Abstract
Social Contexts are specific types of recognised social situation for which specific norms, habits, rules, etc. are developed over time. The unconscious and embedded nature of these make them difficult to change – becoming deeply entrenched over time. How cultures relate can be effected, in detail, on whether contexts in one culture are identified with ones in the other, carrying along with these engrained assumptions and expectations. This chapter explores the implications of social context to the problem of integrating cultures, examining each of the possible subcases in turn. It concludes by noting that how social contexts in different cultures map onto each other (or not) matters greatly in terms of both the outcomes of meeting cultures and the steps that might be taken to facilitate their integration. However the possible interactions are complex and dynamic, so the chapter ends by considering simulations that might start to explore such complexities and outlining some ways to approach this.
Similar content being viewed by others
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
1 Introduction
Context is everywhere in the human social and cognitive spheres but it is often implicit and unnoticed. When trying to understand and model the social and cognitive realms it has become increasingly clear that context is a crucial factor. In particular this chapter will argue that is it vital if one is going to map how different cultures can relate to each other.
This chapter starts by discussing context in general to distinguish some of the many meanings of the word “context”. It then goes on to briefly discuss the pervasiveness of context-dependency in human cognition and how social contexts acquire their distinct identity, becoming entrenched within a culture. The next sections, which analyse how cultures might map onto each other in terms of their different social contexts, are the core of the chapter. These sections go into some detail about the various possible cases of match/mismatch both in terms of the identification of a social context as well as the assumptions, norms, habits etc. that are associated with these contexts. The chapter ends with the implications of this in terms of integrating cultures and studying such integration using agent-based simulation.
2 About Context
The word “context” is used in many different senses and has many different analyses (Hayes 1995). It is somewhat of a “dustbin” concept, in that if a theory or idea does not work the reason may be assigned to “the context”. Thus for many (e.g. linguists) context is a subject that is generally avoided due to its difficulty. I cannot touch on all the approaches to and models of context in the literature, but will give a brief introduction to context in general, including some major conceptions of it, and a few of the issues surrounding it. This will, hopefully, clear the ground for the main suggestions of this chapter and avoid some of the possible confusions.
2.1 Situational Context
The situation context is the actual situation where some events or other described phenomena take place. This could include the time and location, but could include all that is the case about that situation, including: who was there, the knowledge of those people, the history of the place and the objects present. In this sense the context is indefinitely extensive, it could include all the circumstances in which an event occurs.
Such a context may be able to be specified adequately (if rather uninformatively) by giving the time and place of the event,Footnote 1 but the relevant details might not be effectively retrievable from this. For example, the fact “I was reminiscing about our summer holiday” might well not be detectable from the time and place except by the person doing the reminiscing. Thus when talking about the situational context it is common to abstract from this to that which is relevant about that context, or what might be commonly understood. Thus the phrase “the context” (as in the question “what was the context?”) may mean “those factors that are relevant to understand this particular occurrence” even though it may refer to the original situational. Thus to understand what someone is saying to you, you might ask “what was the context?” and get a description of the circumstances, e.g. “I was on the train with my mother”.
2.2 Linguistic Context
Whilst the situational context could include anything, at least in theory, the linguistic context is composed of the words that surround an utterance or phrase. This typically indicates the words that precede or frame the target of understanding, but could also include common knowledge that could be reasonably be expected to be known by the listener/reader, e.g. elements of the relevant culture. Sometimes this is taken to be the same as all that which is necessary to understand some natural language.
Historically this has been what one appeals to if there seems to be no detectable foreground features to explain some particular meaning. However more recently more positive attention has be focused on context in linguistics. For example, Peter Gardenförs has said (“pragmatics” being a term that includes contextual considerations in linguistics):
Action is primary, pragmatics consists of the rules for linguistic actions, semantics is conventionalised pragmatics and syntax adds markers to help disambiguation (when context does not suffice). (1997)
Clearly the linguistic context could refer to almost any of the language or culture that surrounds an utterance, and hence is not something that can be captured in its entirety. Often context is thought of as linguistic context because the interactions that are being considered consist of linguistic communication.
2.3 Cognitive Context
Clearly many aspects of human cognition are context-dependent, including: visual perception, choice making, memory, reason and emotion (Kokinov and Grinberg 2001). What seems to occur is that the human brain categorises kinds of situation which it is able to later recognise, largely without conscious effort. A lot of recall, learning and inference is with respect to these recognised kinds of situation. This abstraction of a situation in the brain – the recognised kind of a situation – is the cognitive context. It is the cognitive correlate of the situational or linguistic context, abstracting and limited by what is relevant. Such cognitive contexts could be identified using a description of the kind of situational context that invokes them or else by the set of all the knowledge, norms, expectations, habits etc. that are immediately accessible once recognised.
It is essential that the circumstances under which different cognitive contexts apply can be effectively and reliably recognised but this does not mean that they have to be consciously recognisable as distinct contexts and labelled. They may be unconsciously recognised by all the members of a community but never named; maybe they their features are distinctive and consciously recognisable but too complex and fuzzy to be completely specified.
That we flexibly learn to recognise contexts and what is appropriate to them that allows for the culturally-specific development and entrenchment of social contexts.
