Seeing Quickly: reasoning in the age of Trump

President Trump’s behaviour is often a lightning rod for contemporary issues. Take the altercation between him and the CNN reporter Jim Acosta. There is one video going the rounds on social media that shows Acosta pushing the Whitehouse intern who seeks to remove the roving microphone he is holding; he resists her efforts. This video suggests that Acosta is, indeed, bad mannered and ‘horrible’, as Trump puts it; ‘Look at how he treats an intern!’ Democrats are saying that this video is ‘false’, based on cut segments replayed in sequence to exaggerate Acosta’s bodily actions. One can readily believe it. One only has to think of the reputation of the AltRight crowd.  But, just as that faction in American politics seems to falsify with video, think of how the left does it too. Recall the evisceration of Prime Minister May at the last general election in the UK, for example. Then the so-called Cornbynistas devised video ‘memes’ that showed Mrs May saying the same words again and again like a malfunctioning robot. These too were based on splices of videoed action cut together to create a false image.

In short, truth seems to go out the door with the affordances of digital video.

Should we no longer trust video? If not, then what do we trust? Hasn’t video become the lingua franca of our times? Certainly, news reports are replete with video, with tales from the field not being written but shown; our social media feeds are populated by video too.

It is not so much that one wants video, though. It seems to me, and this is what this blog is about, that we all want to see. Seeing seems more important, better than other modes of understanding – whatever they might be (reading, listening, and so on).

Why has seeing come to have this status? Do we presuppose that what one sees should be given more credibility than, say, what one reads? Do things have to be seen if they are to be taken as truth? If this is the case, it begs questions about who sees, what they see and who is us doing the seeing for us.

Consider journalists: one might say that they act as our seeing agents; they are representative of our eyes. Given this, is it better that they capture what they see with video, rather than interpret what they see with words? Do the resulting videos stand as proxy for our own line of sight, a line of sight we would have if we were there instead of them? As I say, our news services do seem replete with video, not with written analysis.

Perhaps this is how journalism is being shaped at the current time. If this is so, then what does it say about such things as our justice system? Here, too, people act as our representatives: judges act on the public’s behalf. Are they, likewise, shifting their practices to make seeing increasingly central to the judgments?

Judges do look at the visual, that is for sure; but it seems to me that their looking has some subtleties that we can readily acknowledge make it different from our own ways of seeing. Judges don’t simply look, they look and they listen to what a witness says about what something looks like and juxtapose this account against other witness’s accounts. A judge puts these accounts alongside other sources of information. These may including other visual materials but also documentary phenomena, such as traces of action stored in computer systems – records of bank transactions, say, written documents too. Visual witnessing is only part of what judges judge on. Judges subsume questions of seeing into questions of evidence and its heterogeneity – where seeing is only one mode of truth.

So, should we approach the farrago over Trump’s altercation with the CNN reporter as nothing more than an example of ‘troubles with seeing’ when all we have is a certain type of seeing – it’s the gaze enabled by the media, by click-thru’s on social media. It is not seeing in the general that is at issue, but specific social practices.

Perhaps. But surely there is a deeper question here bound up with how easy it is, with digital tools, to ‘crop’, ‘highlight’, ‘remove’, ‘insert’ and ‘recast’ the narrative arc of some ‘seen event’ and, given this, the power of the visual as a mode of information gathering. The visual is powerful, despite being easy to falsify.  The visual can be said to anchor everyday veracity, for example. A judge treats the visual differently because he/she is not in the everyday mode. Indeed, many psychologists would say that what a person sees in everyday life functions more deeply in the processing of consciousness than, say, ideas or thoughts; in this view (no pun intended) to see does have precedence in questions of knowledge. In my own field, human computer interaction, this psychological view is often used to justify the merits of visual memories and recall enabled by digital tools. One consequence of these efforts is that those who have less interest in the cognitive but more in technology and its consequences, such as media studies theorists, are coming to suggest that the ‘visual’ is measure of what we have become. The digital is making us visual, they say.

Yet these arguments from psychology and media studies miss something that, I think, is more important still. When it comes to the visual, to the use of video, it is not the ease with which what can be seen can be altered and made false that is at issue, it is the efficacy of conveying messages and eliciting responses quickly that is. What one should note about our current world is not that falsification is now more commonplace and easily achieved with video. It is not truth that is threatened by making visual concerns the anchor of our knowledge. It is, rather, a question of the pace of our judgements. We see and judge at a glance and do all this quickly; that is the point. We do not see and ask for more information, we do not think slowly. In this regard, the use of video in debates about politics reflects a desire for quick responses, not questions of objectivity and truth. We want quick actions in politics, not slow, ponderous choice making. Politicians must act, we can hear ourselves say, however compromised their choices as a result.

One might add that it is not just video that affords this rapidity. Think of Trump’s tweets. Whether they are true or not doesn’t matter, it’s the affect they have that does, and the affect is instant, unreflective, accusatory. They are sent quickly, read quickly, and have their affect quickly. It would appear that they are created quickly too.

And that is the point. We are all like Trump now. We might not tweet but we make judgements with a glance; we see, we believe; we judge instantly. It’s not the technologies that are undermining questions of truth, but the tempo of our thinking. And it is not just politicians who are thinking quickly. It is all of us.

Advertisement

The ‘new A.I.’ and the future of paper: a strange juxtaposition

Fifteen years ago, it was imagined that the emergence of personal computers, network systems and the World Wide Web would totally transform work and home. The experiences people have in these settings would become completely digital. At work, paper would cease to be important in messaging or reading, with documents of all kinds being created, exchanged and stored online. The location of work would be transformed as well, with digital connections allowing business to be done anywhere (and indeed anytime). Home life would be revolutionised, with news and lifestyle information no longer being delivered in paper form, in newspapers and magazines say, but online, via the PC. Where people would shop would alter, too, with people using e-commerce to purchase from the couch instead of, say, on the high street or in a shopping centre.

In the past three or four years, further technological change has occurred. The emergence of cloud-based infrastructures has led to Big Data, and this, combined with the new machine intelligence, has led to user data analytics that are allowing some to claim that what people want can be predicted to a fine degree. Some have proposed that this combination will transform the relationship between people and technology. Employment levels will be reduced as intelligent machines move from the factory to the office and replace important decision-making professionals; machines will become experts in work life. As regards the home, data analytics will allow businesses and online providers to deliver content and marketing materials precisely when the consumer needs them. Search engines will not wait for a command, but will deliver what they predict their user seeks. What people want, when they want it, and how to elicit this through marketing, will be understood by intelligent systems interrogating vast sets of data. In effect, the choices people make either in the home and in the workplace will be predictable; choice will be tamed by technology.

On the face of it, these trends would appear to suggest that the use of paper for various purposes will inevitably decline and the digital will take over. Yet, and at just a common sense level, this is evidently not happening in any straight-forward or linear way. To take some examples: the much-hyped arrival of the e-book which some said would lead to the eradication of paper textbooks and novels, seems to have stalled (1). In a similar vein, although sales of paper-based magazines have shown some decline, they have not been replaced by e-content (2). Other research, this time looking at attitudes, suggests that people are returning to a more favourable view about paper as a marketing medium (3). In the first two examples, the issues are related to the ways paper gets used, its ‘interactional properties’ as it were, in the third, it has to do with the changing landscape of perceptions. Paper no longer evokes a dying medium; on the contrary, it has a positive place in people’s attitudes towards how they conduct their affairs at home and work.

These examples make it clear that, just as the relations between paper and digital have been complex and have evolved in diverse ways, so they are likely to continue evolving in equally complex and divergent ways. The relationship between paper and the digital needs more than common sense or brief research studies. This ecology needs systematic research; indeed, a science of how that ecology functions and how it will evolve.

