connected with a power source or not, or if a light is on or not.
Relativity of conceptsAfter perception through the senses, we're left with something that could be called the scene, the experience of the moment now. The scene as it appears in the conscious mind is a whole, the unified sum of all sensual information. Concepts are then the way we structure and classify the scene, how we break it down into smaller parts, and concepts ultimately structure how we think about the situation and how we act. Beyond that, we may think of the situation as something which contains everything that could be experienced in the scene. For example, when we see a cup on the table, we could detatch it from the table and lift it, even if we don't try every time. The situation might be different from out idea about the situation - for example we may expect to be able to lift a cup, whereas it actually happens to be glued to the table. One might argue at this point that we do not know that cups are detatchable unless we try, and that hence it would be wrong to ascribe intrinsic properties to the scene unless we experience them. I'll leave it to the reader if he wants to go down that path and argue a philosophy in which coffee cups have no properties unless they are experienced. I'd make the guess however that this is not how even postmodern philosophers organize their lifes.
Central concepts are then those which are the basic building blocks of definitions. For example, in structuring a scene, we may perceive objects, persons, spaces and boundaries, and these would be central concepts. Other concepts would then be defined in terms of these, for example 'friendship' is usually seen and defined as a relation between persons. It would be somewhat weird to define a person as something that can (among other things) participate in friendships - and that shows that 'person' is more central than 'friendship'.
Normative conceptual relativism is then the philosophical thesis that there is no system of concepts which matches a corresponding structure in the world. In other words, there is no set of concepts which divides things into groups that corresponds to an order how things are divided up in reality as it is - some sets of concepts may be more convenient than others, but that is just convenient in relation to the conscious mind and does not correspond to how things are.
Hilary Putnam expresses this idea as follows:
'Objects' do not exist independently of conceptual schemes. We cut up the world into objects when we introduce one or another scheme of description. (...) If, as I maintain, 'objects' themselves are as much made as discovered, as much products of our conceptual invention as of the 'objective' factor in experience, the factor independent of our will, then of course objects intrinsically belong under certain labels because those labels are just the tools we use to construct a version of the world with such objects in the first place. (Hilary Putnam 1981)
To give a more concrete example, consider seeing a shelf with a book and a vase on it as situation. We of course divide the scene into different objects - we are aware that we could take the book or the vase from the shelf and place these objects somewhere else. But it may just be useful to think in this way about the scene for us, because we have hands and can manipulate objects. Imagine the same scene seen with the eyes of a cat: the cat is not interested in grabbing and displacing the book or the vase - instead in all likelihood the cat conceptualizes the scene in terms of a path across the shelf. So, for the cat, a book (or any object) may just be a secondary concept, to be defined in terms of the primary concepts of paths, and characterized by the way it obstructs or complicates a given type of path. For the cat, having paths as central concepts would just seem more useful than objects.
The relativity of concepts to languageA very prominent version of relativism of concepts claims that language is the essential factor which shapes thought. This is known as Sapir-Whorf Hypothesis. To quote a few passages by Whorf:
We are thus introduced to a new principle of relativity, which holds that all observers are not led by the same physical evidence to the same picture of the universe, unless their linguistic backgrounds are similar, or can in some way be calibrated. (...) The relativity of all conceptual systems, ours included, and their dependence upon language stand revealed. (...) We dissect nature along lines laid down by our native languages. The categories and types that we isolate from the world of phenomena we do not find there because they stare every observer in the face; on the contrary, the world is presented in a kaleidoscopic flux of impressions which has to be organized by our minds - and this means largely by the linguistic systems in our minds (...) no individual is free to describe nature with absolute impartiality but is constrained to certain modes of interpretation even while he thinks himself most free. (Benjamin Lee Whorf 1956)
A cornerstone in Whorf's claim is his 'investigation' of the Hopi language. Whorf concluded that the Hopi language is seen to contain no words, grammatical forms, construction or expressions or that refer directly to what we call 'time', or to past, present, or future and linked this to the way the Hopi conceptualize the world. The supposed timelessness of the Hopi language is extremely popular with writers, especially combined with claims about the predisposition of the Hopi to grasp the essentials of the physics of Special Relativity.
