Carol Upadhya


Biography

Carol Upadhya is interested in theoretical and historical anthropology; economic development and social change in India; globalisation; entrepreneurship; middle class. She has widely published and also made three ethnographic films 'Coding Culture: Bangalore's Software Industry' in collaboration with Gautam Sonti (NIAS-IDPAD project 2006). With A R Vasavi she edited the revealing study '"In an Outpost of the Global Economy: Work and Workers in India's Information Technology Industry" (New Delhi: Routledge, 2008). In 1988 Upadhya obtained a Ph.D. in social-cultural anthropology from Yale University. She was Research Fellow, Indian Institute of Information Technology, Bangalore, 2002-2003, Reader in Sociology, Department of Post-Graduate Studies and Research, S.N.D.T. Women's University, Mumbai, 1997-2001 and Editorial Consultant for Economic and Political Weekly, Mumbai, 1994 - 1996. With Mario Rutten she edited the volume "Small Business Entrepreneurs in Asia and Europe: Towards a Comparative Perspective" (New Delhi: Sage Publications 1997). Carol Upadhya grew up in the Unites States of America and lived most of her adult live in India, where she married and raised her family.

1st December 2009, Bangalore, India

Carol Upadhya is considered to be one of the authorities on the culture of the software industry in India. From an anthropological perspective she has observed and interviewed many people involved. Therefore we were very happy that she was capable of meeting us at the National Institute of Advanced Studies. The National Institute of Advanced Studies (NIAS) was conceived and established by the vision and initiative of the late Mr.J.R.D Tata, who sought to create an institution which would conduct advanced research in multidisciplinary areas, and also serve as a forum that will bring together administrators and managers from industry and government, leaders in public affairs, eminent individuals in different walks of life, and the academic community in the natural and social sciences. We meet Dr. Upadhya at the end of the afternoon.

Summary

Dr. Carol Upadhya and colleagues have conducted a considerable number of interviews in India's IT industry in order to understand the culture and social economic structures of this industry. The fundamental aspect of the IT business in India is that it is mostly an outsourcing industry. The Indian companies, which work mostly for larger clients outside India, have developed the so-called 'Global Service Delivery' model in which they take advantage of various time zones around the world. Collaborations and work move from one place to another, so that 24 hours a day the work in India is going on. In addition to the big multinational companies operating in India, and the large Indian software services companies such as Infosys, there are also small start-up companies, so-called high end product companies. Some of these companies self-consciously replicate the Silicon Valley model of management. It is not that they are doing this replication, but they think of themselves as doing a kind of replication, which is to Carol Upadhya, being anthropologist, very interesting. Upadhya argues that the idea of cultural difference actually has become a mechanism for control over labour. She supports this argument in a variety of ways.

Big Indian outsourcing companies like WIPRO, Infosys, TCS have developed their own models and they have perfected outsourcing as a type of business. Partly as a matter of management and partly as a matter of marketing, the problems that these companies face is the 'Indian' label. It is only recently that India has gained a reputation in the global economy as good in software or good at IT. One has to first build a reputation to be able to market services in the global market. To be able to interface with the western clients, Upadhya noticed that on the one hand there is a kind of moulding to fit into the global market and also to fit to the specifics of the client side whether it is Europe, France, Germany or the US. On the other hand there is the notion of being Indian. When looking at the training programs and the way managers talk about this, somehow they have not made up their minds about what people are supposed to be. Workers have to be an Indian in certain contexts, and have to be like a European in some others and so on. Upadhya finds that it appears to be very confusing to be a software engineer in this context, yet at the same time Upadhya and colleagues observed in the many interviews they did, that in some sense it is not a problem because people have a clear sense of self in their private lives, which is not eroded by all the stuff that they are meant to do at work.