Research of this kind has certainly been done in some respects. The impact of, for example, the first of these transformations, deriving from networked systems, PCs and the Web, has been shown to be not as predicted and the reasons for this explained scientifically. Research reported in, as a case in point, The Myth of the Paperless Office (Sellen & Harper, 2002), showed that paper would continue to have a role in the work setting. Through extensive observational data and experimental examinations, this study demonstrated that the affordances of paper could not be substituted by digital alternatives of the time. For certain tasks, particularly related to information analysis and comprehension, for what one might gloss as essentially human processes of decision-making, using paper was more effective; it enabled people to ‘interact with’ information in ways that best suited the mechanics of their cognition.

Nevertheless, this and other research also showed that digital technologies were altering workplaces in important ways, combining with paper to create new opportunities for information gathering and use, altering the landscape of decision-making as it did so. Networks provided more information to people, enabled widespread transmission and almost infinite storage. This, in turn, altered the role of reports and documents as they came to articulate greater amounts of information; less attention being given to particularities and more to breadth, to navigation. New forms of documents came to emerge that linked information resources on the Web to internal organizational archives. Shared repositories of information in applications like, for example, Microsoft Sharepoint, came to reflect this alteration. Meanwhile, paper continued to have a role in comprehension and document provision, attested to albeit crudely in the increasing volumes of office papers supplied (4).

Early research in home settings, reported in collections like Inside the Smart Home (Harper, Ed. 2003) explained why paper would persist in the home for reasons similar to those applicable to the workplace. This research highlighted also how technologies were creating changes in the patterns of people’s domestic activities, particularly to do with the experiential form of their decision-making. For example, and while it was agreed that search engines and e-commerce would open up when and how people shopped, research made it clear that this would not result in substituting traditional forms of consumption, on the high street, say. Rather, browsing with a search engine would come to extend the footprint of shopping to include both the high street and the couch. As a result, where people chose, to use a shorthand for consumption practices, would broaden. As it did so, there would be a greater role for paper. More particularly, as the couch became part of the shopping domain, so ways of conveying information in ways appropriate to the leisurely manners of the couch (and the domestic setting more generally) would become more important. Paper brochures and magazines were predicted to have an increasing role because of this. Recent investigations seem to confirm this research, certainly if this is to be measured in terms of gross print volumes. This view has been given further weight by attitudinal findings and small-scale observational evidence. This latter research has mostly been undertaken by the marketing industry, however, and this brings into doubt matters of objectivity. The evidence might be right, but the authority is not scientific.

Whilst studies of the first of these revolutions were comprehensive, far less has been invested in examinations of the new A.I. either by scientific investigators or the industries that might be affected by the new data analytics. With regard to the workplace, some initial research has focused on how machine intelligence might replace the professional in decision-making processes (see for instance, Frey, et al, The Future of Employment, 2013). This substitution will result in a reduction in the numbers of such staff a company might have overall. As a byproduct, there will also be a reduction in the need of paper given that it supports the cognitive processes of these decision-makers. However, this research has not, in any detail, examined whether there are variations in decision-making that will constrain the potential impact of new systems. In some domains these new systems may well replace human decision makers since the choices being made are well suited to probabilistic solutions; in other cases less so. Besides, there is also a lack of evidence about the state of the technology, with ease of use and cost being taken for granted, for example. One particular problem is making intelligent machines intelligible to the user. This is not a simple matter.

More generally, however, the state of research about the ‘new A.I.’ echoes the kinds of excited but often un-evidenced claims that prompted the research reported in The Myth of the Paperless Office. Many of these claims are discussed in Harper et al’s critical assessment of decision-making theories in their book Choice (2016). As they show, many of these claims ignore important complicating factors that will reduce the potential impact of these technologies, or at least make their role less than clear-cut. For example, many of the claims assume that professional decision-making is individual work whereas in the workplace most decisions are the output of joint and co-operative activity. Because of this the new systems may not have the impact some expect.

The role of new predictive technologies in the home has been investigated somewhat more comprehensively, though the concern here has been more to show how these technologies alter the kinds of experiences people have and less on how the new A.I. is making choices for them. For example, considerable change has occurred in the games space with a range of new technologies being used to support camera-based interaction – this has created new experiences for people. That new machine learning techniques do this is largely invisible to the user. Meanwhile, a similar set of techniques have had a much more demonstrative role in helping players find game partners on line. However, attempts to leverage these new techniques to support human decision making elsewhere in the home, with regard to cooking, for instance, or in the management and purchasing of domestic provisions, food, cleaning materials and so on, have been less successful. For a variety of reasons, people prefer to cook as they see fit and not with automated systems; similarly, they prefer to manage their own cupboards and fridges even if this means they sometimes end up eating ‘the same old thing’.

Nevertheless, and in balance, home life has been altered and this is reflected in the titles of books on that subject matter – homes are no longer being thought of as smart with technologies taking over important decision-making activities. Instead, homes and their occupants are connected (Harper, Ed. 2011; also Neustaedter, et al, Eds, 2013). Meanwhile, the impact this move toward the connected home will have on the role of paper and the intersection of paper and digital has only been sketched. What is clear, though, is that just as it is with the workplace, the evolution of paper and digital in the home setting will require careful examination; it will not be straightforward to understand the combinations of motivations, attitudes and the ‘affordances’ of digital and paper media that will shape the future. Scientific investigation into the home is required just as much as it is with regard to the workplace.

 

References: Papers and articles

Frey, C.B. Osborne, M.A (2013), The Future of Employment: How susceptible are Jobs to Computerisation? Oxford Martin Institute, Oxford.

The Royal Mail, (2015), The Private Life of Mail.

Harper, R. (2010), Texture: Human Expression in the Age of Communications Overload, MIT Press, London and Boston

Harper, R. (Ed) (2003), Inside the Smart Home: Interdisciplinary perspectives on the design and shaping of domestic computing, Springer Verlag, Godalming and Heidleburg.

Harper, R. (Ed) (2011), The Connected Home, Springer, London.

Harper, R., Randall, D. & Sharrock, W., ( 2016) Choice: The science of reason in the 21st Century: a critical assessment, Polity Press, Cambridge.

Neustaedter, C. Harrison, S. & Sellen, Eds, (2013) Connecting Families, Springer, London.

Sellen, A. & Harper, R. (2002). The Myth of the Paperless Office, MIT Press, Boston, Mass.

 

Footnotes to Web references

1) http://authorearnings.com/report/print-vs-digital-report/

2) http://www.theguardian.com/media/2014/aug/14/uk-consumer-magazines-print-sales-2014).

3) http://www.printweek.com/print-week/feature/1156257/alpha-mail-print-is-retaking-ground-in-the-marketing-mix?utm_content=&utm_campaign=070316PWPrintBuying&utm_source=PrintWeek&utm_medium=adestra_email&utm_term=http://www.printweek.com/print-week/feature/1156257/alpha-mail-print-is-retaking-ground-in-the-marketing-mix

4) http://www.computerworld.com/article/2511279/computer-hardware/business-paper-use-remains-high.html

Is the internet making our social lives more dynamic?

Introduction

Along with two former colleagues of mine, Lynne Hamill and Nigel Gilbert (both of the University of Surrey), I have been working on some research into friendship and the internet. The interest  has been in sociology of friendship and the deployment of new computational tools  for investigating that topic. This research is suggesting some surprising things, so much so that I thought I might share some of that here: what is presented below is a shortened version of a paper we hope to have accepted in the British Journal of Sociology.

