Download This Chapter As A PDF


Please ensure you have
the latest Adobe Reader

Knowledge in the social sciences

Issues of knowledge in the social sciences are widely canvassed in every discussion of meta-theory. It is often supposed that Foucault has established that there can be no one truth about reality since every “truth” is defined within a particular socially constructed “regime of truth”. This kind of perspective is sometimes referred to as “relativism” because all perspectives are relative to the position of the speaker.  According to this use of Foucault, in every society a “regime of truth” operates to declare certain things true and to define a range of acceptable methods to discover the truth – and these regimes of truth always operate to back up social power.

Social science does not escape this general account. Foucault’s suggestion is that in modern societies, knowledge of society is produced to gain power – there is no knowledge without power and vice versa. Consequently we should be looking at the way power constrains us to find the truth and act according to the truth we uncover.  According to Foucault, uncovering the truth is not always a radical strategy. Foucault repudiates Marx’s idea that true knowledge is always on the side of the oppressed – while oppressors hide the truth in ideology. 

There is some of this that is very salutary, especially in terms of political strategy.  While it may be good politics, writers can draw implications for meta-theory that are a problem. Before I go on to discuss this, I want to present a radically different view of the acquisition of knowledge from that which is common sense in social science writing. This view takes it that there is no essential difference between our everyday perceptions and the knowledge that comes through these perceptions and our attempts to uncover the truth through social sciences, with all their array of abstracted theoretical concepts. In so far as there is any difference at all, it is in our degree of certainty, the extent to which we feel confident that we are really getting at the truth. 

It is not that we are dealing with a radically different way of knowing, appropriate to a radically different set of objects – social objects. Nor is it the case that in the social sciences we are dealing with a radically different kind of knowing because of the social factors that impact on our knowledge creation. It is just not true that in the social sciences our knowledge is influenced by social context or socially created theoretical constructions – and in our everyday perceptions and the natural sciences we are not so influenced. Nor does the fact of this social influence and theoretical context make it a vain exercise to attempt to uncover “the truth” about society; or about anything else for that matter. 

The kind of position I am taking can be called “realism” because it argues that there is a social reality and social science is an attempt to understand the nature of that social reality.

In fact, as we shall see, the arguments against true knowledge in the social sciences are the same arguments that have been used by philosophers for centuries to call into question our everyday knowledge and perceptions. I will make use of Armstrong’s discussion of theories of perception to get this going (1961) and then move on to analogous arguments in the social sciences.

Our naïve view of perception

In our naïve pre-philosophical condition we may think that when we perceive we have direct access to a real thing in the physical world. If we see a tree it is a physical object that we are looking at. Or “the object of awareness is never anything but a physical existent which exists independently of the awareness of it” (Armstrong 1961: xi). This position, which Armstrong refers to as “direct realism” is the starting point.  Other theories of perception come about as we make critiques of this view.

Attacks on the naïve view – the argument from illusion

The argument works through three stages.

1. There is sometimes a difference between what I may perceive and what someone else may perceive. Or I may perceive something at first and later find out that I was mistaken – what I thought was a cat dashing out of the room was a shadow moving across the wall. Berkeley, an English philosopher of the eighteenth century, gives as an example a case where you put a cold hand and a warm hand into the same “vessel of water in an intermediate state; will not the water seem cold to one hand, and warm to the other?” (Berkeley 1965: 142).  Here, there is an incompatibility between your perceptions, so how could you be perceiving the same thing with each hand.

2. According to the argument from illusion, what is going on in these cases is that we are actually not perceiving physical objects at all. What we are directly perceiving and are not mistaken about, are sense impressions. If you think you see a cat running from the room what you are really seeing is a sense impression of a cat. 

3. The argument from illusion is then nailed down. It is not just when we are mistaken that we are really perceiving sense impressions. This is all that we perceive all the time. A sense impression, which is really an idea of the observer’s, is the direct object of perception. 

Berkeley presents all three stages of this argument in this example:

In the jaundice everyone knows that all things seem yellow. Is it not therefore probable that those animals in whose eyes we discern a very different texture from that of ours, and whose bodies abound with different humours, do not see the same colours in every object that we do? From which, should it not seem to follow that all colours are equally apparent, and that none of those which we perceive are really inherent in any outward objects? (Berkeley 1965, 149).