2.4 Social Context
Many of the cognitive contexts we have learnt seem to correspond to recognisable kinds of social situation. Examples include: greeting, lecturing, and a political discussion. Once established these seem to be self-perpetuating: habits, conventions, norms, terms etc. are developed by people who recognise the context, but the fact that there are specific habits etc. that pertain to this makes them more recognisable. Thus social contexts can be co-constructed over time and passed-on to others. Individuals co-learn the cognitive correlates along side others so that these correspond to the kind of social situation (similar to how we learn to use the same word for the same colour).
When people are asked to describe the context, they will often do it in social terms. Thus it is that the social context, although it is a special case of situational context is closely linked to the synchronised cognitive context that participants have learnt to associate with situation, because it is often the social aspects that are important in terms of communication and understanding. It is, of course, social contexts that we will be primarily concerned with in this chapter, although these must largely correspond to the cognitive contexts in the minds of the society’s.
2.5 Identifying and Talking About Context
One of the difficulties in discussing context is that they may well not (a) be accessible to us (b) identifiable even if they are accessible or even (c) definable in precise terms even if we can identify them. This is due to the complex and largely unconscious way in which context is recognized. Rather we often have to try and deduce what the relavant contexts are by introspection and other observation.
Despite this, we often talk about contexts as if they were discrete “things”, however it needs to be understood that for our conscious selves this may not be the case. Thus “the” context is an abstraction of the aspects of those background features that define it, whether or not this is a meaningful or reifiable entity for us. To simplify the discussion I will generally talk about contexts in the sections below as if they are well defined identifiable entities, but the caveats just mentioned need to be always taken into account. This difficulty means that the context for any situation is often not made explicit or represented – those involved may well not be aware of the cognitive context they are assuming.
Thus although we may not be able to describe or specify cognitive contexts in general due to their fuzzy, complex and inexact nature, social contexts are more identifiable due to their institution into the fabric of our society and the fact that they need to be readily recognizable by all actors.
To aid the discussion to follow I will use some shorthand terminology. The context means the recognized kind of situation where specific rules, norms, language, dress, habits etc. may hold. The contents of a context are those specific rules, norms etc. that are associated with that context that will be widely known by the social actors participating in them. The analogy is of a set of containers of these specific items of knowledge that are taken off the shelf and accessed depending on the recognized kind of situation.
3 The Context-Dependency of Human Behaviour
3.1 The Pervasiveness of Context-Dependency in Human Cognition
People behave differently in different situations – the rules, norms, expectations and decisions that people apply in “a lecture” will be very different from those they apply at “a celebration”. I would claim that this is not a coincidence, nor only a surface difference masking underlying universal patterns of behaviour – but that context-dependency is fundamental to the way humans deal with the world. There are several reasons why this claim is a strong one. Firstly, there is the simple observation that the same people behave very differently in different social contexts and that different people behave in similar ways in the same social context. Secondly, there is a lot of evidence that human cognition is context-dependent in many respects. Kokinov and Grinberg (2001) list some, including: visual perception, choice making, memory, reasoning and emotion, preferences, and language comprehension. Context-dependency seems to be hard-wired in our cognition, and it would be very surprising if this did not result in a context-dependency in terms of behaviour. Thirdly, one can see why it might have evolved in our species. If, as seems likely, a significant part of the evolutionary advantage that our brains provide us is in our ability to organise and adapt in social groups – as suggested in the “Social Intelligence Hypothesis” (Kummer et al. 1997) – then the ability to learn to behave in a highly context-dependent manner can be explained in terms of the significant advantage that would result from groups being able to develop different shared norms, habits and protocols to suit different tasks and situations.Footnote 2 So, for example, how to behave on a fishing expedition can be developed to suit the conditions and technologies available for that activity, but the patterns of the whole group would change quickly and simultaneously if a conflict with a competing group arose, or a storm was coming.
Human cognition seems to involve the combination of rich, unconscious, fast and vague context recognition with relatively simple, conscious, slow and precise reasoning and learning relative to the context (Edmonds 1999a). Dividing the world into similar kinds of situation and dealing with it on that basis makes the conscious reasoning, learning, and decision making feasible (Greiner et al. 2001). The flexibility of this combination or rich Machine Learning kind of mechanism with slow but specific Artificial Intelligence style reasoning and adaptive mechanisms seems very effective and powerful for the particular environment and social existence that we have (Edmonds and Norling 2007).
One of the features of this combination is that the recognition of the kind of context is usually unconscious (unless, for example, we misidentify the appropriate context). Thus our brain automatically (and apparently seamlessly) is providing us with the knowledge, expectations, habits, vocabulary to deal with the kind of situation we are facing, without apparent cognitive effort. This rich and unconscious context recognition can make it hard to indentify or even talk about context, which is perhaps why it has not had the attention that it deserves.
Also, it must be said, that there has been a bias against context-dependent understandings of social and psychological phenomena on the grounds that it is not “scientific”. It is true that universal models and understanding is preferable if they are possible, but there is no reason to supposeFootnote 3 that the world has been so conveniently arranged for us in this respect (Edmonds 2012).
Regardless of our ultimate philosophical views on the existence of universal underlying mechanisms, in practice it is sensible to understand and model human behaviour as context-dependent. This is particularly true in the social and cultural sphere where reductions to putative universal underlying mechanisms are currently no more than a theoretical commitment.