A sociology of Friendship

Though one might imagine friendship is a concern for psychology perhaps more than sociology, friendship has in fact always been a central theme. Webs of connection made around friendship were said by Simmel (1922/1955), for example, to be both the consequence of institutional and professional bonds and the source of those bonds. In his view, work affiliation could lead to intimacy, and intimacy could be the motivation to join professional and organisational groupings.  Given this, the term friendship is at once a label that distinguishes those who are friends from those who are not and a label for a connection that leads to action. More recently, reknown British sociologists Pahl and Pevalin (2005) use longitudinal data to affirm that this is a better way of thinking about friendship; they explain that friendship is both a categorization of a relation in time and something that evolves through time. While Simmel was interested in the move from friendship to institutional relationship, and in how webs of sociality lead to webs of economy (and thus ultimately with the formal properties of social relations), Pahl and Pevalin focus on emotional connections, not economic ones. Moreover they are interested in both the making and the breaking of relationships: how individuals start as acquaintances, and then gradually, with the passing of time, become friends; and how friendships can weaken with the passing of time and the shock of life events.

Friendships can also be viewed in another way: the relations can be thought of as a social network that is “fluid, shifting” (Boissevain, 1974: 48). Confirming this, Grossetti (2005) demonstrated that there is “a constant turnover” in personal relationships, developing from family at birth through to friends at school, and then changing as co-workers and neighbours come and go in adulthood. In this respect, social networks are affected by social mobility, not in the sense of movement in economic class, but insofar as people experience changes in their social context and geographic location. Key life stage events, such as marriage, cause perturbations affecting both the size and structure of the network (see for example Kalmijn, 2003; Wellman et al., 1997). However, kin relationships are more likely to persist over time than relationships with non-kin, even if contact is infrequent.

People have few friends compared to the number of people around them; that is, social networks are of low density, despite the fact that most modern life is spent in urban settings where people are in constant close proximity.  Many studies show that physical proximity increases the likelihood of social closeness, especially with non-kin (Heider, 1958: 188-189; Fischer, 1982; Cummings et al, 2006; Mok et al., 2007).  In 2000 just over half of British adults had close relatives living nearby and three quarters had nearby close friends.  Many of these were seen daily, suggesting that geographical nearness was a property of the relations in question, for otherwise this frequency of contact would not have been possible (Coulthard et al., 2002: 54). Being together is what friends do, it would appear, even if the social geography in which these friendships occur is one that is populated by many strangers, that is, in Simmel’s anonymous modernity.

Communications technology and human connection

The way these physically close connections with friends and kin are maintained is less well understood. What is sure is that the frequency of face-to-face meetings falls dramatically with increasing distance (Smoreda and Thomas, 2001; Quan-Haase and Wellman, 2002: 305; Licoppe and Smoreda, 2005; Larsen et al., 2006: 112; Frei and Axhausen, 2009). Nevertheless, it would seem obvious that technologies that enable some amelioration of the effect of distance will affect social networks – even if how they actually do that is manifold, and even sometimes opaque.

New transport and communications technologies have enabled people to interact over increasing distance. But, those interactions are diverse and subtle. Roads not only allow more frequent visits but also allow speedier sending of the gifts of friendship; postal systems deliver content but also help create a cultural sensibility to make social bonds through the written word (Henkin, 2007). Telephones do not just allow voice to be conveyed over distance, but foster the desire to chit-chat and thus make friendship in new ways (Fisher, 1992). Research on the impact of communication technologies shows that they can increase the strength of friendship connections in rather particular ways (Schiano et al., 2002; Boase, 2008).  The frequency of phone calls, fixed or mobile, becomes less frequent as distance increases, though they are of longer duration; but phones are important in maintaining friendships, especially strong ones, regardless of the frequency or ease with which face-to-face meetings can occur (Wellman, 1996; Wellman et al., 1997; Cummings et al., 2002; Quan-Haase and Wellman, 2002: 305; Coulthard et al., 2002; Licoppe and Smoreda, 2005; Larsen et al., 2006: 112).  And Carrasco et al. (2008) noted the importance of email in maintaining contact in a way not facilitated by phones.

The internet is the most recent technology to affect the process of friendship. Early studies, in the 1990s, used rather simple measures that suggested that the more time people spent on the internet, the fewer friendships they had, because spending time on the internet was treated as an alternative to investing time in friendships. This led to the formulation of the so-called ‘internet paradox’, the inverse relationship between time spent on the internet and friendship (Kraut et al., 1998). However, social networking technologies were then less advanced and less widely used than they are today, and internet behaviour often entailed playing very crude online games where little communication with other players was possible. It is hardly surprising therefore that the internet paradox was refuted by the same researchers a few years later (Kraut et al., 2002) when new social networking applications began to appear. By this time motives for using the web had altered too. The later research suggested that internet interaction helped foster friendships across the board.

Another set of researchers drew a distinction between types of experience people have with one another and the friendships that resulted. Friendships deepen and sustain themselves when ‘quality time’ is invested in them, these researchers asserted (Nie et al., 2000). In this view, good friendships exist when people spend time together. Other forms of connection, in which the parties are physically apart, were less rich and hence less consequential. Accordingly, use of internet-enabled techniques to communicate across distance could undermine friendship if that were the primary mode of contact, especially if it led people to spend less ‘quality time’ with each other. This research showed some concern with the patterns of friendship and friendship networks through time, although this was implied rather than researched. A greater concern was revisiting the internet paradox argument.

However, a growing body of literature has emerged that suggests that the impact of the internet on friendship is related to social type: people who are more sociable online are more sociable offline too (Di Gennaro and Dutton, 2007; Wang and Wellman, 2010). Those who do not make use of connections online are also more likely to have few friends offline (Dutton et al., 2009: 5). This research suggests some of the reasons why friendship networks vary in size and in density, with some people having consistently more friends on the internet through time than others: it is because they would have more friends whatever the technological infrastructure at hand. This infrastructure eases the work of ‘keeping in touch’, allowing those who have a propensity to leverage such opportunities to do so, while leaving those with less inclination to do otherwise. Claims about the internet paradox have come to be seen as somewhat orthogonal to these (and indeed other) sorts of questions.

Investigating the impact of the internet

One such question is the topic of this paper: how has the duration of friendship ties been affected by the coming of the internet? Answering this question poses some difficulty, however. Despite the increasing sophistication of these debates it is becoming apparent that the standard sociological data typically evoked to explain social action –gender, age, income, education – when combined with such things as internet access times and site usage, are not sufficient to analyse the ways that friendship is being shaped by this technology. Nor are these data rich enough to explore how the technology in turn is being shaped by friendship (Di Gennero and Dutton, 2007). Other factors need to be uncovered.

Solutions to these concerns may be at hand, however. New kinds of data are being made available by the internet beyond the enervating counts of access volumes and duration that have been hitherto relied upon. As Savage & Burrows (2007) note,  social network data can provide opportunities not just for researching the scale of friendship but for a whole host of sociological topics including ‘points of view’ within capitalist society (2007: 891).  While agreeing, we would add that the kinds of evidence that are being garnered through analysis of social connections made through services like Facebook does not suggest that the essential material of sociological inquiry is altering as much as might have been hoped. For example, Ellison et al. (2007) note that there is a strong link between the extant social capital that people bring to bear when they engage with others through social networking sites (SNS) and the duration of that social capital. SNS increase the lifecycle of human connection.  On the other hand, Henson et al.’s research (2010) is uncovering new forms of sociality and social identity, and they bring to bear huge aggregates of data to support their analysis. These data say little about the experience or process of friendship however, being more allied to the question of civic role in the age of networked technical support.

Discerning new characteristics in internet-mediated human friendship is not easy. As yet, no clear and comprehensive patterning governing how friendships are made, sustained, or come to wither on SNS  and other forms of mediated connection has been found (for a review see Author B, especially chapter 4). The relation between modes of contact and the processual character of friendship has also not been completely researched. It is easy to point out that more new connections are made via SNS than via other more traditional modes (Di Gennero and Dutton, 2007), but what happens thereafter is less well understood.  Although some years ago Urry (2003) implied that there might be a natural prosody to how often people would need to meet face-to-face in order to sustain close connection, more recent research shows that no such clear cut distinctions can be made (as I show, in my book Texture, 2010). Different modalities of communication afford different opportunities and constraints and people appropriate these in various ways, sometimes resisting and altering those affordances to use the technology in new ways (Papacharissi, 2011: 304-318). As Sosik et al. (2011) illustrate: although Facebook only affords asynchronous and primarily textual modalities of expression, these limitations do not weaken friendships. Users put effort into making their acts of communication within Facebook more adroit and powerful because of these limits. There is still much to learn about how different sorts of communication media affect the process of friendship.