Only after a considerable amount of work in which trust is built through a rigorous monitoring and witnessing of every little step, companies and Indian centres of larger multinationals get more interesting high end contracts. Even though the reputation of the Indian IT industry has been established by now, most workflow processes are still orchestrated in a 'low trust' manner. In larger distributed projects there are online mechanisms, which record the work in different places and these statistics are fed back to the manager in packages. It is in a highly technical quantifiable way that projects are monitored. Very often on the client side the project is also being monitored every single day. The client does not just send an order to India and say come back in 6 months and show me what you have done. Every single day, nearly every hour the work of every person is monitored. This 'low trust' environment affects the people who work in it. Upadhya describes the dynamic as follows: "Because of cultural stereotypes it is implied that there is a certain level of incompetence that needs to be managed. So there is this whole bunch of things that always get said about Indian engineers, especially from the European perspective, which is where I did the work. One is that they are too subservient, they don't assert themselves, they don't take ownership, they're not taking initiative in doing, they are very passive. But on the other hand in a sense that's 'good' because they are 'very good at taking directions'. So you can give them this work and then just go on and let them do it, but you cannot expect innovation from them. So that is the way they are seen. But at the same time if that is the way they are seen they are almost expected to behave that way, right? Because they're in a situation where they're not given any responsibility. And if you talk with the Indian engineers one on one, they will never see themselves like this. They see themselves as being very intelligent, very capable and very much capable of doing much better work than what they're getting. But they're not getting that kind of work because there is this low trust situation." Nevertheless, Upadhya noticed, even though the machine is in many ways controlling the workflow and controlling the work process, people have ways of tricking the machine and they enjoy tricking the machine, getting around the system. Not logging what you're supposed to log or finding ways to fiddle the system so it doesn't log what you were doing. And these are computer engineers, so they can do stuff like that. There's a kind of subversion that happens also sometimes. People don't talk about it much, but Upadhya thinks it's very common.

In the IT industry people talk about 'becoming global', referring to a meaning of the word 'global' that somehow includes moulding behaviour into a model, which everybody can understand regardless of where they are from. But in spite of 'getting global', Upadhya finds that it really matters where a person comes from. In cross-border projects a lot of the problems that crop up are identified as communication gaps or as a cross-cultural problems by the people involved. In order to understand the problems they face, people talk in the language of differences in culture. Instead of understanding themselves as working in a common culture, the culture of IT maybe or digital culture or global information technology culture, they see themselves as being located in these very distinct traditional cultures. To Upadhya this shows that people, even when they are working in cyberspace most of the time, are in fact rooted in their localities. But at the same time she notices that in these dispersed teams and dispersed projects, people are not in equal positions. Upadhya finds that a system of hierarchy operates the system of control in this entire scenario.

When inquiring about the nature of the work, very few people are self-conscious and can articulate such things. By reading in between the lines, Upadhya got the notion that people think about their work as very much driven by machines, as very mechanical. Software production is a very mechanical process around the world, not only in India. They do not see themselves as being adjunct to the machine, but certainly they see themselves being stuck in a way in a very routine factory type of a setting. Hierarchy is very much there and yet there is a whole story about denial. People will say that they have flat structures and a lot of flexibility, giving the workers a lot of space and freedom and so on. This is again based again on the Silicon Valley thing. It is true in some surface kind of way, but Upadhya argues that actually it is a highly structured system and if it were not highly structured they could not do the work they are doing. They have huge software projects with hundreds of people working in them, everyone is working on a little piece and someone has to pull it together. So it is like a car assemble factory where the guys in the car factory probably have more autonomy than the software workers in the outsourcing industry, Upadhya suggests.

In Upadhya's research many people commented about the fact that there is a lot of reaching out in terms of information sharing. Most of this knowledge sharing is self-organized within the limits of certain tasks and deadlines. Interesting enough, this knowledge sharing happens not only between and within teams, but also between workers of different companies. Every company has this knowledge management system that is supposed to capture the knowledge which is created, but it actually doesn't. A lot of knowledge is circulating in informal ways and those informal networks reach beyond the company. People will post a question on the Internet and people from other companies will answer. However, most companies are concerned about their Intellectual Property Rights and try to prevent this by closing down external networks like the Internet. Because the companies make this security ring, this is something people will go home and do, as became evident in Upadhya's research. In this information sharing the whole question of promotion and salary is always in the background. People do not want to share too much, on the other hand you do help each other and you don't want your boss to know when you are in trouble, you reach out to other people. This is where the knowledge sharing becomes complex, because if one shares too much knowledge with colleagues, this may give them one step ahead for promotion, so the competition is always there.

As a sociologist going into all these companies, it is interesting for Upadhya to think about presence and absence. Of course there are various ways one has an immediate boss, who is there physically and sees what the workers are doing. There is a client or a remote manager who is in another place and with that person one interacts via phone, Internet, email. Often all are brought together in the same networked system of machines. But then many people say that it is really hard to do a whole project like this. There are certain points in which you need to have this face-to-face communication'. And people feel this very strongly, Upadhya noticed. It is usually in the beginning and at the end of the project that people want to meet. In the beginning they are planning and designing or get the requirements. When clients give the specifications that is always face-to-face. And then at the end it is often what they call 'post mortem', everything that went wrong in the project, people sit together and figure that out.