Berkeley is hardly the only philosopher to have made this argument. Husserl gives as an example the way a table looks different from different angles. These different perspectives are incompatible. Consequently what we actually perceive is never a table itself but merely a set of perspectives. The physical object is then “transcendentally” perceived or constructed by our mind from these various perspectives (Husserl 1931).

If you buy this argument from illusion you may be in the market for two theories of perception that are supposed to offer a better deal than direct realism. I will describe these and then talk about some of their problems.

Representative realism

In representative realism, what we immediately perceive is sense impressions or ideas in our heads. However lying under each of these sense impressions is a physical object which in some way or another gives rise to our sense impressions. 

This seems like a very common sense view today when we are aware of light waves and the like striking our sensory organs and electrical charges passing up our nervous system to the brain. So it is like there is a little observer there in the brain looking at these images as they come in. It is these images or sense impressions that themselves provide evidence for the existence of the physical objects that spark them up. 

Armstrong presents a number of arguments against this view but I will stick to just one. We end up having no reasons to believe in the existence of the physical objects supposed to underlie these sense impressions. 

What are the reasons we may normally have for thinking there is a physical object present that we do not immediately perceive? For example, we are asleep and wake up to the sound of a train. We think there is a train out of sight down the end of the paddock because on other occasions when we have heard this distinctive noise, we have had immediate perception of the train through our eyes as well as our ears. 

But if all we ever experience is sense impressions; what previous experience of a link between physical objects and sense impressions could we have? Our seeing and hearing of the train in the past were only linked sense impressions – neither of them reveals an object lying behind them. We could never check up to see whether a physical object was behind our sense impressions because all we would ever be able to come up with would be another sense impression. Nor could we have any way of knowing what physical objects were like, because they would never be perceived – only sense impressions are perceived. 

Berkeley himself comes up with the knock down argument against representative realism as he moves into an even more extreme view of perception:

In short, if there were external bodies, it is impossible we should ever come to know it; and if there were not, we might have the same reasons to think there were that we have now. Suppose, what no one can deny possible – an intelligence, without the help of external bodies, to be affected with the same train of sensations or ideas that you are, imprinted in the same order and with vividness in his mind. I ask whether that intelligence hath not all the reason to believe the existence of Corporeal Substances, represented by his ideas, and exciting them in his mind, that you can possibly have for believing the same thing? (Berkeley 1965: 68)

Phenomenalism

In the social sciences the view that is analogous to “phenomenalism” in the philosophy of perception is described usually as “extreme relativism”, but sometimes goes under other names such as phenomenology or postmodernism.  But more of that later. 

According to phenomenalism, all that exists are sense impressions or at most sense impressions and the possibility of having them. We talk about a physical object being present when a certain group of sense impressions takes place – or when a certain group of sense impressions would happen if we acted in a certain way. So if I say there is a table in the kitchen it means that if I walked into the kitchen I would have table sensations. There is nothing more to it. A consequence is that the world of physical objects cannot be independent of the minds that perceive them. There is no such thing as a physical object other than a set of mental events.

As the reader is probably not surprised to hear, Berkeley had a word for it:

… all the choir of heaven and furniture of the earth, in a word all those bodies that compose this might frame of the world, have not any substance without a mind; that their being is to be perceived and known; that consequently so long as they are not actually perceived by me, or do not exist in my mind, or that of any other created spirit, they must either have no existence at all, or else subsist in the mind of some Eternal Spirit; it being perfectly unintelligible and involving all the absurdity of an abstraction, to attribute to any single part of them an existence independent of a spirit. (Berkeley 1965: 63)

So is there a problem with this conclusion? Well, one is that you end up inhabiting a world all of your own. The mind of the speaker is that on which everything else depends for its existence – the minds of others are mere constructions from our own sense impressions.

Also there are some counterintuitive conclusions about the material world. Without a mind to observe it; the material world cannot exist – because it is just a set of sense impressions and the possibility of having them. So life on earth could not have begun before animals developed consciousness. There are problems about space and time.  How are my sense impressions related to your sense impressions? We cannot say that one of us is a metre north of the other one. This already supposes we are objects in space. What about time?  How can two showings of a film be separated in time, if the same sense impressions take place?

Armstrong’s solution

None of this seems a very attractive way of dealing with the problem of illusion.  What if we assume that perception is the acquisition of a belief about the physical world? 