This brings us back to Savage and Burrows. They propose that sociology should invite new methodologies and tools. Lynne, Nigel and I all agree that concerns deriving from apparently  premature judgements about internet-mediated changes on social connection, common agreement about the limitation of current data taxonomies, and deficiencies in understanding the relation between the internet and other technologies, lead us to suggest that one new method is especially worthy of investigation . Though it does not transform the source of sociological data (something that drew the attention of Savage and Burrows) this method uses computational techniques to treat data in novel ways. This technique is computational agent-based modelling.

Agent-based Modelling

This kind of modelling is not an alternative to the traditional sociological methods of observation, interview and survey, nor of those new sources of data that Savage and Burrows propose. Indeed, without such data collection, modelling of any kind would be impossible. Agent-based modelling is rather a way of consolidating the data that are available, and can bring together the qualitative and the quantitative in ways that were not possible before.

A major limitation of the ‘traditional’ qualitative and quantitative sociological studies is that they generate data that represent essentially static moments in social processes: they show a snapshot at one point in time. Longitudinal studies lasting over several years are rare, but again offer links between what are essentially static points. Agent-based modelling, meanwhile, not only captures the outcomes of process, but presents those processes as inspectable phenomena, insofar as investigators can alter the variables so as to test the adequacy of the model against various known or certain data samples.

Relatedly, the act of building an agent model itself can help investigators think about a problem and clarify their own hypothesis or motivating questions. Adjusting the model can expose implicit assumptions that might not otherwise have been appreciated, can identify variables that had not been considered, and can even raise questions of definition about the form or dynamics of relationships. All of this can help investigators better assess the relative importance of various factors suggested by more traditional forms of data gathering and theory.

Furthermore, modelling can be used to test theories about dynamic social processes by facilitating experimentation that for practical or ethical reasons is impossible to conduct in any other way. Modelling permits researchers to address ‘what if’ questions that simply cannot be addressed by any other means.

To sum up, following Epstein (2008), there are four key reasons to model:

•to test theories of explanation;

•to explore dynamics;

•to formulate questions (and thereby guide data collection);

•to examine possible outcomes.

There are of course many different types of modelling. However, the newly emerging computational agent-based modelling has two characteristics that seem especially useful in relation to the impact of the internet on the duration of friendships. First, it is good at tracing out the dynamics of social relations such as the processual concerns in relation to friendship, for example.

Second, agent-based modelling facilitates experimentation, allowing tests of the importance of different factors. As mentioned above, recent research about the internet has shown that an increasing number of diverse factors are important. Understanding of the relations of these data is often inadequate. Agent-based modelling can help test which factors would seem to be most likely to explain the emerging evidence about internet use and its relation to friendship, and it can do so with the limited data that is available. As a case in point, Casilli and Tubaro (2010) combine ethnographic data about friendship enabled through Facebook with agent-based modelling to explore how different types of individual action can affect the overall macro structure of a social netwo Friendship is one of the most profoundly subjective of experiences. But friendship also has what Simmel called ‘formal properties’: patterned dimensions manifest at an objective level. This paper has investigated one of these properties: namely the duration of friendship. In particular, it has addressed the question: how has the duration of friendship ties been affected by the coming of the internet? Because of the problems of using traditional sociological techniques to answer this question, this paper has used agent-based modelling. This modelling provides a quantitative assessment based on a set of plausible, consistent assumptions that can be varied thus permitting experimentation. It provides both a framework for answering some questions and a method to investigate concerns that cannot be addressed by more conventional means.

What do we find?

On the basis of the assumptions we made, this model suggests that the internet is unlikely to increase the number of core friends, but it may make these relationships more stable. The model indicates that:

Between 1998 and 2009, the average number of core friends had increased from 5.4 to 5.6. Although these averages suggest little overall change, some individuals have been affected significantly. By 2009, almost 1 in 5 had at least one friend in their core network who would not have been had there been no internet effect: 1 in 25 had more than one such friend.

Onliners were more likely to maintain at least one core member over the 11 years. Without the internet, 40 per cent would have had a persistence rate of zero i.e. none of the members of their original core network would have still been there at the end of the period; while with the internet, this fell to 36 per cent. Although the internet has had little effect on the average persistence rate of core ties so far, in the longer term it could affect it significantly, even raising it to as much as two thirds.

In sum, the modelling suggests that the number of friends that people might call close, or intimate, will likely remain fairly similar with the internet. This may seem surprising, given the hyperbole that often goes with discussion of the internet suggestive that social ties are weakening, that historical stability is being replaced by social fluidity (See for example Bauman 2005). What the internet appears to do is slow the pace of change, so friendships last longer. The oft-heard idea that the internet is creating change would seem to be egregious if by that is meant change in sociality. Our modelling suggests the reverse: the internet creates more stability through time. If this is so, such stability may help to assuage the loneliness of modernity that Simmel describes. One may also reflect on Giddens’ explorations of the reflexivity of identity, and his view that people have to negotiate who they are through the myriad relations they form with others (1991). Our model suggests that this might not be such a burden as Giddens implies.

Relatedly, the modelling suggests that the importance of geography is reducing with the internet, but again, not greatly. Certainly one might say that friendships sustain themselves longer when geographical distance is increased. This is evidenced by Dutton and Blank’s (2011: 38) finding that the internet has increased contact with friends and family who live further away. But being near still counts. Again, how does this relate to the claims about the move away from the door-to-door society? Cairncross’s book (2001), The Death of Distance is evidently offering an erroneous but commonplace view.

And this in turn suggests that the internet is not simply a means of making, keeping and moving on from contacts. It, rather, affords particular forms of sociality. This sheds a different light on the arguments about the internet paradox. These suggested that there is a difference between the kind of relation enabled by face-to-face and by internet-mediated connections. It was proposed that there is a contrast between the ‘quality time’ delivered by bodies being co-proximate as against the weak and anodyne bonds made through the keyboard. What our modelling suggests, in contrast, is that the links made possible through the internet are as vital as any other, but that they may be of another kind. And what we are thinking of here is not a distinction between, say, the virtual and the real. We need to distinguish those who can foster human connection, whether it is mediated or not, and those who do not or cannot foster such mediated connection. It is Bourdieau’s habitus that is evoked, not arguments about space, time or volume of connection (see Mistzal, 1996: especially 102-156). This term is now somewhat old, but perhaps it could be brought up to date with the protocols of the internet social network site as its mis en scène. It is De Certeau’s ‘practice of everyday life’ (1984) when that entails routine use of Facebook. What is required is a move from observing behaviours in Paris to observing them on the world wide web of activity, the habitus enabled by Palo Alto.

Bibliography

Bauman, Z (2005) Liquid Life, Cambridge, Polity Press.

Barnard A (2009) The effect of taxes and benefits on household income, 2007/. Available at: http://www.statistics.gov.uk/downloads/theme_social/Taxes-Benefits-2007-2008/Taxes_benefits_0708.pdf

Boase J (2008) Personal networks and the personal communication system. Information, Communication & Society. 11(4): 490-508.

Boissevain J (1974) Friends of Friends: Networks, Manipulators and Coalitions. Oxford: Basil Blackwell.

Cairncross F (2001) The Death of Distance 2.0: How the Communications Revolution will Change Our Lives. London: Texere Publishing.