Upadhya's observation is that every project faces a crisis at some point. And often it is a problem of time, a time crunch, so what really happens when they go 'firefighting' is that they just go in day and night for over 7 days or on weekends to put in extra hours to get is sorted out. A dynamical hierarchy is defining the work. When it is a marketing question then the marketing guys come in and will be dominant. And if it is a technical problem than the engineers would come in. In a small startup company there are regular group discussions and concerning the body language for example, you see the person who is the president, the CEO, his body language is very much that of the boss. You can see others around who seem to be subservient to him and others who are more his equal and will challenge him. They kept talking about the Silicon Valley model, because the whole idea is that this kind of an organization, that free open space and all, is what leads to innovation. So here they are told we are a high-end company, we are a product company, we are innovation driven and we are venture capital funded, therefore we are like this. And therefore all the people in the company start becoming like that. There is a kind of a studied casualness, informality, the jeans, the T-shirts, the way in which people talk to each other, very informal, very we all together. To Upadhya it seems that a lot of it is very engineered. It is not just you and me and all together we are doing something, there is engineering behind that freedom, that so-called freedom.

In her research Upadhya found that in the IT industry a very specific kind of personal performance is staged all the time because people have to mould to the Global Service Delivery model. They have to develop 'global' communication skills to manifest themselves in a free innovative environment yet they operate in a highly controlled context where they are monitored all the time. They are witnessed and stage an identity to be witnessed with, yet the space in which they can take responsibility and be a witness themselves is very limited. This leads to a staging of professional identity that reflects an image, inspired by Silicon Valley models, of how a software engineer should be, and this deeply invades the self-reflection of the people involved as well. As a result, Upadhya argues, there is a great discrepancy between the actual practice and the self-image of the software engineers involved. However, people seem to resolve this by making a sharp distinction between professional and personal realms of life. Upadhya thinks that specifically in the Indian context this rupture is not problematic because traditional culture (enacted in personal and domestic life) is still very strong unlike in the West for example.

Transcript

The following is an edited transcription of the conversation. Film fragments of the conversation are included to illustrate parts of the transcribed text.

[Sequence 1] [Sequence 2] [Sequence 3] [Sequence 4] [Sequence 5] [Sequence 6]

[Top] [Transcript]

Sequence 1

CN: You have done exquisite research on entrepreneurs in Asia and Europe, people who work in small businesses and big businesses. If you look at the practice of entrepreneurs how would you rephrase what they do in terms of social engineering? They made a business, they design processes and actually they deeply influence how people communicate, how hierarchies are shaped, and for I know from your work, you know a lot about this.

CU: Can I talk first a little bit about the IT and software industry here in India, because that is what I have been working on most recently. Social engineering, that is really a big question, I am wondering how to approach that. The problem with trying to understand how IT business is working in India, the most fundamental aspect of it is that it is an outsourcing industry. Primarily the Indian companies are doing work for clients outside India, either in North America, Europe and so on. How the industry functions has been shaped by that history and that fact. And because of that there has been tendency to replicate in many ways the systems and organizational structures which you find in the West, which have been brought into India in the context of the software industry. Within a larger context you have these smaller companies, start-ups started by India entrepreneurs, working in various fields. Within that you will find a whole range of different types of companies. Not only in terms of what they are doing and the products they produce, but in the way in which they understand themselves. So when you look at these small start-up companies, so called high end product companies, like the one we show in the film July boys, they really self-consciously replicate the Silicon Valley model. So you have in this industry a notion of culture to which people shape what they do. They think of themselves as being Silicon Valley Company or an Indian company or some other kind of company. There are all these cultural models, which are being brought into this mix, but in a very complex way. It is not that they are doing this replication, but that they think of themselves as doing a kind of replication, which as an anthropologist I find very interesting. So I think that is the larger frame to which you have to understand some of the more specific things that are happening. For example, if we talk about communication, we hear a lot of talk about communication problems in the outsourcing relationship. You have teams of Indian software engineers working for a client or a manager say located in Germany or the US. They have to be in touch with each other all the time because they are working on a common project. But they come from different cultures, societies and sometimes come from different languages. So they identify a lot of the problems they face in getting projects going as communication gaps. When they identify this as a communication gap or as a cross-cultural problem, in order to understand the problem they always talk in the language of culture. So you get Indian culture is this, German culture is that, that culture is this, and than they try to negotiate these differences through this understanding. But the problem of the whole thing to me is that they have very stereotyped notions of all these cultures. Instead of understanding themselves as working in a common culture, the culture of IT maybe or digital culture or global information technology culture, they see themselves as being located in these very distinct traditional cultures. Understanding all of this and then placing the question of communication, there is a whole complex background, which makes it very hard for the observer to understand what is happening.