So, some apparent contradictions in the perception of objects are not really illusions but merely different ways of acquiring beliefs about the same object. We know that when we walk around a table we get different views of it and this is part of an unconscious theory of tables that informs our perceptions of tables. We accept different views of a table as being views of the same table because we have a theory about tables and what happens to our view of them as we walk about. Yet each view of the table contributes to and is itself a process of acquiring a belief that a table exists in the place we are looking at. Or even a stick that appears bent as it enters the water.  We know the stick is still straight; it just looks like a bent stick as it enters the water.  We do not acquire a belief that the stick is bent but merely acknowledge that it looks like a bent stick in this circumstance.

In other cases there is merely a mistake. We acquire a mistaken belief about the physical world. We think the shadow is a cat moving across the room. We think we have read a certain set of words. We acquire this belief and then check it against other beliefs to decide this set of words is unlikely. We check and note our mistake.  But the first false perception did not feel any different from a true perception in the first instance. 

In this view, perception is always a direct relation between the perceiver and the object, both of which are independent. 

An objection to this view is that our impressions of things do seem to be qualitatively unique; a phenomenon of consciousness that is a bit different from the acquisition of other beliefs. Armstrong deals with this by saying that our consciousness is an awareness that we are acquiring these beliefs – this gives perception its vividness.  Other philosophers talk about these phenomena of consciousness as “qualia” – they  have a certain “whatness”!   Well, whatever, the basic idea that perception is the acquisition of beliefs and that illusions are merely acquisitions of false beliefs is what is useful in Armstrong’s take on this. 

A key consequence of this perspective is that perception is never “incorrigible”; the belief we acquire when we first perceive something can later be corrected. What you always perceive is a physical reality and you can always be wrong about it. By contrast, representative realists and phenomenalists have a reassuring sense of certainty that at least our perceptions of sense impressions can never be wrong. But as we have seen, that reassurance comes at a price. 

Why this theory helps the social sciences

The main hidden virtue of this theory for us in the social sciences is that it completely does away with the gap between our everyday perceptions of the world and the kinds of theory that we use in the social sciences to account for things. Both perception and making theories are attempts to acquire knowledge about the world. Both can be wrong. 

What is more important, everyday perception of the physical world is mediated by theoretical knowledge. The beliefs we acquire through our senses are formed to make sense of what our senses are giving us. This process of making sense relies on other information and ideas that we have developed through our experience. In a way, all perception is theoretical. We acquire beliefs about the world in relation to theories about the world that we have already developed. The grounds of any belief that we acquire when we perceive are not just a set of firing synapses occasioned in that instant by the presence of a physical object. The grounds of perceptual beliefs are in fact “endlessly deferred” through a raft of connected beliefs. 

A recent example from my own life. I was in a hurry to get to class. I went down two flights of stairs and as I was about to exit the building I remembered I had forgotten my course guides. It was week one and I was worried the students might leave the tutorial if I was late. I rushed up the stairs and automatically turned right and started to enter the hall that I thought would lead to my room. Instead the corridor in front of me seemed like something out of science fiction. I felt as though I had walked into a time warp or another dimension. The corridor seemed very long and forbidding, with a set of strange pictures hanging on the walls. In fact I had only gone up one flight of stairs and was on the floor below my room – a corridor I have often been in. My theory that I was on my own floor created a perception of illusion and strangeness even though there was nothing odd about what I was actually looking at. 

A very common move in the social sciences is to make a distinction between everyday facts, which we perceive with our senses and the kinds of theory, which social scientists develop. The first are supposedly common sense and imply a common physical reality that we all inhabit. The second are influenced by theories and ideas that we receive from society; they are fallible and can be so different as to imply there is no social reality as such – just different perspectives. But what the above shows is that our theories and pre-existing beliefs are just as relevant to everyday perception as they are to our social theories.

How do we come to acquire beliefs about the physical world – if that is what perception is? There is no answer to this question that does not make use of theories about the physical world that we have developed. There is no “general” account of such things. Today, in terms of our current understanding of physical reality, we talk about sound waves and molecules of gases bouncing off each other and so on. The same with the social sciences. We have theories about the world which are theories about how we come to have theories of the world. 