Of course, the latter statement should trigger some alarm bells with everyone who knows anything about Special Relativity. It is not a timeless theory, it does not argue a difference between space and time in terms that space can be measured but time cannot - quite the opposite. Special relativity relates space and time, and argues that there is no difference between them - whatever you can do with space also applies to time. But it turns out that there's more wrong with Whorf's claims about Hopi. Whorf had never learned Hopi or lived among the Hopi people - he gained his knowledge by interviewing a single individual. In contrast, Ekkehart Malotki has done an in-depth study of time concepts in Hopi by living among Hopi and learning the language ('Hopi Time' (1983)) and concluded that Hopi expresses time pretty much like every other language - it can be used to count days and it has similar metaphors for space and time measurements. So, some more spectacular claims of linguistic relativity are just plainly not true and can be traced back to faulty research methods.
Returning to the more general theme of this section, it is of course quite true that there's a connection between language and the way people think. For example, it has been demonstrated that the way a question is worded can change the response of people from a large majority for an issue into a large majority against the same unchanged issue. Euphemisms illustrate the issue fairly well - people tend to think and talk about a veterinarian 'putting a horse to sleep' rather than simply 'killing the horse', and although the action is not different, talking about it in a different way makes the thought more acceptable to the mind.
It can also be observed that there are languages which are able to express a concept which is characteristic for one society, but is not readily found in another. Consider for example Japanese: The Japanese language has an elaborate set of forms to express different levels of politeness. One would address an acquaintance different from a customer, but yet different if the customer is a learned man, and this is reflected not only in the title of address but also in verb forms. Japanese women often use different words from men to refer to the same thing. These grammatical and lexical distinctions correspond with a complicated social hierarchy which is characteristic of Japanese society. As a result, a Japanese text describing a man going to work and greeting his customers, friends, collegues and his dentist could not be adequately translated into an English text of comparable length and structure - English simply lacks the grammar to make the same distinctions. This is sometimes referred to as the translation problem. However, it would still be possible, using explanations and footnotes in the text, to let an English-speaking reader understand what happens in the Japanese text. It is not impossible to express the Japanese ideas of politeness in English, it is just not efficient.
However, all that is a far cry from claims that language 'constrains individuals to certain modes of interpretation' or that 'observers are not led by the same physical evidence to the same picture of the universe, unless their linguistic backgrounds are similar'. The link between language and concepts is not that solid.
Consider a few examples: The Finnish language has no future tense - and yet it would be a quite mistake to claim that the Finnish people are constrained to have no concept of future events - of course they do. In German, the word 'spoon' has male grammatical gender ('der Löffel') - and yet the Germans do not view spoons as particularly male symbols. People do not feel particularly constrained by the grammatical gender of words. The moon has male grammatical gender in German ('der Mond'), and yet the Neopagan community often feels that this should be female - thus they coined a female expression ('die Mondin'). All this would be quite impossible if language substantially constrains the way people think.
As for practical experiences, I've been doing physics with colleagues from e.g. Japan, Russia, China, South Africa, India or Brazil, i.e. non-native speakers of English. Although their linguistical background is not similar at all, they were all led to the same picture of the universe by the same physical evidence. There were no comprehension problems having to do with different concepts in practice.
Stepping back a step further, it is even remarkable how similar all natural languages are. True, at first glance, they appear anything but similar, but once one starts comparing with artificial languages it becomes readily apparent that natural languages represent a tiny fraction of what could be there. Consider Tolkien's Entish, which is a very slow language. When referring to a hill in Entish, one would recount the entire history of the hill, so it takes hours to even exchange greetings. There is no natural language like Entish. There is not even a natural language in which information would flow ten times slower than in English (or ten times faster for that matter). The speed at which natural languages transmit information is rather fixed - it is set by the rate the conscious mind can absorb and process information. Or consider computer languages: They allow to define a word to mean a long string of instructions, then supersede the global definition by a local one for one paragraph, and then come back to the global definition afterwards without note. Natural languages don't do anything like this (although it would be very efficient) - our memory simply isn't accurate enough to return to definitions that have been made an hour ago, but have been used differently since. So, this is quite different from thinking being relative to language - here we see that essential properties of natural languages are determined by the way the mind works!
The bottom line of linguistic relativity is probably something like the statement that language determines habits of thinking - but as habits can be broken, thinking can exceed the concepts present in language and use language to express thoughts in a complicated way, or modify language by coining new terms, or even by inventing a completely new language that is more suitable to express thoughts. Strong versions of linguistic relativity such as advocated by Whorf sometimes are therefore not true.