3.2 The Development and Entrenchment of Social Contexts
Whilst general types of environmental situation may well be identified and learnt by people (paths, clearings etc.) it is in the social sphere that context is delineated in the most obvious manner. This is because situations that are recognised as a kind of context become entrenched as the result of social processes. Thus if a situation occurs, such as a lecture, then it may be recognised by others too. Over time particular rules, norms, ways of behaving and language might be invented to suit that kind of situation. The more particular things pertain to it, the more clearly it is recognisable. The more clearly it is delineated the more it is likely that things will be invented or adapted for that kind of situation – the context becomes socially entrenched via the co-development of its identity and specific content.
For example, consider the lecture. It is likely that early lectures were held outside or simply in people’s houses. They might well have been much more fluid than the lectures we are used to, with people coming and going during it, and more of a dialogue, or even barracking, during the lecture. The lecture may not even have been distinguished from other kinds of teaching or discussion. However over the years the lecture has developed into a sharply defined institution. We build special rooms to hold lectures in. People are trained from an early age how to behave in a lecture, that it is not allowed to disrupt a lecture, that one can expect the lecturer to have some expertise in what they are talking about, that limited amounts of questions will be allowed to be put but not free-form dialogue. It is now trivial to recognise a lecture and we all know how to behave and what to do in them.
All cultures, through necessity or invention develop a series of kinds of situation that have different purposes or provenances and which are correctly recognised by the members of the culture. That is, it is not only the expectations, habits, language, actions, norms, dress etc. that will have developed differently in different cultures but also when different sets of these pertain. These kinds of social context structure the different cultures, in ways that its members recognise. In other words, social context has a crucial role in the social embedding of individual action (Granovetter 1985).
This is not to say that all culture reduces to considerations of context, but just that context-dependency so permeates human social action that it is impossible to ignore. Within a single culture, among acculturated members, knowledge about its social contexts will be common, so there is no need to make the contexts explicit. This and the fact that context is omnipresent and unconscious means that little is written about the relationship of context and culture. When considering how cultures might relate to each other, one cannot leave context out of the picture. The “contents” of social contexts will also differ profoundly from those of other cultures, even within closely identified contexts. However this chapter is focussing on the extra dimension that emerges when each culture has different contexts, when different cultural elements are relevant, as well as that of how these elements might differ or relate.
4 Implications for How Different Cultures Map onto Each Other
The principle point of this chapter is to point out that the existence and identity of the entrenched social contexts in cultures will have a huge impact as to how those cultures relate to each other. That is to say, the relative structure of cultures matter.
Sometimes the social contexts that have developed within different cultures will roughly coincide. That is to say the broad identification, function and style of a social context in one culture will be identifiable in another, even if the content and scope differ somewhat. Thus a service of religious worship may vary greatly from culture to culture in terms of what is expected of people within that service, when it happens, its place in relation to other aspects of society etc. etc. but is still broadly recognisable as such. Other examples of this include a court of law or a wedding.
Why these are identified as being of the same kind, or indeed why they are even identifiable as the same kind is complex and sometimes mysterious. It may be that the social institutions have travelled across the world with the spread of technology, people, trade, empires or religion. It may be that the commonality can be traced to a need or function that is the same for people everywhere, for example a funeral or cooking. It may even be that situations with different roots come to be identified as belonging to the same category, for example a musical performance. Of course, it may be that some social contexts that are identified as the same have, in fact, little in common but some relatively unimportant surface features. A meal might involve some eating by definition, but the meal of a solitary diner experiencing a variety of flavours in a haut cuisine restaurant might have nothing much to do with the ceremonial eating of unleavened bread at the Jewish Passover festival.
In other cases the social context in one culture may not have anything that corresponds in the other. So the social context of a commuter train may not have much in common with travelling in a nomadic community. Sometimes a new social context appears that does not spread to other cultures, or has not spread yet. There may be no need for a babysitting circle in places where childcare needs are dealt with purely within an extended family. What is acceptable in one culture might not be acceptable in another, so a “rave” (where large numbers of young people gather to dance to hypnotic music with an accompanied use of drugs and alcohol) would simply not be tolerated in many countries.
Of course, deciding whether or not a particular social context in one culture corresponds to another in a different culture is highly problematic. Social phenomena are horrendously complex, changing and subjective. It may even be that, ultimately, it is impossible to come up with precise identifications that can stand up to rigorous questioning. However this is not the point. However problematic the identification of social context across cultures is, the fact is that people do identify some social contexts as similar across cultures and this effects their judgements and reactions to elements of different cultures. How these social contexts are perceived, and the social consensus on this matter does have a definite consequences for how and when cultures can integrate.
The central point of this chapter is that integrating an element of one culture with another will be very different in cases where the social contexts are widely considered to be of the same kind to those cases where the social context of one of the cultures is not recognised as having an equivalent in the other. That is to say that the structure of social contexts matters, as well as the “content” of those contexts. The reason for this is that the scopes of social contexts (when they are considered to occur) are very difficult to change, often deeply embedded within a culture, and to a considerable extent unconsciously assumed.
This analysis implies that cultural integration will be fundamentally affected by the structuring of social contexts in each culture. The case where the contexts largely overlap (e.g. an academic lecture) will be very different from when they do not (e.g. some religious contexts). This, in turn, implies that there will be some very different kinds of cultural integration corresponding to these different cases.