How sociologists repeat these accounts of perception

Sociologists are concerned with the way whole groups of people view the world.  They are also concerned with particular kinds of objects that the social scientist and others can have views about – social objects. For example, things like people, social classes, social structures, discourses, subjectivities, identities, subcultures. In the social sciences, there are theories that resemble representative realism and phenomenalism in theories of perception. Just as those theories of perception tend to think that objects in the world boil down to our perceptions of them; analogous theories in the social sciences treat societal objects as having a less than objective reality. We can find all these arguments in Mannheim’s work and more recent discussions just repeat these arguments; sometimes with an annoying implication that these wonderful new insights are unique to our age. 

Naïve Realism

Durkheim certainly represents a naïve realist position in the social sciences, and his unsophisticated empiricism could certainly put you off realism. Durkheim maintains that the social scientist must be on their guard against mistaking their beliefs for reality itself. He (sic) “must emancipate himself from the fallacious ideas that dominate the mind of the layman” (Durkheim 1964: 32).  He should study social phenomena “objectively as external things” (Durkheim 1964: 28). This is because social facts “are things by the same right as material things” though they may differ from them in type. 

My main problem with this is the idea that we can emancipate ourselves from fallacious ideas. The ideas of the social world that we come up with are necessarily influenced by a great range of ideas that come from our accumulated experience – just like the ideas about the physical world that we come to acquire through perception.  We cannot abstract ourselves from our experiences to do science of any kind or even to observe a sunset.

On the other hand I will agree with Durkheim’s view that social phenomena are things and are objective in just the same way as material things. One way of putting this is to say that they are things and their existence does not depend on whether we observe them or not. 

This is very controversial in the social sciences. One view is that because we affect social situations when we observe them; they have no reality independent of our observations. I will consider this again later but to begin with a few points. Whether our noticing of social phenomena causes them to change depends on whether we influence a social situation through observing it. We may and we may not. This is very case by case. If we are not doing meta-theory, we do not have any trouble recognising that our influence on what we observe depends on the situation and the kind of observation we are talking about. 

Also, we usually have no hesitation in talking about what may be happening in social situations that we are not observing; and describing events that are taking place or have taken place, even though we know full well that we can be having no influence on them. For example what happened in Ancient Greece. We constantly make judgements about the nature of social realities that we are not observing; making inferences based on what we have heard from third parties, making an educated guess based on our theories of how things work in that social situation and so on. All this implies that we think social phenomena are things that are in many cases independent of our observations. 

In cases where we believe that our observation does change the social situation; then that is a fact about the world too. If I arrive in a pub to observe local culture and I am the only person there who is a stranger, my presence may end up influencing what takes place. However this is itself a fact about the world. We can think of an analogy in our observations of the physical world. A billiard ball goes into a pocket because we strike it with a billiard cue. No less a fact because we caused this event to happen.  Our causal influence on the billiard ball does not lead us to conclude that its going into a pocket did not objectively happen; since we had a role in this causal sequence!  We make a distinction between our influence on the event that took place and our observations of that event. No less the case for the impact of our social presence on the things we are observing in the social world. Our observation (as it takes place as a sequence of social events) has an impact and we observe that impact.

And of course, we do not just look at social objects (such as people; social classes; states; interest groups; discourses) as Durkheim suggests, but also at relationships (A coerces B, C influences D, E exploits F, G loves H) and events (the signing of the Magna Carta; the Second World War). Then there are those most reviled targets of social analysis – “powers” (libido; attachment; hunger; wealth; hegemonic masculinity; class interests; social power; ideologies). All these are no less “realities” because they are not objects. More on all these later.

Mannheim’s argument from illusion

Mannheim begins by arguing that for the sociology of knowledge it is “not men (sic) in general who think, or even isolated individuals who do the thinking, but men in certain groups who have developed a particular style of thought in an endless series of responses to certain typical situations characterising their common position” (Mannheim 1968: 2).  He goes on to make two arguments to show that knowledge in sociology cannot be knowledge of social realities – it must be knowledge of the different perspectives on society – of “ideologies”.  The first of these is the argument from illusion that we are now familiar with and the second is based in the specific qualities of social knowledge.

Mannheim presents his argument in three stages, corresponding roughly to stages in history. In our naïve view of the world, we believe that our own view of society is true and nothing in our social experience leads us to question that. However, as we become more au fait with different cultures and with the sweep of historical ideas, we discover the “alarming fact that the same world can appear differently to different observers” (Mannheim 1968: 5). 