Relativity of concepts in scienceFor a scientist, the relativity of concepts is not a particularly new or disturbing idea, but rather something that is central to modelling. First, consider something which could be termed scale hierarchy of concepts. A surgeon may look at the human body and find it useful to conceptualize it in terms of organs - he finds the liver, the heart, the kidneys, and so on. A cell biologist would approach the same body stating that really cells are the most useful concept to describe what happens. And on a cellular level, the distinction in terms of organs are blurry - one cannot really say that this cell belongs to the liver and that to a blood vessel running through the liver. Nevertheless, on a scale larger than a cell, there is a useful ordering scheme of how cell types change corresponding to organs.
A biochemist would of course state that the most useful concepts are not cells, but rather biomolecules, enzymes, DNA and their reaction chains. On this scale, the distinction what molecule belongs to what cell becomes blurry and can't really made. Nevertheless, on a scale larger than molecules, cells are a useful ordering scheme to describe how different types of molecules are distributed. Further down in scale, a physicist would argue that biomolecules should really be described as bound states of atoms, and that atoms are composed of electrons, protons and neutrons, which in turn consist of the elemental quark and gluon fields, and that there is no way one could tell if a particular gluon is part of a proton or part of the vacuum fluctuations, but that on a scale much larger than that of a quark, protons, atoms and molecules are useful concepts to describe a particular ordering in the elementary fields. Scientists really select concepts according to the size scale of the phenomena they are interested in, and they are aware that a transition to a different scale usually also involves a transition to a different set of concepts.
But there is even more: In physics, it is quite common to exploit the freedom of cutting a situation in different ways to select the conceptual framework which makes the problem easiest. Consider for example the problem of computing the orbits of the planets around the sun. It is possible to do this by conceptualizing the problem as a set of rigid massive bodies with gravitational forces acting between them. Then one solves the equations of motion of each planet with the force of the other planets and the sun acting on it. But is it equally possible to derive the so-called action of the system, given by its Lagrangean function, which involves the gravitational potential rather than forces. The solution in this case is obtained by extremizing the action, i.e. by solving completely different equations for different concepts. Or consider problems in Quantum Mechanics - they can be formulated in Schrödinger's wave equation picture, in Heisenberg's Matrix picture, or in Feynman's path integral formalism, each of which involves a completely different set of concepts, equations and solution techniques. Yet they provide the same answer in terms of what is observable - so the physicist picks the concepts for which he can solve the problem easiest. If there is a wave function in Quantum Mechanics or not is a question of choice - but there is always something which can be described by a wave function - and if the description is not in terms of a wave function, then the essence of that something must be captured in a different way.
Relativity of realityThere are several claims by postmodern thinkers with regard to the relativity of reality itself. The idea of social construction of reality by Peter Berger and Thomas Luckmann (1966) for example is a rather prominent one, but there are numerous others:
The fact of the matter is that the 'real world' is to a large extent unconsciously built up on the language habits of the group. No two languages are ever sufficiently similar to be considered as representing the same social reality. The worlds in which different societies live are distinct worlds, not merely the same worlds with different labels attached. (Edward Sapir, 1929)
(...) the formulation in terms of 'comparison', in speaking of 'facts' or 'realities', easily tempts one into the absolutistic view according to which we are said to search for an absolute reality whose nature is assumed as fixed independently of the language chosen for its description. The answer to a question concerning reality however depends not only up that 'reality' or upon the facts, but also up on the structure (and the set of concepts) of the language used for the description. (Rudolf Carnap, 1949)
There is, I think, no theory independent way to reconstruct phrases like 'really there'; the notion of a match between the ontology of a theory and its 'real' counterpart in nature now seems to be illusive in principle. (Thomas Kuhn, 1970)
What characterizes these statements is a certain lack of clarity. Take for example Sapir - what he says is that the fact of the matter is that there are no real facts of the matter - that's a bit self-defeating. What Carnap says can have a spectacular and a trivial interpretation. The trivial one is that any answer to a question concerning reality must of course be given in a language, languages use concepts, so one needs to select a set of concepts if one wants to formulate any answer to anything - in other words, communication doesn't work in a language-independent or concept-independent way. However, that in itself doesn't imply that reality is concept dependent, only that its description is concept dependent, and it's not quite clear where the more spectacular claim that the 'absolutistic view' of an 'absolute reality' is incorrect comes from. As seen above, physicists regularly select different concepts to describe what they view as 'the same thing', based on the notion that the observable outcome is the same regardless of the chosen concepts. Kuhn states something fairly obvious, i.e. that since we never experience things without experiencing things, any statement of what is there whenever we don't experience it is a theory. But in practice that's not a problem, as we have ways to sort wrong theories from good and bad theories, hence we do not need to prove that we have a 'match' or the 'correct' theory. One doesn't really need to show that theory A is right to adopt it if it is vastly more successful than all its contenders.