Carrasco J-A, Hogan B, Wellman B and Miller EJ (2008) Agency in Social Activities Interactions: The Role of Social Networks in Time and Space. Royal Dutch Geographical Society, 562- 583 Available at: http://homes.chass.utoronto.ca/~wellman/netlab/PUBLICATIONS/_frames.html

Casilli A and Tubaro P (2010) Légitimation intersubjective de la présence en ligne et formation de réseaux sociaux: une approche ethno-computationnelle. Available at: http://www.bodyspacesociety.eu/wp-content/uploads/2010/03/CasilliTubaro_Article_RT26.pdf

Coulthard M, Walker A, and Morgan A (2002) People’s perceptions of their neighbourhood and community involvement: Results from the social capital module of the General Household Survey 2000. Office for National Statistics. London: The Stationery Office.

Cummings J, Butler B, and Kraut R (2002) The quality of online social relationships. Communications of the ACM 4(7): 103-108. New York: ACM Press.

Cummings J, Lee J, and Kraut R (2006) Communication technology and Friendship During the Transition from High School to College. In Kraut R, Brynin M, Kiesler S (eds) Computer, Phones and the Internet. Oxford: Oxford University Press: 265-278.

De Certeau M (1984) The practice of everyday life. Berkeley: University of California Press.

Di Gennaro C and Dutton WH (2007) Reconfiguring friendships: social relationships and the internet. Information, Communication & Society 10(5): 591-618.

Dutton WH and Blank G (2011) Next Generation Users: The Internet in Britain. Oxford Internet Survey 2011. Oxford Internet Institute, University of Oxford. Available at: http://www.oii.ox.ac.uk/publications/oxis2011_report.pdf.

Dutton WH, Helsper E, and Gerber M (2009) The Internet in Britain 2009. Oxford Internet Institute. Available at: http://www.oii.ox.ac.uk/publications/oxIS2009_Report.pdf.

Ellison NB, Steinfgiedl C and Lampe, C (2007) The Benefits of Facebook “Friends:” Social Capital and College Students’ Use of Online Social Network Sites, Journal of Computer-Mediated Communication. 12, pp1143-1168.Epstein JM (2008).

Why Model? Journal of Artificial Societies and Social Simulation. 4(12).  Available at: http://jasss.soc.surrey.ac.uk/11/4/12.html

Fischer CS (1982) To Dwell Among Friends. Chicago: University of Chicago Press.

Frei A and Axhausen K (2009) Modelling the Frequency of Contacts in a Shrunken World. Available at: http://www.ivt.ethz.ch/vpl/publications/reports/ab532.pdf

Giddens A (1991) Modernity and Self-identity. Oxford: Polity Press.

Grossetti M (2005) Where do Social Relations come from? A Study of Personal Networks in the Toulouse area of France. Social Networks 27(4): 289-300.

Heider F (1958) The Psychology of Interpersonal Relations. New York: Wiley.

Helsper E (2008) Digital Inclusion. Department of Communities and Local Government. Available at: http://www.communities.gov.uk/documents/communities/pdf/digitalinclusionanalysis

Henkin DM (2007) The Postal Age: The Emergence of Modern Communications in Nineteenth Century America. Chicago: University of Chicago Press.

Kalmijn M (2003) Shared friendship networks and the life course. Social Networks. 25: 231-249.

Kraut RE, Patterson M, Lundmark V,  Kiesler S, Mukhopadhyay T, and Schlerlis W (1998) Internet paradox: A social technology that reduces social involvement and psychological well-being? American Psychologist 53(9): 1017-1032.

Kraut RE, Kiesler S, Boneva B, Cummings J, Helgeso V, and Crawford A (2002) Internet Paradox Revisited. Journal of Social Issues. 58(1): 49-74.

Larsen J, Urry J and Axhausen KW (2006) Social networks and future mobilities. Report to the Horizons Programme of the Department for Transport, , University of Lancaster & ETH Zürich. Available at: http://www.ivt.ethz.ch/vpl/publications/reports/ab330.pdf.

Licoppe C and Smoreda Z (2005) Are social networks technologically embedded? How networks are changing today with changes in communications technology. Social Networks, 27: 317-335.

Hansen D, Shneiderman B and Smith MA (2011) Analyzing Social Media Networks with NodeXL, Elsevier  Burlington,  MA.

Mistzal BA (1996) Trust in Modern Societies: The search for the basis of social order. Cambridge: Polity Press.

Mok D, Wellman B and Basu R (2007) Did Distance Matter Before the Internet? Interpersonal Contact and Support in the 1970s. Social Networks. 29(3): 430-461.

Nie N, Hillygus S and Erbring, L (2000) Internet use: interpersonal relations and sociality. In Wellman B, Haythornwaite, C (eds) The Internet and Everyday Life, Blackwell, Oxford: 215-45.

Pahl R and Pevalin D (2005) Between family and friends: a longitudinal study of friendship choice, The British Journal of Sociology, 56(3): 433-450.

Papacharissi Z (Ed) (2011) A Networked Self: Identity, Community and Culture on Social Network Sites, London: Routledge.

Quan-Haase A and Wellman B (2002) Capitalizing on the Net. In Wellman B,  Haythornwaite C (eds) The Internet in Everyday Life. Oxford: Blackwell: 291-324.

Savage M, Burrows R (2007) The Coming Crisis of Empirical Sociology, in Sociology, 41: 885-898.

Schiano DJ, Chen EP, Ginsberg J, Gretarsdottir U and Huddleston M (2002) Teen use of messaging media. Conference on Human Factors in Computing Systems CHI ’02 extended abstracts on Human Factors in Computing Systems, Interactive Posters. Minneapolis, 20-25 April. New York: ACM Press, 594–595.

Simmel G (1922/1955) Conflict and The Web of Group-Affiliations. Translated by Wolff, KH and Bendix R. London: Free Press-Macmillan.

Smoreda Z and Thomas F (2001) Social Networks and Residential ICT Adoption and Use. COST269: EURESCOM Summit. Heidelberg.

Sosik V, Zhao X, and Cosley D (2012) See Friendship, Sort of: How Conversation and Digital Traces Support Reflection on Friendships, in Proceedings of CSCW 2012, ACM Press: 1145-1154.

Urry J (2003) Social Networks, Travel and Talk, British Journal of Sociology,  54(2): 155–175.

Wang H and Wellman B (2010) Social Connectivity in America: Changes in Adult Friendship Network Size from 2002 to 2007. American Behavioral Scientist, 53 (8): 1148-69 Available at: http://www.chass.utoronto.ca/~wellman/publications/social-connectivity/social-connectivity.pdf.

Wellman B (1996) Are personal communities local? A Dumptarian reconsideration. Social Networks. 18: 247-354.

Wellman B, Wong RY, Tindall D, and Nazer N (1997) A decade of network change: turnover, persistence and stability in personal communities. Social Networks. 19(1): 27-50.

The concept of communications overload and the sociology of mobile communications

I prepared this piece as a sketch of one of the arguments in my book Texture (MIT, 2011) and, since it is brief, I think it is worth sharing. Besides, a version of the following is to appear in the Sage journal ‘Mobile Media & Communication’, full citation to follow, and that journal may well be of interest too.

http://www.sagepub.com/journalsProdDesc.nav?prodId=Journal202140

 

April 2012

Our lives are busy. All of us are busy. Too busy, we all complain. And then there is communication, so much of it indeed that, with everything else, we are overloaded. ‘Enough Already!’ one can hear us all exclaim. One can hardly list the numerous self-help books written to deal with this dilemma.