Our initial response to this situation is to decide that our own views are true but that the false views of others can be explained by their situation and the interests, which develop out of that:

One does not call his own position into question but regards it as absolute, while interpreting his opponent’s ideas as a mere function of the position they occupy (Mannheim 1968: 68)

This is Mannheim’s interpretation of Marx’s position in the German Ideology. In my way of looking at it, Marx, like other empiricists tries to remove the possibility of his own position being tainted by interest by engaging in a fool proof scientific method, that supposedly guarantees the truth of his findings. He sticks to material realities and what can be known about them through rigorous historical research. However the ideas of the ruling class are revealed to be ideas based in their interests – they are ideological. Those ideas, unlike Marx’s own, are a function of the social position the ruling class occupies.

What is crucial about Mannheim’s reading of Marx is that he interprets Marx as saying that these ideas of the ruling class are false because they are dictated by class interest. Mannheim says that in this view the “validity of the adversary’s theories is undermined by showing that they are merely a function of the generally prevailing social situation” rather than taking them “at their face value” (Mannheim 1968; 66, 49). I will go onto this later, but I think that in fact there are other ways to read what Marx is saying.

In the third stage of our understanding, according to Mannheim – that which we have now reached – we realize that this analysis must apply to our own ideas as well. Even ideas that we have regarded as established by science are found to be constructed in specific social situations in relation to social interests coming out of those situations.  This is what Mannheim calls the “general form of the total conception of ideology” (Mannheim 1968: 69).  It comes about when the researcher “has the courage to subject not just his adversary’s point of view but all points of view, including his own, to ideological analysis” (Mannheim 1968: 69). We discover that the thought of every group arises out of its life conditions. We cannot see other views as false and our own as true, but must recognize a multiplicity of perspectives, all of which have an equal claim to validity. All views of society must fall short of absolute truth because they are all “ideological”, coming out of a social situation and representing some type of social interest. 

This third stage immediately leads Mannheim into problems analogous to those of phenomenalism in the theory of perception. We end up isolated, unable to communicate with other people. It becomes difficult to even explain how we could communicate with others and share ideas or gain understanding of the ideas of other people and times. As Mannheim says, “the very principles, in the light of which knowledge is to be criticized, are themselves found to be socially and historically conditioned. Hence their application appears to be limited to given historical periods and the particular types of knowledge then prevalent” (Mannheim 1968: 259). We “begin to suspect that each group seems to move in a separate and distinct world of ideas” (Mannheim 1968: 88). 

Mannheim’s solutions

Mannheim has a number of solutions to these problems of the argument from illusion and I will briefly consider these before going back to that argument. 

Often Mannheim writes about this as though he takes a representative realist view of social phenomena. We can work with the different ideological perspectives and see through them to a social reality. We can suspect that these “different systems of thought, which are often in conflict with one another, may in the last analysis be reduced to different modes of experiencing the ‘same’ reality” (Mannheim 1968: 88).  And later he says; “It is not intended to assert that objects do not exist or that reliance upon observation is useless and futile but rather that the answers we get to the questions we put to the subject matter are, in certain cases, in the nature of things, possible only within the limits of the observer’s perspective” (Mannheim 1968: 269-270). 

But how can we use the points of view to which we do have access – ideologies which must all be false – to infer an underlying reality? According to Mannheim we are lucky that there is a specific group in society (the intellectuals) whose social situation is such as to help them to see beyond ideologies. What leads intellectuals to an understanding of society that is more profound than that found in past epochs is the current period of class conflict. As the contending classes clash, they throw up different ideas about what is going on. These ideas are of course responses to their differing class interests. However each of these opposing views may be seen as revealing the limitations of the other position:

It seems inherent in the historical process itself that the narrowness and limitations which restrict one point of view tend to be corrected by clashing with the opposite points of view. (Mannheim 1968: 72)

The class position of the intellectuals is on the sidelines of this match. They can see both sides and cancel the biases of each. A group:

… whose class position is more or less definitely fixed already has its political point of view decided for it. Where this is not so, as with intellectuals, there is a wider area of choice and a corresponding need for total orientation and synthesis. (Mannheim 1968: 143)

This is very flattering but seems a bit too good to be true. Surely intellectuals have their own class interests; a position explained in great detail in Bourdieu’s Distinction and the Ehrenreichs’ writings about the professional managerial class. Intellectuals are hardly unbiased witnesses to social conflict.