The Anbaric Demon and objective experiencesTo illustrate the difference between what has been termed above the situation and the description of the situation in terms of concepts, let us introduce the Anbaric Demon. It is an experience that a computer needs to be plugged into a power source to work - if it is not connected to a pwer source, it does not work. We are used to think of this in terms of electric current, and on face value the experience would argue for the reality of electricity.
However - we don't know. Imagine someone believes in the Anbaric Demon which can be summoned into the computer and makes it work. To summon the Anbaric Demon, it is not necessary to make an incantation, but rather to draw it into the computer via a wire. More things are necessary - for example talismans need to arranged in the proper order (think of the fusebox...) and sacrifices need to be made (think of the electricity bill...). To the degree that the concept of the Anbaric Demon describes all the experiences for which we usually invoke the electric current, the concept of the Anbaric Demon is as real or valid as the electric current. Interestingly, no physicist would view electric current as something which is fundamental reality - instead a physicist would see it as a useful concept to describe quantum electrodynamics at large distances and discard the concept in a second as soon as he reaches a situation in which it is too simplistic.
Nevertheless, the equal validity of Anbaric Demon and electric current isn't an 'anything goes'. The concepts used are tightly constrained. One can not invoke an Anbaric Demon for which the wire must be connected once to the computer so that the Anbaric Demon can get in, but then the Demon resides in the computer and the cable can be pulled out again. This is contrary to experience. The situation in which the cable is plugged into the computer is different from the situation where it isn't - and regardless of what concepts are used to describe the situation, be it a path-oriented language in which the cable is seen as something obstructing a way in a particular way, be it the Anbaric Demon, be it electric current or be it quantum fields, the chosen concepts must reflect this.
No amount of social construction, discourse, language habits or choice of different concepts has ever led to a computer working without power source. Nor will it ever - regardless of what Sapir may claim when he refers to different worlds. These world are tightly constrained by something - which is reality. What this reality is not contained in the concepts, in 'thing' or 'space' or 'person'. Reality is much more elusive - it is contained in the properties which remain invariant when one changes from one system of concepts to the other.
Relativity of subjective, interactional and communicational realityThe example of the Anbaric Demon has referred to the objective quality of a situation. One might ask about the relativity of other classes of reality.
Communicational reality is obviously rather different. In essence, here the mutually agreed on description is the reality, there is no reality to communication except the concepts which are communicated. Ideas like 'natural right' or 'money' are in essence concepts, and because a large number of people share and acknowledge the validity of the concepts, they become real in the sense that they cause actual experiences, but as soon as one deals with a group who does not share the concepts, their reality becomes less obvious. Discourse or language habits might not make a computer work without power, but for sure they create laws and define the notion of a natural right. Social construction is certainly true for communicational reality.
It is less obvious what happens within subjective and interactional reality. We think in concepts - so in what way could these concepts be different from the reality of our mental state? In what way could we choose wrong concepts to conceptualize thought? The key insight is that the mental state is more than just verbalized thought, and that the way we think about our own mental state might be at odds with the way the mental state is. Consider the concept of (romantic) love. Typically, societies maintain that love is exclusive - one can be in love with one partner, one can eventually fall in love with another, and that implies that one is not in love with the first partner any more. Yet sometimes people experience that they fall in love with someone new without ceasing to love their present partner in any way. This often leads to a discrepancy between what one 'should feel' and what one 'feels' and thoughts about 'what the true feeling is' and 'whom one really loves'. And yet - the mental state of such a person would be better described by a non-exclusive concept of love and by acknowledging that the usual concept of love is not an objective reality. Thus, a socially constructed concept can be at odds with the individual mental state - i.e. subjective and interactional reality is not necessarily defined by social construction. People have the ability to rise above the conceptual system they are used to, and re-conceptualize for a better description.