4.1 Different Kinds of Integration that May Occur
Let us consider some of the possible cases of correspondence, or lack of it that might occur when different cultures have to co-exist and hence encounter each other. Table 9.1 lists four basic cases of match/mismatch: where a social context in culture A does/doesn’t have a corresponding social context in culture B, and where the “content” of the social context in culture A is compatible/clashes with culture B. Of course the author recognises this is a simplistic categorisation but these differences come from pragmatic social considerations and facilitates discussion of each case. I have given each case a simple label to make the exposition clearer.
4.1.1 Case 1: Compatible New
In this case there is a social context in culture A that does not correspond to anything much in culture B, but the content of that social context is largely compatible with the norms, ethics, habits etc. of culture B. Thus a new pastime that is transplanted with the movement of people to a new culture can simply add to the menu of choices in the receiving culture, and recognised as such. Indeed it may be that it becomes a cherished context within the receiving culture and, over time, becomes embedded within that culture. This case is largely unproblematic, since there is no confusion that the context is new, nor any competition as to the detailed identification of that kind of situation. There may have been no such thing as “homework” (a specific piece of work for the pupil to do out of school and handed in to be marked) in some cultures, but as a potentially useful addition to the life of children can be simply added-on if it does not conflict with other patterns or duties.
4.1.2 Case 2: Roughly Compatible
In this case there is a social context in culture A that many people would identify as essentially the same (or indeed “the same”) as a social context in culture B, and the nature of what happens in the context in culture A is roughly compatible with that in culture B (it is almost impossible for it to be identical, even with an institution trying to standardise them). This case is problematic in so far as people might simply assume the content and identification of the situations to be completely the same when there are, in fact, small differences. Thus instead of dealing with a situation using well-entrenched habits associated and triggered by the context, some conscious thought and adaption might be necessary. Thus if one goes to a religious service in another country, even one within the same global institution, then there might be differences in norms about lateness, talking during the event, style of dress, etc. An unthinking reaction might lead to a negative reaction to these differences (e.g. “they had no respect, talking and chatting all through it!”).
In this case bringing the differences to the foreground, pointing them out and the different reasons and roots of the differences, i.e. explicitly educating people about the differences can help ease any dissonance that might have occurred. The trouble is that the habits and assumptions associated with particular social contexts are largely automatic and unconscious otherwise. Thus, in both US and UK cultures there is a well-recognised social context of “greeting and getting to know a person” within such events like parties, meetings etc. However (to generalise broadly) it is largely the norm within US culture to tell the other about oneself as a way of opening up the conversation and in the UK it is the norm to ask the other about themselves for the same purpose. This can lead to the case where after a first meeting between a US and UK citizen they come away with negative impressions of each other (“the US are always bragging about themselves and never once asked me what I do” and “the UK people are so snobby and close he did not tell me anything about himself but kept me at a distance” being stereotypical reactions in this case). Sometimes simply pointing out the differences can be enough to sort these misunderstandings out, in other cases habits are so ingrained or beloved that people are unwilling to adjust possibly leading to minor mutual irritation.
4.1.3 Case 3: Clashing New
In this case, there is a social context in culture A that does not correspond to anything much in culture B, i.e. it is “new”, but the content of that social context is either incompatible with the norms etc. of culture B or it is perceived as being problematic from the point of view of Culture B. The “intruding” social context is not identified with social contexts in the receiving culture by many of those in that culture. Of course such an intrusion may be a matter of perception and not, ultimately, a matter of practical, legal or moral incompatibility. Thus there may be a neutral or positive reaction from people who identify a Mosque with a Church, and Islamic prayer and worship with Christian prayer and worship to the plan to open an Islamic Centre near the site of the 9/11 terrorist attack, but a negative one from those who see Islam as basically alien to them – an intrusion into their society. Whether the incompatibility is real or perceived, the conflict it can trigger might be very evident in the form of peoples’ actions and rhetoric. Without making a judgement there are several possible outcomes to such a situation (in general).
The receiving culture might decide to ban or discourage the new kinds of social situation from taking place. This might or might not be successful, depending on the lengths to which they are prepared to go, the level of conflict with the sending culture that they are willing to tolerate, and the importance and embedding of the social context within the sending culture. There are cases where this has been largely successful (e.g. monogamy even for immigrants from countries where polygamy is practiced and legal), and cases where it has not worked (e.g. the past attempted suppression of churches in China).
After a time it might be that the new context is accepted in the receiving culture. As new generations grow up with the “problematic” social contexts being part of their social environment, it may not seem so threatening. Of course, this depends on the incompatibilities not being fundamental, so this acceptance will occur if either: the perceived incompatibilities become accepted over time, or the incompatible elements are adjusted by the incoming culture so that it is acceptable. Such a process of acceptance and/or adjustment can be facilitated with the involvement of the receiving culture’s people in the introduced context (either directly though participation or indirectly via social contact) both in helping correct misperceptions but also in the introduction of the receiving culture’s values to those in the incoming culture.