More commonly, what Mannheim proposes as a solution fits with phenomenalism in theories of perception. We cannot transcend ideology – that is all we have access to.  So what we have to do is make some kind of a synthesis of ideologies. Where all views are comprehended in roughly the same framework (the naïve pre-sociological understanding) then errors are considered to be “everything that deviates from this unanimity” (Mannheim 1968: 220). However, where there is awareness of conflicting frameworks, the social scientist has to make an effort to “discover a common denominator” for the varying perspectival insights. When we find this common denominator, we are able to separate this common element from the “arbitrarily conceived and mistaken elements” and attempt a “synthesis from the most comprehensive and most progressive point of view” (Mannheim 1968: 270, 168). 

Yet all this contradicts his own premises. There is no standpoint outside history where we can determine what is a comprehensive point of view. Or a “progressive” point of view. To survey ideologies is just another attempt to discover the nature of reality and hence it is just as much affected by our own unique socially constructed viewpoint as any other exercise in understanding society. If these other attempts to understand society are compromised by the influences we receive from our background position in history; then so is this effort at synthesis. 

Reviewing Mannheim’s position

We can approach Mannheim’s position by questioning the argument from illusion.  Baldly speaking, we are in exactly the same position that we find ourselves in with illusory perceptions. 

The fact that there are a range of contradictory positions and different ways of understanding society can be reconciled with a realist view of the social world in one of three ways.  Either

• these viewpoints can be reconciled to a degree and seen as different takes on the same reality

or

• some of them at least, and maybe all of them, are false

or

• maybe there is much about society that we are not sure about, so it makes sense to have a number of different and contradictory perspectives.

Secondly, we can take it that our theory of the social location and construction of viewpoints on society is part of what we think we have discovered about society. At least, this is part of our current picture of the way society really is.  

Going back to Mannheim’s argument, the crucial mistake in the argument is this step.  The “validity of the adversary’s theories is undermined by showing that they are merely a function of the generally prevailing social situation” (Mannheim 1968: 66).  The validity of someone’s theories cannot be undermined by showing that these theories suit their interests. An everyday example. We tend to assume that a parent’s views of their children are positively biased. This is an understanding of the world based on experience. Yet we do not assume that everything a parent says about their child is false. If we think they are wrong about their child, we give reasons for our own view – we do not merely point to the fact that a parent said it. Oh, well that cannot be true, their parents think that.

Whether or not a person makes a true statement is of course causally dependent on their situation. When realists say that the truth of a statement and its genesis are independent, this obvious fact is not being denied. What is claimed is that what makes something true or false is some state of the world and not some state of the person who at any particular time asserts it.

These considerations can also be applied to social perspectives. How someone comes to acquire their views about society is one question about social reality. The truth or falsity of those views is another question. There is no general answer to the question – why do we think our own views are correct when they could be influenced by our social situation to be false? We could explain what we think the social reality is and what we think is good evidence for those ideas. We could also look at what it was in our own history that led us to take the path of making those investigations and coming to those conclusions. These histories and this evidence are open to the inspection of others whose own history and investigations could lead them to quite different conclusions. But what we do share is an attempt to understand an objective social reality that is out there in the social world. 

Since Mannheim sees Marxists as making this mistaken step, I will briefly outline another reading of the German Ideology. Marx presents it as a fact that has been discovered about social reality that “men (sic), developing their material intercourse, alter, along with this their real existence, their thinking and the products of their thinking” (Marx & Engels 1942: 14). It is this fact about the real world that makes “morality, religion, metaphysics and all the rest of ideology false” (Marx & Engels 1942: 14).  How I read this is as follows. What is false about ideologies in general is not that they are influenced by people’s material situation. What is false about them is that they claim that thought is causally independent of people’s material life and works in a real causal world of its own. The fault common to them is to think that ideas are more independent of material life than they are in fact. 

Marx then goes on to argue that his own method ensures that he will not make similar errors and that he will present a true picture about social reality. “The premises from which we begin are not arbitrary ones, not dogmas, but real premises … They are the real individuals, their activity and the material conditions under which they live … These premises can thus be verified in a purely empirical way”  (Marx & Engels 1942: 6-7). Using this method, the next step is to show empirically the ways in which ruling class ideologies fit the social interests of ruling classes. 