Relativity of epistemic standardsThe term 'epistemic standards' refers to the set of criteria and techniques one uses to discriminate correct from incorrect statements. The need for this is given by the observation that things may turn out to be different from what one thought them to be, but they may also turn out just as one expected them to be, so there are ideas can be sorted a posteriori according to how well they anticipated events, but epistemic standards are also used to judge the value of ideas a priori, i.e. before an event takes place, given what is known at the moment. In other words, epistemic standards define what is a reasonable conclusion, what is rational what is a justied belief and what is a good technique to gain knowledge. Ultimately, the relativity of epistemic standards is connected with the relativity of truth. However, as the scientific method doesn't really make claims to truth, the relativity of truth will not be discusse prominently here.
There are various claims that such epistemic standards are dependent on society and change over time, naturally often brought against the scientific method as a toolkit to investigate the objective:
(...) specific canons of rationality are time-dependent (Larry Laudan, 1977)
Science is a highly elaborated set of conventions brought forth by one particular culture (our own) in the circumstances of one particular historical period; thus it is not, as the standard view would have it, a body of knowledge and testable conjecture concerning the real world. It is a discourse, devised by and for one specialized interpretive community, under terms created by the complex net of social circumstance, political opinion, economic incentive and ideological climate that constitutes the ineluctable human environment of the scientist. Thus, orthodox science is but one discursive community among the many that now exist and that have existed historically. Consequently its truth claims are irreducibly self-referential, in that they can be upheld only by appeal to the standards that define the scientific community and distinguish it from other social formations. (Paul Gross and Norman Levitt, 1998)
It is obviously true that different cultures historically had and continue to have different epistemic standards. Aristotele believed that deduction from self-evident basic premises would be sufficient - today we are much more cautious as to what is considered self-evident and why. Probability theory with its ideas of effect significance or correlations, which is central in the design of many present-day experiments and studies, did not occur prior to the 17th century. Before, experiments were often what we would consider today as There is a large number of people who are convinced that holy scriptures are a better source for knowledge than experiments. So, there is no question that the community of scientists actually is just one among many others that now exist and have existed historically.
However, it is not true that the standards upheld by the scientific community are peculiar for the scientific community, or that its truth claims are self-referential. Let's deal with the latter first: Scientific epistemic standards are justified by the ability to make successful predictions. If a society wants to build a bridge, a scientist can compute the stresses and possible maximal load and provide a suitable design - scientific method as far as it relates to problems of statics is then justified by the existence a stable bridge which everyone can experience. Philosophy, religion or other communities can make (and have historically made) no such claims with regard to problems of statics. If philosophy could predict bridge designs, it's truth claims with regard to statics would be as justified as those of science. If science could not predict bridge designs, it could not make any claims with regard to its validity for statics. To call this a self-referential claim is certainly misleading - bridges are used by many people outside the science community who can all testify that the bridge is stable.
With regard to the scientific community defining itself through a particular set of epistemic standards, it has been pointed out above that core principles of logical reasoning are present all across humanity in everyday life. Some of these principles would be life threatening to abandon. Take again the example of inductive inference: Imagine the last 100 people who have tried to pet a tiger have died. Inductive inference then argues that you should not try to pet a tiger if you see one and argues that it is reasonable to assume that if you try, the tiger will attack. A rational system in which inductive inference does not hold would argue that all tigers are different, and we don't really know anything about this one, regardless of what happened the last 100 times - so in the absence of such knowledge, it's as reasonable to pet the tiger as to run, so if you feel like it, you can just as well go ahead. Tigers (and other dangers) would quickly lead to the demise of a society based on such notions of rationality. Even postmodern philosophers take a jacket if they see snow outside - regardless of what they write, they don't actually think that just because it was cold the last hundred times when there was snow outside, there's no reason to believe it will be cold today.
It is, however, immediately obvious where the advantage of making claims to the relativity of epistemic standards lies for a thinker: He can claim immunity from falsification. Any scientist must face the possibility that he did insufficent research, a bad experiment, or made a mistake in deductions, none of which is pleasant. But by invoking the relativity of epistemic standards, one can always argue that one hasn't done anything wrong, it just appears wrong from a particular point of view, but this point of view is really just one opinion among others. Consider the comparison between Whorf's and Malotki's study of the Hopi language above: Strong epistemic relativity says that Whorf's study is really as valid as Malotki's and that they both are just different narrative constructions. But in practice it makes all the difference if I interview a Chinese speaker for a weekend and claim to understand Chinese, or if I actually learn the language. In the first case, I am in for a nasty surprise should I ever travel to China. Such a 'reality check' immediately reveals the superiority of one method over the other. It's perhaps not surprising that the same community which continues to hold Whorf's study in high esteem advocates ideas of science as narrative construction - it's precisely what this community does, and so they think everyone else does the same. More often than not, claims to epistemic relativity appear little more than a smokescreen to avoid critical assessment of ideas.