So what is it that is making us so busy? What one can say is that it is not work that makes us so frenzied and rushed. Historians show that we give less time to work than ever. Today, most of the time people have is given over to things other than work. Some of these things haven’t altered with the centuries. People have to sleep, for example, and this takes up more time than almost anything else (about 37% of a day, according to some measures). They also have to eat and clean (another 9%). These are body related things, biological needs one might say. There are also needs related to society, ones that one might imagine alter over time, but this is not always the case: people spend as much of their lives travelling to and from work as they did a hundred years ago (about 6%). And then there are the contemporary things that we do: fiddling with our Facebook status updates, reading Twitter; emailing. What is sure is that the time allocated for these sorts of things is increasing, though the figures that might enable us to judge by just how much and with what velocity of change are difficult to interpret. There are gross figures for communicating that suggest that 8% of the day is given to it but these don’t allow for multitasking. Apparently, these time-measuring statistics are based on the assumption that individuals don’t do more than one thing at a time – or rather they have to assume this since their data is somewhat rough: multitasking is invisible in it. It is no wonder, then, that when one adds up all the time consumed by such measures, the doings of the day take longer than 24 hours (for discussion of these numbers see Harper, 2011, p37-45).

One can readily understand the methodological difficulties in this area – who wants to count the seconds one spends texting while one is watching the TV? If one leaves these difficulties aside, however, and simply looks at the overall balance of time, one will see a paradox: one sees that, when all the things that have to be done are done, the sleeping, the eating, the cleaning, the going to work (even though we spend less time at work than we used to), there isn’t much time left in an ordinary day for anything. Those activities we complain about as especially onerous time consumers – as filling our days up – don’t actually consume much time for the simple reason that there isn’t much time for them to consume. Here I am thinking of communicating: are our complaints about it overloading us really only about 8% of the day?

So, while it seems a truism to say we are currently burdened by overload – who would disagree? – careful consideration might lead us to discover that we don’t, in fact, really give much time to communicating, not to mediated communicating, the expressions that require some kind of transport in their functioning. Indeed, we don’t seem to give much to anything at all, apart from the remarkable amount of time we give to things we don’t really have a chance to alter – to the need of our own bodies for rest, for their upkeep, and presentation (eating, washing and dressing). Yet it seems to me that, and I admit this seems contrary, this doesn’t mean we are wrong to complain about communications overload. Though it is certainly the case that we might muddle things up, confusing the burdens of communications with the burden of all of things we have to do, we do indeed worry about how much we put into communication and for good reason. But time measures are not the issue here. Or, rather, they are not a helpful way of approaching what the grounds for our complaints and concerns might be.

It seems to me that when the phrase communication overload is used, we should assume that what is being alluded to is simply a quantitative phenomenon. Quantitative measures are sometimes useful, but only occasionally and even then have to be treated with care. Certainly, the way we ordinarily talk about some kinds of communication does imply the consumption of time, and sometimes too much time: ‘They do go on’, one can hear people say about another’s conversation; ‘Their emails are so prolix’ in another instance. And sometimes the use of the phrase alludes to the exchange of stuff, such that the more there is (of the stuff) the better it is for us involved in the communicating but by the same token also implying that sometimes there might be too much exchanging. Gossip and gossiping comes to mind. Here quantification might help. But these are instances of the many and varied things that acts of communication constitute and there are many other acts where such metrics don’t help (or fit).

There is a bigger issue afoot, however: we are starting from the wrong place. Instead of thinking of the phrase communication overload as merely empirical, and strictly arithmetical at that, we need to think of it as first and foremost a concept, as a tool used by people, a mechanism that lets them understand their world through describing it. Treating it this way will lead us to look at all its uses, and stop us confining our interest to just one such use (one that implies counting say). It will let us to unpack the many properties the concept has, the things implied when it is used, the things assumed in that use, the links it has to other concepts. It will let us come to understand what it helps people do, how it describes, how it organises their affairs. The view I am taking is Peter Winch’s interpretation (1958) of Wittgenstein rather than the approach offered by Strawson’s (1992) analytic philosophy; it is to show a concern for the forms of life that concept are related to and not merely their logical form.

One of the first properties one might want to highlight from this point of view is that the concept ‘communications overload’ is not very accurate or specific. It labels lots of things. When people complain of communications overload they are using a catch-all concept. I have already listed some examples of the things it labels. Here are some more: when people ‘communicate’ to other people they are showing, for example, a sensibility for family life. Sensibility seems some way from communication so what am I getting at? I am alluding to the fact that one of the things one does in families is listen, listen to one another irrespective of whether one has any interest in the topic at hand. Family life is about chatting, amongst other things. In another instance, people communicate because they are delighting in friendship: words might be exchanged, ideas remarked upon, but it is being with another that is at issue: that is what friendship can sometimes entail. In a third, when someone posts an update on an SNS, they are seeking ways of characterising themselves.

So, when people say ‘I am overloaded’ just what is it that they are saying? Do they mean to confine their remarks to the specific acts of communication, or to the point of a communication? Is about family life they are whinging, or are they thinking of friendship, or are they thinking about how little time they have to paint a portrait of themselves on their Facebook account? In other words, empirical referent is one concern that has to be treated carefully when one considers the concept.

This is not the only property of the concept that is important. Another has to do with the fact that when the term is used it refers to uniquely human affairs – this sounds tautological, since all the examples above are of human activities. What I mean is that the concept is used in a way that turns on the assumption that it is the particular and peculiar properties of human affairs that are at issue. These can be summed up by the term morally implicative. Human communication is always about what consequences individual acts of communication have for the relations between people, the character of those relations and their nature.

I cannot overemphasise the importance of this nor the delicate complexity of this fact. Consider some examples from my research. I have looked why old folks think communications acts say something about them and their relations, for example. They treat such acts as sets of doings that are judged and oriented to in particular ways. I have found that old folks condemn broadcast messaging (such as posting on Facebook) because it shows little deference for the individual recipient of a message, for the singularity of friendship. They prefer longer letters or emails. Thus the moral implications of acts of communication are central to how such acts are understood, selected, avoided, counted, ignored.

One example might seem insufficient to prove of my claim. Let me elaborate on another example which shows how delicate, complex but nevertheless moral are the consequences of communications. (All these are taken from my book Texture, 2011). I have mentioned above the sensibility that family life requires. In my studies of technologies like Whereabouts Clocks I have found that people use that technology to finesse their acts of family tenderness and affection – using the clock to know when to make tea just before someone comes home or to let them be more aware of where partners have been and thus better able to make small talk with those partners when they come home and need to unwind. The richness of communication acts are, then, great indeed; but this richness points towards the richness of social relations, their properties, their patterning through time: in commentaries and analyses we tend to sterilise them or at least offer descriptions that seem to lose the tendernesses they entail, the thoughtfulness they enable.

And here is the rub. If this is so and if this moral implicativeness is so delicate and complex, so rich and so vital to human affairs, so easy to misrepresent, so hard to characterise, what is the research agenda that is appropriate for this space? Or is it simple care that is demanded? Care is obviously requisite. But we still have much to learn. Despite many years of effort in mobile communications and media research, we are confronted by a number of difficulties.

They ensue when we rush too quickly to explanation before we have uncovered what the acts in question are essentially about or properly understood the concept used in relation to them. The examples above have been selected to show how there is an obvious link between the concept of overload and counting but how, at the same time, when people use the phrase overload, they might in fact be pointing towards the moral aspect of messaging. It’s not the counting of the messages that might be at issue, it might be friendship, or it might be family affection. If so, then the word overload might be a synonym for guilty indifference – in both contexts this would apply though the resonance in each would be different. A marriage partner is not the same as a friend, after all.

How does one examine this topic (or topics)? Can one count these forms of guilt? To be sure we can if we treat it in certain ways, asking people to give a number to their feelings, for example, surveying a population to see how many of them ‘feel guilt’. But doing so can obscure important properties of that emotion – especially when the concept is used, as in this illustration, in relation to friendship and –or family life. Take friendship: is it to be conceived of in terms of scale? Frequency of contact? Or its reverse: infrequency? Such a model seems calculative. One would not savour a friend who counted in this fashion. One wants them to feel – to feel for us: that is what friendship ought to be. All the more so in family life,  one might say.