4.1.4 Case 4: Internal Clash
In this case the social context of the incoming culture is identified with that in the receiving culture, but there are incompatibilities in terms of what happens within that social context. For example, “waiting for a bus” is a social context that is recognised widely throughout many cultures, but the norms of how one behaves when the bus arrives might differ. There is an obvious and direct conflict between the norms of queuing (the people enter the bus in the order they arrived at the waiting place) and walking on without regard to order as quickly as possible. The former might well resent the latter as “pushing in” illegitimately, and the later might be frustrated at what they perceive as the “unnecessary” formality concerning a simple action of entering a bus, especially in cases where there is room for all. Both views may react on the basis of deeply entrenched habit and norms, and resent the other pattern of behaviour, often attributing onto the others bad motivations and character.
In some of these cases there is no easy resolution, but that one or other pattern of behaviour will win out. It may be that in the long run social influence determines the outcome, either newcomers are persuaded to adopt the norm of queuing in the UK, or it may be that this norm falls into disuse. Sometimes these clashes are decided by enforcement, with reference to “fundamental” rights and duties of a society that incomers or inhabitants must abide with. In France it was decided that wearing the hijab in public was illegal – thus in this case that the incomers were forced to adapt to the existing norms. In the US it may be that the principles of freedom of religion and expression would make such a law impossible, and it be the duty of the receiving culture to be tolerant.
However another response to this case (Internal Clash), is to encourage the differentiation of contexts, so that the contexts with respect to culture A and culture B come to be considered as separate kinds of situation and thus avoid the conflict due to the expectations of people that they are the same. Perhaps an example of this is the variations of the academic lecture that may be found in some of the more traditional Islamic world, where the lecture hall is arranged so that male and female students are screened from each other.
4.1.5 Case 5: New Contexts
A case that is not covered in the above classification is when a new social context is created that is separate from those in either culture. Such a context might be created to ease the interaction of different cultures, since neither culture will have engrained expectations of such a context. This might reduce the misunderstandings that might arise, since all participants are aware they are not in any of their “home” contexts. An example of this might be the international business meeting, which has gradually evolved to be distinct from a normal business meeting in any one country.Footnote 4
4.2 The Implications for Attempts to Promote Social Integration
If it is indeed the case that the social contexts, along with their associated habits and expectations are relatively difficult to change (once entrenched), then this has consequences for what might be effective at promoting cultural integration or, conversely, avoiding cultural conflict. What is likely to work will be dependent on which of the cases above one has. Case 1 is largely unproblematic. In Case 2, education (explaining the differences) and making the unconscious assumptions explicit, bringing them into the open might be effective. In Case 3 there might well be no adjustment possible and the receiving culture simply has to decide whether it will tolerate the “intrusion”. In Case 4 the only thing to do is adjudicate as to what can occur and what the fundamental rights are, hoping that time makes the alien familiar. Examples of Case 5 should simply be encouraged as the easiest medium-term approach to establishing working interaction and dialogue.
Of course, these prescriptions are simplistic and imprecise, more of a starting point for investigation rather than its conclusion. The dynamics of perception, context and interaction can easily make them otiose. Also the fractal nature of social contexts (sub-contexts within contexts etc.) can make a useful analysis of how contexts might correspond complicated. Finally what a social context is depends crucially on how they are perceived, and so changes of perception might result in apparently radical and quick changes in the scope of social contexts. However, it is also equally clear that a generic approach to considering how cultures relate and/or integrate which does not take their contextual structure into account might be woefully inadequate.
5 Simulating the Integration of Cultures
5.1 Representing Both Cognitive and Social Aspects of Social Context
Clearly there is a lot that is not known about: how social contexts and social norms interact, how they develop and fall into disuse, how people recognise social contexts, how group identity and signals and social contexts interact… etc. etc. To start to understand and unpick these complex and complicated interrelated factors we need to simulate them – since there will be a severe limit to how much one can keep track of these implications informally. The micro–macro link here is important, in other words it is essential to understand how the abilities, biases and intentions of individuals determine and are determined by the higher-level social.
If the above analysis is at all correct then (a) the awareness, identification and representation of social context is essential to fully understand how cultures might integrate and (b) the dynamics of cultures, with their constituent individuals, social contexts, norms, habits, assumptions and interactions could be highly complex. Together these indicate that in order to get a fuller picture, agent-based simulation is the most appropriate tool (Edmonds 2010b). This is because this technique can (a) represent some of both the cognitive and social aspects involved in social context and (b) track some of the complex interactions between social context, perceptions, habits, norms and actions of the individuals concerned.
However simulations that take seriously both the cognitive and social complexity in terms of what is represented is rare. It has tended to be that in many social simulations the KISS principleFootnote 5 rules as far as the cognitive model of the agents is concerned, concentrating on how complexity can emerge from the interaction of many relatively simple individuals. This is the approach exemplified in (Axtell and Epstein 1996). Clearly one can use such simulations to discover possible ways in which social complexity could occur, but this does not tell us how complexity in observed human societies occurs.Footnote 6 There is a community that takes the representation and simulation of human cognition seriously – the cognitive modelling community. This community does seek to represent in detailed simulations how we think. However it is very much from the individual point of view – understanding how an individual thinks. The social situatedness of human cognition is rarely touched upon here, and thus explorations of how the social embeddedness of human social artefacts, such as social context emerge, are maintained etc. are not possible.