CN: I was puzzled when I red your article from 2006 on this, I was puzzled whether people work online and remote all the time and they use not-remoteness to understand their un-comfortability of remoteness. I was also wondering because in any collaboration between different disciplines or different countries, genders, differences make people feel uncomfortable and they find solutions for un-comfortability. I find it interesting that the uncomfortablity goes back to where you live or where you are from, to being in a place?

CU: I think that shows that if you talk about cyberspace and having a special IT or global culture - in this industry they always talk about 'getting global' - global somehow means than moulding your behaviour or the way you do things into a model, which everybody can understand regardless of where you are from. But in spite of all that, it really matters where you come from. I think it shows the fact that the people who are working in the cyberspace or whatever are in fact rooted in their locality. But at the same time I would say that they may not be as rooted as they think they are. That is the argument I try to make, which is in that book that you mentioned. I see that in this global information economy, where you have these dispersed teams and dispersed projects happening, people are not always in equal positions. For example one of the companies we work with is Philips, maybe I should not use the name but I am just taking the Dutch example. We visited the centre in India and we also visited the centre in Eindhoven. The software centre in India is not on par with Philips Netherlands; they are actually in a subordinate position. That is something, which in this whole talk about the global thing, has to be ignored. My argument has been that this idea of cultural difference actually becomes a mechanism of control over labour. So what you really have here is this huge army of software workers, program developers or software engineers - Indians -- who do work of a certain kind for all these companies located outside of India. By encapsulating them with a certain form of labour, these are the India software engineers who are already creating a space, and you are locating them in a certain space and you are also therefore preventing them from entering another space. So we have to remember the system of hierarchy in the system of control, which operates in this entire scenario.

CN: Can I go one step back before we go on about the hierarchy? You mentioned two times the model people mould their behaviour to. How do people find the model? The outsourcing industry was not like this a few years ago, so the models have been invented. And really fast. The models were also not there in the eighties in the US?

CU: I think this is the question I also address in my writing. Where does the model come from? Look at India, what were the first companies that would come into Bangalore? First GE, then Texas Instruments, and then a couple more of the big American companies. In a sense maybe the IT models started with that. But then there is the fact that many Indians have gone to the US into the IT industry and some have come back. I suppose the Silicon Valley model is something which people also absorbed as a result of their experience there. Then the third kind of model are the Indian outsourcing services companies like Wipro, Infosys and TCS. These are the big companies and they actually have developed their own models and they have perfected outsourcing as a type of business, what do they call it, the 'global service delivery' model in which they take advantage of all the time zones. So the work moves from one place to another, so that 24 hours a day the work is going on. Partly as a matter of management and partly as a matter of marketing, the problems that these companies face is the 'Indian' label. It is only recently that India has a reputation in the global economy as good in software or good at IT stuff. Earlier it was like very hard. You have to first build a reputation if you want to market your services in the global market.

CN: Already for 10 years Indian engineers have a good reputation?

CU: Yes, 10 years is of course long in the IT, but not really.

CN: When I visited MIT and TJ Watson in 1996 there were already plenty of Indian students and engineers. Also at Xerox.

CU: Yes, plenty of Indians. But I think that the Indians were there in the US or Europe, but for an American company to send its work to India that is quite something else. They had to overcome a hurdle of reputation and trust to get the clients to do this. And now this had been built up, people begin to look at India and say this is a good place to bring our work. Maybe 10 or 15 years ago it was difficult.

[Top] [Transcript]

Sequence 2

CN: You literally write somewhere that the 'prefabricated understanding' defines the self-image of the IT workers. This prefabricated understanding is created by management and intercultural training. It is really an intervention. So people are in a situation and have expectations and then there is an intervention by the industry to make the understanding. I once visited WIPRO and got a whole tour and saw a presentation of how they work and literally socially engineer, make analyses of politics and cultures and have scales how to measure these things, and these define the education of their workers when they go abroad or others come here. It is very refined in how people have to adapt to the system. Do you have any idea or notion about this prefabricated understanding? The knowledge is something that people can have themselves being part of the industry, but somehow this became an intentional intervention?