So, does that mean that feeling is really the issue, the one that should be researched when people complain of overload? No,  of course not. I think the real task of research is to show how topics like this are linked to and made visible by talk about communication. The problem is to preserve that complexity, its delicate form, despite our desire to simplify, theorise, and distil – even the words we use can handicap us. Consider, I used the word essential: this implies that something is at the core of something, when all I meant was that one concern is more important that others, more salient perhaps, more at the heart of things and it is that which we are after, not platonic essentials.

Take another example of how easily our research can lack finesse. One thing obviously associated with the word communication is time, but not simply clock time. The counting of time hardly does justice to human experience. The full prism of social connection is manifest in how social connections are at once prospective and retrospective, in the here and now and sensed as things to be done and things that were done. Individual acts of communication need to be understood in terms of the topography of connection through time. But how does one research that? It might seem easy, merely a question of empirical traces. Let me return now to the question of guilt. Traces might not be the issue. An interest in communication, with acts of communication, can lead one to ignore those social relations that don’t appear to entail any act. If it is the case, as I mention above, that one can sometimes feel guilty about not responding to a communication, or acting appropriately as a result of one, there are some instances, some occasions, where the absence of such acts – of any kind – doesn’t foster guilt. Quite the reverse: the absence of contact is viewed as good behaviour. As Simmel noted in his essay, How is Society Possible? Part of the good grace that is required in modernity is the capacity on the part of every individual to ignore strangers they come to share space with as part of the contingencies and necessities of life. One sits next to strangers on the bus or the metro; one queue’s besides anonymous others in the shops; one laughs in the theatre along with people one doesn’t know; one gracefully looks away from the unidentifiable man making a mobile phone call in earshot. In our focus on the physical acts of communication, or in seeking traces of acts in the past and planned for ones in the future, we must avoid neglecting these silences and gentle looking aways, those moments when people chose as it were not to speak, to not communicate at all, even to glance, now, or indeed ever. Their choice to not act expresses in itself an alertness to the fact that they have no right to communicate to, for example, the present other. They choose not to communicate as a way of showing respect.

When we choose to communicate and when we chose not to, in other words, we are making many distinctions. Thus, when it comes to communications overload, we need to be clear about the distinction that pertains. When we assert ‘Enough Already!’ is it enough communication with those who have rights to communicate to us that we are thinking of? Is it their rights to demand of us a response to every act they make that irks us? Or are we thinking of the burdens we place on others, some of which might have been mistakes, when we sought communication with those we do not know and hitherto had no right to chat to, to call, to email, to text and thus found, by default as it were, that we had opened up a system of communication: first this, then that in an never ending circle? Are we guilty for the burden we have placed on others, one that can last for the eternity of our lives: ‘I introduced myself, I can hardly ignore him now’.

The distinctions I am drawing here are, of course, curiously one way: one cannot undo the rights to communicate, even to those one has fallen out with. For if we choose to ignore their messages, they know, as our mutual friends do, that our doing so is an act of enmity. We speak volumes by not answering. But thus it is that we need to be careful about who we allow to become part of our social world, who we allow in, for in the future their demands for a communication might burden us down.

In sum, when the phrase communication overload is used, so many things are meant, so many possibilities implied. It would be wrong to say that analysts of the age of communication should avoid the term themselves, seeking perhaps to come up with their own labels and measures. I do not think that the everyday use of the term is poor, or inadequate or confused. I do not think it needs replacing by something better. My point is that its use is incredible complex and subtle and adroit. We would do well to investigate those complexities with similar adroitness: it’s what people do with the term that matters, they why of it, the how of it, the texture that results. This is the business of the human act of communication. It should be ours, too, as analysts of that business.

References

R. Harper (2011) Texture: Human expression in the age of communications overload. Cambridge, Mass: MIT Press (book).

G. Simmel (1971) ‘How is Society Possible?’ in On individuality and social forms, Chicago: Chicago University Press (book).

P.F. Strawson (1992) Analysis and Metaphysics, Oxford: Oxford University Press (book).

P. Winch (1958) The idea of a Social Science and its relation to Philosophy, London: Routledge and Kegan Paul (book).

Culture versus modern analytic philosophy

I have been bringing together another edited collection, this time on philosophy, sociology and technology – in particular cloud technologies and their impact on the internet (prior collections have looked at the Future Home, at  SMS, and Wireless Connectivity, amongst other things: I have previously mentioned my most recent, on the Connected Home). The philosophy arguments have been drawing my attention not because I think that the latest arguments in philosophy are appealing; I don’t think they are: but they are current and people are espousing them: their oddness dismays me and I feel as if I ought to say something. ’Hey, that can’t be right,‘ I want to shout.

Let me be boring on this: the reason why I don’t find the appealing is because these philosophical arguments seem to distract attention from the enormously rich tools of culture we have in such a way as to diminish those tools, almost to discredit them. A current concern in philosophy, for example, has to do with how you explain, through experimental evidence, how people understand each other (Bratman’s work comes to mind as an example of this (1992) – though I think it mostly derives from Donald Davidson’s essay ‘Action, Reasons, and Causes’ of 1963). The starting premise of these enquiries is oftentimes the idea of free will, how this is to be accounted for, and how this in turn leads to questions about how the acting agent understands (and thus acts upon) other’s intentions; thus to collaboration, etc, the theme of Bratman.

Leaving aside the details about that, what I am suggesting is that the premise of freewill turns into enquiries about a sort of  solipsism (though this word isn’t used in the narratives in question -I am inclined to think that using that word or alluding to it would be too literary for current ‘empirical philosophers’).

Let me put it another way: some philosophers commence their thinking with the idea that people do not share a world on common; people are all essentially isolated (if not physically) – certainly without the gift of understanding the motives or purpose or aspirations of others. People are Cartesian entities, if you will, that exist without an intimate sense of others (and needless to say without a common God who can bring them together – there isn’t much religion in the mis-en-scène of contemporary philosophy).

Thus one will see already that the new breed of philosophers don’t look at what people do and experience in the real world – since having a sense of what others are about is the cornerstone of how we arrange our affairs even if we often do a bad job of it. Instead what lots of philosophers do is a kind of weird science: they put people in laboratories – Bratman is exemplar here – and set them tasks that they have to do together – playing a game, solving a puzzle – and then the philosopher’s watch what happens. The philosophers claim that through the evidence they garner thus they can discover how people come to understand each other.

So far it sounds sensible, if a bit odd: why put people in a peculiar setting to understand their normal affairs?

But now, hear this: the really curious thing is this (not merely the fact that philosophers think you can understand people by putting them in labs): it is that the philosophers transcribe what the subjects say to each other (in the experiments) and then ‘analyse’ those words. How bizarre: the philosophers are pretending that they cannot listen to what is being said by their subjects (i.e. as these poor compliant individuals who are getting on with the tasks they have been given and talk about it as they do); they cannot listen to them in an ordinary way but rather they, the philosophers, can only ‘understand’ what these ‘subjects achieve’ through a form of scientific magic, the mechanism of transcription and ‘analysis’. It is as if philosophers make other people, not themselves, other than human and somehow like Martian’s issuing forth grunts and animal calls that need to be ‘interpreted’. The philosopher’s art is to figure out how these sounds (words to you and me) point towards how people, these Alien- like animals, come to understand each other and ‘collaborate’. It seems to me that these philosophers think they are doing something like a cognitive science allied with an analytic philosophy but really they are corrupting understanding, the understanding that is constitutive of the world.

In my mind, these views draw attention away from the possibility that people already have tools that let them (us, you , me, all of us) understand each other. These tools have taken centuries to work out and cultivate. They are not the things one finds in laboratories or experiments.