However there are some projects that are starting to include representations of both cognitive and social complexity in their simulations. An example (though not about social context) is the EMIL project (Conte et al. 2010) that sought to simulate the twin cognitive and social aspects of social norms, allowing the exploration of norm emergence in terms of both mental perceptions of obligations and the social patterns that co-developed with these. A similar project is needed to start detangling the co-dependence of individual learning and perceptions of social context and their institution and embedding within the practices and artefacts in society.
5.2 Some Existing Simulation Work that Points in These Directions
It is notable that very few social simulations represent any of the processes for dealing with such context-dependency. That is to say, the agents in social simulations tend to be endowed with cognitive processes which are not sensitive to, recognise or use context. If the situation in which the agents are being represented can be considered as a single entity, so that all interaction can be considered as taking place within a single agent, then this is reasonable since one then does not need context.
However many simulations aspire to be a more general theory of social interaction. In this case, one has to assume that either the simulation is to be taken only as an analogy or that the simulator thinks that people’s behaviour, norms etc. will be so similar between social contexts that including mechanisms of context recognition, dependency, etc. are unnecessary (Edmonds 2010a).
In the former case where the simulation is used only as an analogy, then this is valid because humans are experts at applying analogy in a context-dependent manner, adjusting its assumptions and form to be appropriate to its domain of application.
In the later case, where an essentially context-independent algorithm is used to represent a highly context-dependent process must, at least, be the legitimate target for doubt. Whilst the psychological realism that is necessary in a social simulation does depend upon the purpose of the simulation and the level of aggregation (Gilbert 2006), it is certainly not the case that the results of a simulation can be assumed to be robust against changes in the cognitive model being used (Edmonds and Moss 2001).
There are not many simulations which represent aspects of context-dependency in their agents, but there are a few: (Edmonds 1998) used a cognitive learning model specifically because it included some aspects of context-dependency; (Schlosser et al. 2005) argue that reputation is context-dependent, (Edmonds and Norling 2007) looks at the difference that context-dependent learning and reasoning can make in an artificial stock market, (Andrighetto et al. 2008) shows that learning context-dependent norms is different from a generic adaption mechanism, and (Tykhonov et al. 2008) argue that the definitions of trust mean that trust is also context-dependent. (Alam et al. 2010) present a model of the exchange and family structure within a Mexican village based on the cognitive model of choice called “endorsements” (Cohen and Grinberg 1983). They justify the choice of this particular mechanism by the way it can result in context-sensitive choice.
These show that, at least in some cases, that context-sensitive cognition can make a difference. The fact that it can make a difference is not very surprising given the important role it plays in human cognition and society.
There are approaches to including cognitive context within the learning and decision-making of agents. (Andrighetto et al. 2008) use an approach based on social norms, whereby some of the habits and knowledge of agents are dependent upon the social context, in the sense of which group they are part of. (Edmonds 2001) suggests a particular algorithm and approach to learning appropriate cognitive context (which is discussed further below). This showed that the knowledge could be learnt and recognised in a way so that distinct contexts emerge, but it did not show that agents co-learnt the same social contexts, due to the anti-coordination motivation provided by the artificial stock market environment the agents inhabited.
It must be said that cognitive contexts that implement cognitive dependency are thin on the ground. This indicates that more work, both foundational and applied is needed if social simulations that can start to represent social context are to be become useful. However the next section looks at some possible ways forward in this respect.
5.3 Towards Implementing Simulations that Incorporate Context
Broadly there are three feasible approaches for incorporating context-dependency within a simulation. These are as follows.
-
1.
Within a Single Context. That is focus on one, identifiable social context and only model agents acting within this.
-
2.
Within a Fixed Set of Known Contexts. Identify N social contexts and model the behaviour and interaction within each of these, along with the agents’ recognition of when a shift in context is appropriate.
-
3.
Learn to Identify Context with the Behaviours. Let the agents in a simulation learn what the appropriate contexts and context-specific knowledge are.
The first of these is “modelling as normal”. A context is identified (or assumed) and the behaviours of agents are simulated in the normal way. The behaviour of agents is based upon a fairly simple cognitive model, incorporating those elements deemed to be relevant for the behaviours in that context. All of the interactions are presumed to occur, essentially, within that context. As long as the context within which the model is valid is reliably recognised by all concerned, either explicitly or implicitly, then the context can be taken for granted, as long as extra-contextual influences do not significantly affect the outcomes (Edmonds 2009).
The second case requires the relevant contexts to be identified, with an essentially separate behavioural model for each context (though there may well be commonalities between them). In addition the rules for when the actors switch contexts need to be modelled. This makes for a lot of work, however by restricting each behavioural to a particular context, it may be that the context-specific models are fairly mundane and simple. This is, after all, what Herbert Simon observed in his observations of administrative behaviour, what he called “procedural rationality”.
The human being striving for rationality and restricted within the limits of his knowledge has developed some working procedures… These procedures consist in assuming that he can isolate from the rest of the world a closed system containing a limited number of variables and a limited range of consequences. (Simon 1976)
Here standard elicitation and ethnographic techniques from the social sciences could be used to lay bare the procedures that people have developed, maybe starting from first-person accounts of what they do normally and some variations from this. However it is necessary to “prime” interviewees with each intended context – maybe presenting each of them via vivid scenarios – to get the full information. Although a number of techniques exist to help analyse and “make sense” of such narrative data, there still lacks a complete methodology to bridge the gap from such data to agent rules in a simulation. Currently this tends to be done informally by the programmer.