SummaryFrom the discussion of the various branches of relativism above, one can perhaps see a pattern. Reality usually is relative to a framework if it happens to be of the communicational type, but it is far less mallable if it is of the objective type, i.e. the way we talk about things is very dependent on cultural context (we might talk about electricity or the anbaric demon) but the underlying reality is not (both electricity and the anbaric demon require the power chord to be plugged in for a device to work). Thus, it is correct to say that there is no such thing as 'electric current' outside a particular framework of thought since 'electric current' is just a way to talk about a phenomenon - however there are properties which won't change no matter what other words we use to talk about the same thing.
Most of the more spectacular claims of postmodern philosophy are illustrated and made plausible for communicational reality and then assumed to transfer to objective reality, in the process usually equating the phenomenon itself to some degree with the way we talk about it. That, however, is a grave reasoning error, as there is no reason to assume this transfer is possible. While for a sociologist it may perhaps seem that physicists believe in the existence of, say, forces, the physicists themselves know quite well that they can re-formulate the problem such that no forces appear and yet the outcome of a calculation remains the same observable, i.e. that forces are merely a useful tool to describe reality, not the thing itself.
Distinguishing between the various shades of meaning of 'reality' and between the various branches of relativistic claims is, as seen above, crucial if one aims to understand the value of the relativistic ideas of postmodern philosophy.
Appendix: A collection of red herringsLet us conclude with a commented list of red herrings often used in debates involving postmodern philosophy arguments:
However, while it is in principle correct that lack of complete certainty implies that other things can happen, in practice the question is a quantitative one rather than a qualitative one. It becomes a problem of estimating likelihood, and the reasonable strategy is to believe in and to act on the most probable ones of all possible alternatives. That is what people actually do in their life. Consider the following example: It is certainly possible that a meteorite destroys your workplace while you sleep. It is just an exceedingly unlikely event. If this happens, there is no point in getting up and going to work in the morning. However, no one, not even die-hard postmodern philosophers, acts on this possibility - they all get up and go to work. This is because it is not reasonable to believe in something or to act on something just because it is possible - it rather needs to be, all things considered, about as likely as the alternatives. For example, it is reasonable to take a raincoat to work even if the weather forecast predicted sunshine, because the forecast has a probability of being wrong of the order of 10-20% - and rain is therefore still a probable outcome, albeit not the most probable one.
Applied to frameworks like scientific method, it is not sufficient to show that the scientific explanation of something cannot be proven with certainty to argue for a different explanation. It must be shown that the different explanation is, all experiments and observables considered, successful in accounting for everything on a level comparable with the scientific explanation.
The consequence is that there are many explanations of phenomena about which we cannot be conclusively sure, but we can be so certain in practice about their correctness that we can discard the possible alternatives for good. We speak of such things as if they were certain, and this is reasonable: No one factors the tiny possibility of his workplace being hit by a meteorite into his plans how to spend the next week and has the alternative schedule worked out should this actually happen. We are beings with a final mental capacity, and we need to make such approximations to get anything done, and treating exceedingly unlikely events in practice as impossible events and exceedingly unlikely explanations as wrong is certainly a good approximation. Anyone who argues otherwise must address the question why he argues differently from what he really does in his life.
But this means that strong claims like 'reality is relative to society' backfire: Either there is objective reality to what 'society' is, then not all reality can be relative, or there is no objective reality to 'society', then the claim doesn't really mean anything if neither 'reality' nor 'society' can be defined beyond an individual's idea of what they are - and this last leads again to Solipsism.
More broadly spoken, if there really are no concepts, beliefs or modes of reasoning, then one cannot claim that people differ with respect to their concepts, beliefs or modes of reasoning. And if speaking different languages or belonging to different cultures leads to different modes of thought, then there must be objective causal connections between speaking a particular language or belonging to a particular culture, on the one hand, and how one thinks, on the other.
Back to main index Back to essays