The kinds of tools I am thinking of are big things and little things. As regards the latter: the kinds of things I am thinking of are, for example, little things like books, novels, poems, all of which are, in my mind, cultural devices that have been devised as ways to let people tell their stories, the tales as it were of their unique voyage, which they then tell (give?, share?) to others. These tales are, of course, both real and imagined.

Novels, poems, books of all sorts, are written around a different premise to the one that philosophers like Bratman start with. They are begun with the understanding that everyone lives in a world shared and known in common. It is also assumed that each person voyages in different places and in different ways and their nature (for want of a word, though this is powerful) colours the experience of this differently. Some philosophers agree with this view, this starting point: Cora Diamond is a case in point in her ‘The Difficulty of Reality and the Difficulty of Philosophy’ (2008).

Be that as it may, let me say more about nature. By that I am thinking of such things as character, personality, education, the stuff that makes us what we are: people. We all know that different people have different reasons for articulating their experiences; we know too that people have different abilities; some have eloquence and artfulness, and some insights that are unusual, even unique.

Amongst the reasons we read books, then, is to hear the tale of how it might have been for us but for circumstances, but also to read the insights of those who see more clearly things that we ourselves experience but can’t articulate as well (and sometimes don’t even see).

Books, poems and such like aren’t then a solution to solipsism or the skeptics dilemma (viz ‘are there others in the world?’); they are then devices to let people explore a world in common. The charm in such things is provided through discovering (learning, reading) that other’s voyages are more interesting, other’s views are more articulate, theirs deeper than one’s own. Otherness here is not a philosophical problem; it’s a poetic one: one that is interesting for the way it is articulated.

I read some literary theorist who put these views forward some weeks ago: he claimed it was Emerson’s view, but I don’t know enough about literary theory to be sure. But anyway  I wholeheartedly agree: I think this view is the correct one. Views articulated by philosophers who think we need to discover what others are about –  and who think they can discover how people ‘do this’  through looking at this experimentally- undermine the credit we should give tools like the novel and thereby also distract attention away from those tools to other modes of practice – ones that offer little on this subject – like experimental ones.

It is for these sorts of reasons that I read current philosophers and despair. I also get a bit annoyed since it seems to me that Wittgenstein more or less argued for a sensitivity to such cultural tools in his own later work, but that seems mostly forgotten in the corridors of philosophy now: Donaldson’s myopia is the rage; Bratman’s research simply the output of a distraction.

So there you go!

Bratman, M. (1992) Shared Cooperative Activity, The Philosophical Review, 101(2) pp 327-341.

Davidson, D. (1963) Actions, Reasons, and Causes’ in Davidson, Action & Events, OUP, 1980: 3-20

Diamond, C. (2008) ‘The Difficulty of Reality and the Difficulty of Philosophy’ in Cavell, S. et al, (Eds) Philosophy and Animal Life, Columbia University Press, New York

The Absence in Kinect

It seems to me that the message Kinect advertisements deliver emphasizes an absence; attention is drawn to how something is missing in the relationship between the human and the technology. It makes the claim that there is no ‘how’ when someone interacts with Kinect.  Indeed, the claim is stronger than this: if the publicity is to be believed, there is no interaction at all. This is because, as the adverts put it, with this technology, ‘You are the Controller’.

Obviously this is simple publicity; advertising does not pretend to be science. Nevertheless, there is much to consider in what is being emphasized in phrases such as this. What is being proposed is that, with Kinect, there is isomorphism between, on the one side, the human and what they want to do, and, on the other, the machine’s ability to understand, grasp, comprehend or, in this case more accurately, to ‘see’ what those intentions are and then respond to them. It is because Kinect can ‘comprehend’ these intentions without any intermediation, no keyboard, mouse-clicks or consul, that the publicity has this odd absence at its heart: it seeks to celebrate the removal of mechanical means for controlling computers.

Whereas once the computer mouse might have been the centre of an advertising campaign, now in a world where the technology can let you be the controller, all those sorts of technologies can go. This future holds less – less technology to get in the way. And thus, so the hype goes, we are allowed to be ‘natural’. All other means of interacting with a computer are, by dint of this phrase, implied to be somehow other than natural. In this view, the keyboard was merely a technological step on a path to this ‘natural’ mode; the computer mouse simply a cute technological aid that, like the keyboard itself, is soon to disappear, to be found in the future only in museums and histories of innovation.

So, what is one to make of the phrase ‘You are the controller’ and relatedly the term ‘natural? The trouble with both is that they are pregnant with numerous meanings and interpretations – it’s not that they are vague; it is that they are too evocative.

For example, the word natural is bound up with what its use implies as unnatural. Given that Kinect does away with a keyboard, one interpretation that goes with this usage could therefore be that people do not ‘naturally’ want to communicate by written words – the primary output of a keyboard. Doing so is an artifice of technology, one might think. Yet one only has to think of linguistics and one of their continual bugbears. As Baron laments in her book Alphabet to Email  linguists are regularly asked to explain to those with little knowledge of their discipline that the written word is not a surrogate for some other human activity that is ordinarily preferred or more ‘natural’. The other activity that is often invoked in these discussions is, of course, the spoken word. Speaking is natural, this vexingly wrong view holds, while writing is a forced necessity when speaking is not practical (see also my own book Texture 2010). As Baron explains, this is completely wrong; writing is a different social act than verbal communication; it is not a replacement or a proxy.

Consequently, when Kinect is said to offer users a more natural form of interaction what is it that is being contrasted? It would appear not to be the creation of text.  Perhaps then, the kind naturalness in question has to do with other tasks. One is sorely tempted to make a pun: Kinect is natural for certain kinds of interaction, when a person wants to a computer to do certain sorts of things, but evidently not for all things – so what then? Just what is the ‘natural domain’ of Kinect? Games? But is natural what comes to mind when one thinks of games? Surely artifice, rule-boundedness, the temporary abandonment of the natural and everyday attitude is what comes to the fore.

Similarly, what is one to make of the phrase ‘You are the controller’? Is a person not a controller when they use a keyboard and mouse? Does an individual somehow relinquish some control when they use these devices? Presumably that is not what is being implied when the phrase ‘you control’ is coined. Alternatively it might be immediacy is being implied: with Kinect one can get the machine to do as one wishes instantly, as quickly as one can move one’s hand: this brings to mind the idea of direct interaction, so beloved by some in HCI.

And then besides these concerns there is the bigger question of what is meant by ‘you’. For it is not the inner self that is being alluded to, assuming that is that there is such a thing. After all, one only needs to recall Turkel’s account of the joy that early programmers felt when they could see how what went on ‘inside their head’ could sometimes, with the right coding, appear to manifest itself externally, on the computer screen. This was certainly her thesis in The Second Self even if she has now recanted. No, what seems to be of concern here is the body; this is what is meant when the phrase you are the controller: for it is what one’s body does that acts as the mechanism to control the system. It’s one’s torso and limbs that do the controlling, not thoughts in the head. But just as this is so, then this also means that the social who, the status of the individual who plays, is therefore not part of the vocabulary of Kinect: bodies are not just thoughtless, they are without status too.

All of these questions highlight then the oddness of the phrase ‘you are the controller’ and relatedly the term ‘natural’. It seems to me that, though they might be odd, one ought not to think that they are therefore wrong; it is more probable that the terms need grounding in rather prosaic particularities. One imagines that what is natural needs to be understood and evoked in references to spaces where a user and their machine can see each other without interfering with other domestic activities, for example, in spaces where it is natural to game, say; one imagines that the phrase ‘you are the controller’ makes sense when one disregards certain aspects of the self, and privileges others.

The bottom line is that even though the interaction that Kinect enables might indeed be immensely appealing and may make users experience a new sense of control and wonder at how easy and ‘natural’ it seems to be, for those in research, ‘just what’ the experience of Kinect is, what it affords and how it can be designed around, needs careful investigation……

to report on this soon..