So his argument can be seen as presented in three stages:

1. Explain the method by which Marx will investigate history and provide evidence.
2. Using this method to investigate social life, discover what is false about ideologies – they think thought is independent of life.
3. Using this method, show that as a matter of fact, ideologies serve the interests of ruling classes.

This excursion is not meant to prove that Marx is absolutely right about any of this.  My own view is that in these passages he wildly underestimates the power of ideas and the arbitrary inventiveness of their creation. He overestimates the ability of his method to establish conclusions, which all those using the method will agree to. But what I am endorsing is his general strategy for social science. He gives us an example of how we can handle the discovery that the world of ideas can be socially constructed. How we can come to know this and go on believing that there is a social reality out there to be discovered?

A niggling problem with Mannheim’s position is the way it undermines its own premises. It tells us as a fact about social reality that there is a conflict between social classes. It tells us as a fact about society that these different classes end up with different viewpoints corresponding to their interests. And then it concludes by telling us that there can be no way of knowing what society is actually like. There is just a range of different understandings of society. So how could we possibly know that it is a social fact that there is a conflict between social classes?

I think that in the social sciences, a key problem with the realist perspective is that it is uncomfortably linked to Western imperialism. The enlightenment develops science to dispel the myths of religion. Western powers go out to conquer the world and to consider the beliefs of other cultures as “mythologies”. It is difficult not to be implicated in this.  

For me the most uncomfortable example is the Dream Time accounts of the creation of rivers and mountains presented by Indigenous Australians. I have heard Indigenous speakers present these accounts as the literal truth. Yet to me, they contradict the accounts of the creation of mountains and rivers that are given in geological science. It is nerve wracking and it feels racist to say that the Dream Time accounts are just false. Yet within a realist framework I cannot merely say that these are just different perspectives, each of which is formed out of the social influences that have impacted on the Indigenous speakers and myself, respectively. This is certainly true but it does not really let me off the hook. As a realist, I have to end up by saying that my own view is that the geological account is the correct one and that the Dream Time account is mythology. Of course I could be wrong and no perspective is infallible. I might well think that the Dream Time account of things made sense within and backed up a social system that had many virtues that Western society has never equalled. I might also understand that as a metaphorical system it expresses a lot of very accurate insights about the links between Indigenous society and environment that all of us need to take very seriously at the present time. I might well understand that making these claims today makes sense for Indigenous leaders as a response to racism. But I cannot honestly claim to share this view as a picture of reality. 

Clearly many radical social scientists are more comfortable with a relativist view of social knowledge because it does not get you into this uncomfortable political position. 

An alternative perspective is to say that relativism about these matters is actually patronising. Social scientists put themselves in the position of Mannheim’s “intellectuals” – the ones who are hors de combat and review a range of views and appreciate them all as perspectives. But of course, some of those whose views are being appreciated in that way do not themselves adopt a relativist position. They claim that their view of things is a true picture of the one social reality. In this case, social science intellectuals end up by presenting themselves as the tolerant citizens of the world – while they see others as understandable in their attachments to parochial and intolerant demands to claim true knowledge of reality. This is very tacky. One can hear the clink of the cups as the tea and scones are brought round. It could be more respectful to acknowledge all these perspectives as attempts to understand social reality and to admit that one’s own understanding of that social reality is different from that of other people (Hammond 2007). 

I find this a difficult moment in theory and I worry about the company I keep. I share my position with clunky and unreconstructed Marxists who are antipathetic to feminism and the environment movement, or with born again western enlightenment rationalists, who seem to have no real understanding of the social roots of religious fundamentalism. Perhaps this is one of those occasions on which Foucault’s warnings against always wanting to tell the truth make sense. However wrong relativism is as social science meta-theory, it could be more useful in fighting the establishment. Or at least that is one way of looking at it and I can understand why people may take that view. Bring round the tea and scones instantly. I invite readers to take their pick from these options.

The direct realist account of social knowledge

In the direct realist account there is no great gap between “facts” about social reality that we can all agree upon and theories about society, where we may disagree. Both statements about social facts and statements of theory are attempts to describe social reality. A study of the incidence of HIV transmission through injection is no more a statement about social reality than Bourdieu’s analysis of distinction.

Just as in our perceptions of the physical world, our beliefs about social reality can always be false. There is no way of making absolutely certain that one is not making a mistake, just as there is no way of being absolutely certain that you are not suffering from a hallucination when you perceive the physical world. But there are certainly some things which we are more certain about.