The third of the options is to equip the agents in a simulation to identify and learn both the relevant contexts and the behavioural rules themselves. In other words, nothing less than allowing the agents to acculturate within the situation (including the society of other agents) that they find themselves in. There are several major difficulties with this approach. Firstly, there are no mature methods for implementing such a cognitive mechanism, athough I sketch the one I have explored below (Edmonds and Norling 2007). Secondly, the process of acculturation is slow, needing a considerable amount of time so that the agents can learn and adapt to their environment. Thirdly, this requires the specification of an environment for them to adapt to. Since social environments are what we are primarily concerned with here, we have the problem of initialisation of a social environment. To produce a social environment one needs socially adapted agents, but one needs a social environment for these agents to adapt to. This is the deep problem of social embedding (Granovetter 1985; Edmonds 1999b). One can gradually co-evolve the agents to be each others’ social environment, but then one cannot easily direct this to be the same as those observed. However, this might be used to reveal some general patterns and issues concerning cultural integration.
I do not have space to go into such an architecture, but only briefly illustrate it, see (Edmonds and Norling 2007) for details. The architecture is illustrated in Fig. 9.1 below. The memory (CDM) is spread over a space, so that given position in this space (which is the cognitive correlate of the context) beliefs are retrieved and laid down near this position – this neighbourhood is the cognitive context. There are three other components which all access the memory: a machine learning module that guesses the cognitive context from a rich selection of perceptions (the CIS above); a belief revision module (the LL above) that updates the set of beliefs within the cognitive context; and a decision making/reasoning module that takes the beliefs within the cognitive context and makes decisions or plans which determine action.
Figure 9.2 shows the emergent clustering of beliefs within an agent in a simulation of a stock market, which uses the above architecture. This is the set of models to predict the future prices of a particular stock by a particular agent at a particular point in time. The memory of this agent is in a space determined by the dimensions of recent volatility and volume of the market, since anecdotal evidence suggest these factors are important in determining the “mood” of a stock market. In this case the agent has developed three distinct clusters, roughly corresponding to high volatility, high volume and low volume (both at medium levels of volatility). The models are most quite simple, e.g. predicting the price by looking at a price last week.
6 Conclusion
Culture is structured in a fundamental manner by social context. Any attempt to understand social integration or promote it will need to take this into account and not just as an afterthought. This chapter aims to convince that this is the case and indicate some of the ways forward in terms of simulation modelling, understanding and promotion. Such advances in modelling need to be part of a broader range of approaches alongside studies to gather evidence as to how and when social contexts “collide”. Without good evidence to validate simulation models at many different levels (Axtell and Epstein 1994) one is limited to exploring abstract possibilities only.
Notes
- 1.
As is essentially the approach in (Barwise and Perry 1983).
- 2.
It might be that just as language might have co-evolved with the brain (Deacon 1998) that this ability to coordinate via a shared social context might have co-evolved with our cognitive abilities to deal with context.
- 3.
Other than sheer optimism.
- 4.
The fact that such a context derived from that in a particular context does not prevent it developing into a new and separate context.
- 5.
“Keep It Simple Stupid!”, the engineering principle that one should only introduce complexity after simpler approaches have failed – the opposite is “KIDS” (Edmonds and Moss 2005).
- 6.
Of course one can make the heroic assumption that the nature of human cognition does not matter when it comes to the social layer (e.g. Ye and Carley 1995; Gilbert 2006) argues that one does not always have to accurately model cognition in social simulation. However (Edmonds and Moss 2001) shows that the cognitive model can be crucial.
References
Alam, S. J., Geller, A., Meyer, R. and Werth, B. (2010). Modelling Contextualized Reasoning in Complex Societies with “Endorsements”. Journal of Artificial Societies and Social Simulation 13(4)6 <http://jasss.soc.surrey.ac.uk/13/4/6.html>
Andrighetto, G., Campennì, M., Conte, R., and Cecconi, F. (2008). Conformity in Multiple Contexts: Imitation Vs Norm Recognition,. In World Congress on Social Simulation 2008 (WCSS-08) George Mason University, Fairfax, USA.
Axtell, R. L. and J. M. Epstein (1996). Growing Artificial Societies – Social Science From the Bottom Up. Brookings Institution Press and MIT Press.
Axtell, R. L. and J. M. Epstein (1994). Agent-based Modelling: Understanding our Creations. The Bulletin of the Santa Fe Institute 9: 28–32.
Barwise, J. and Perry, J. (1983). Situations and Attitudes. Cambridge: MIT Press.
Conte, R. et al. (2010) EMIL-T: The final report of the EU 6FP EMIL Project. http://cfpm.org/cpmrep206.html
Cohen. P. R. and Grinberg, M. R. (1983) A Theory of Heuristic Reasoning About Uncertainty, AI Magazine, 4(2) http://www.aaai.org/ojs/index.php/aimagazine/article/view/393/329
Deacon, T.W. (1998) Symbolic Species: The Co-Evolution of Language and the Brain. Norton & Co.
Edmonds, B. (1998). Modelling Socially Intelligent Agents. Applied Artificial Intelligence, 12, 677–699.
Edmonds, B. (1999a) The Pragmatic Roots of Context. CONTEXT’99, Trento, Italy, September 1999. Lecture Notes in Artificial Intelligence, 1688:119–132.
Edmonds, B. (1999b). Capturing Social Embeddedness: a Constructivist Approach. Adaptive Behavior, 7:323–348.
Edmonds, B. (2001) Learning Appropriate Contexts. In: Akman, V. et. al (eds.) Modelling and Using Context – CONTEXT 2001, Dundee, July, 2001. Lecture Notes in Artificial Intelligence, 2116:143–155.
Edmonds, B. (2009) The Nature of Noise. In Squazzoni, F. (Ed.) Epistemological Aspects of Computer Simulation in the Social Sciences. Lecture Notes in Artificial Intelligence, 5466:169–182.
Edmonds, B. (2010a) Bootstrapping Knowledge About Social Phenomena Using Simulation Models. Journal of Artificial Societies and Social Simulation 13(1)8. (http://jasss.soc.surrey.ac.uk/13/1/8.html)
Edmonds, B. (2010b) Agent-Based Social Simulation and its necessity for understanding socially embedded phenomena. CPM Report No.: 10–205, MMU. (http://cfpm.org/cpmrep205.html)
Edmonds, B. (2012) Complexity and Context-dependency. Foundations of Science. Online First. DOI 10.1007/s10699-012-9303-x
Edmonds, B. and Moss, S. (2001) The Importance of Representing Cognitive Processes in Multi-Agent Models, Invited paper at Artificial Neural Networks – ICANN’2001, Aug 21–25 2001, Vienna, Austria. Published in: Dorffner, G., Bischof, H. and Hornik, K. (eds.), Lecture Notes in Computer Science, 2130:759–766.
Edmonds, B. and Moss, S. (2005) From KISS to KIDS – an ‘anti-simplistic’ modelling approach. In P. Davidsson et al. (Eds.): Multi Agent Based Simulation 2004. Springer, Lecture Notes in Artificial Intelligence, 3415:130–144.
Edmonds, B. & Norling, E. (2007) Integrating Learning and Inference in Multi-Agent Systems Using Cognitive Context. In Antunes, L. and Takadama, K. (Eds.) Multi-Agent-Based Simulation VII, Lecture Notes in Artificial Intelligence 4442:142–155.
Gärdenfors, P., The pragmatic role of modality in natural language. in 20th Wittgenstein Symposium, (Kirchberg am Weshel, Lower Austria, 1997), Wittgenstein Society.
Gilbert, N. (2006) When Does Social Simulation Need Cognitive Models? In Sun, R. (ed.) Cognition and Multi-Agent Interaction: From Cognitive Modeling to Social Simulation. Cambridge University Press, 428–432.
Granovetter, M. (1985) Economic Action and Social Structure: the Problem of Embeddedness., American Journal of Sociology, 91:481–93
Greiner, R., Darken, C. & Santoso, N.I. (2001) Efficient reasoning. ACM Comp. Surveys, 33(1):1–30.
Hayes, P. (1995). Contexts in Context. Context in Knowledge Representation and Natural Language, AAAI Fall Symposium, November 1997, MIT, Cambridge.
Kokinov, B. and Grinberg, M. (2001) Simulating Context Effects in Problem Solving with AMBR. in Akman, V., Bouquet, P., Thomason, R. and Young, R.A. eds. Modelling and Using Context, Springer-Verlag, 2116:221–234.
Kummer, H., Daston, L., Gigerenzer, G., & Silk, J. (1997). The social intelligence hypothesis. In P. Weingart, S. D. Mitchell, P. J. Richerson, & S. Maasen (Eds.), Human by nature. Between biology and the social sciences (pp. 157–179). Mahwah, NJ: Erlbaum.
Schlosser, A., Voss, M. and Brückner, L. (2005). On the Simulation of Global Reputation Systems. Journal of Artificial Societies and Social Simulation 9(1)4 <http://jasss.soc.surrey.ac.uk/9/1/4.html>.
Simon, H. (1976). Administrative Behavior (3rd ed.). New York: The Free Press
Tykhonov, D, Jonker, C, Meijer, S and Verwaart, T (2008), Agent-Based Simulation of the Trust and Tracing Game for Supply Chains and Networks. Journal of Artificial Societies and Social Simulation 11(3) 1 http://jasss.soc.surrey.ac.uk/11/3/1.html.
Ye, M. & K. M. Carley (1995), Radar-Soar: Towards An Artificial Organization Composed of Intelligent Agents, Journal of Mathematical Sociology 20(2–3): 219–246.
Acknowledgements
The research was done under grants GR/T11760/01 and EP/H02171X/1, both from the EPSRC. Its support and that of the MMUBS is gratefully acknowledged.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Edmonds, B. (2014). Matching and Mismatching Social Contexts. In: Dignum, V., Dignum, F. (eds) Perspectives on Culture and Agent-based Simulations. Studies in the Philosophy of Sociality, vol 3. Springer, Cham. https://doi.org/10.1007/978-3-319-01952-9_9
Download citation
DOI: https://doi.org/10.1007/978-3-319-01952-9_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-01951-2
Online ISBN: 978-3-319-01952-9
eBook Packages: Humanities, Social Sciences and LawPhilosophy and Religion (R0)