CU: In this whole question of performing the new global worker or professional, it is so complex. On the one hand again the model. If you want to send your Indian software engineers to Frankfurt, they have to be certain kind of person, they have to behave in certain appropriate ways, and that is what this training is about. For young man and woman who have never been abroad outside of India, they just have to learn the usual customs, manners and etiquette, the simple stuff like table manners and the complex stuff like human relationships. Because for example in Europe the way you talk to your boss is very different from how you do so in India. So these kinds of things they are supposed to pick up. On the one hand there is that kind of moulding to fit into the global market and also the specifics of the client side whether it is Europe, France, Germany or the US, but on the other hand there is still the notion of being Indian. I feel somehow, if you look at the training programs and the way the managers talk about this, somehow they have not made up their minds about what people are supposed to be. So it is not that people are given this prefabricated model. It is there but it is also in pieces. You have to be an Indian in certain contexts, you have to be like a European in some others and so on. To me it appears to be very confusing to be a software engineer in this context. At the same time we observed, and we interviewed quite a lot of people, in some sense it really is not a problem. Because I think people still have a very clear sense of self, which is not that eroded by all this stuff that they are meant to do. Say a call centre worker, for 8 hours a day they are on the phone, they have to be polite and smile and all that, but that does it mean that they become robots in their real life. You can do all that and go home and become something else.

CN: As you know my research is about witnessed presence. About what happens when people witness each other and when people witness the machine and the machine witnessed the people. When being a IT worker and interacting with a machine all day you do not have to become a robot all the way, you say. So what part you have to become a robot and what part you do not have to become a robot?

CU: This whole human machine interaction I should have looked at more closely and I did not in this research. I did try to ask people about their experience. I would ask about things like how much time in the day do they spend on the machine versus other types of interaction. I could never get anyone to articulate that experience. So I wonder as a researcher how you can approach this. Very few people are self-conscious and can articulate such things. Now and then I would get little things like when I asked a question about the machine someone would say, " Well, you know, software engineers are machines". So than I got this notion that people think about their work as very much driven by machines, as very mechanical. And when I looked into it I realized this is actually the case, because the nature of the work in most of the companies and for most of the junior engineers is highly mechanical. It is not the high-powered computer programmer who is this creative and nerdy guy sitting in the corner writing new programs. It is a very mechanical process, as you know software production around the world is like that. And that is what they do. I don't say they see themselves as being adjunct to the machine, but certainly they see themselves being stuck in a way in a very routine factory type of a setting. That is why most want to get out of that work as fast as they can. Nobody wants to make a career as a software programmer (forever). Only a few of them who work in the high-end jobs or in the labs would like to. So the everyday work of getting a website running or a software program out to the customer is not challenging, is low end and just not desirable for most of them.

CN: So if you compare it with normal industry the way people interact is mostly determined by hierarchy?

CU: Yes, and see again every person you ask you get a contradiction. Hierarchy is very much there and yet there is a whole story about denial. The first thing anyone will say to you when you ask them about software work is "Oh we are very flat, we have flat structures and we have a lot of flexibility, we give our workers a lot of space and freedom and dadada". This is based again on the Silicon Valley thing, it is a different kind of company, IT companies are different, and they don't have beaurocracy and hierarchy and so on. But if you look at how they function it is not true. It is true in some surface kind of way, the way they interact they have teams and the team members can argue with their manager about certain things, but actually it is a highly structured system and if it were not highly structured they could not do the work they are doing. They have these huge software projects with hundreds of people working in them, everyone is working on a little piece and someone has to pull it together. So it is like a car assembly factory. And I would argue that the guys in the car factory probably have more autonomy than the software workers have here.

CN: Because their mind is free?

CU: Yes maybe, you are right. Here it is mechanical but still your mind is engaged in this mechanical activity.

CN: What would you say that hierarchy does to how people actually witness each other?

CU: Witness...One of the things I find very interesting as a sociologist going into all these companies is when you talk about the presence and the absence. Of course there are various ways you have an immediate boss, who is there physically and they are seeing what you are doing. You have a client or a remote manager who is in another place and with that person you interact via phone, Internet, email, you are often all the time brought in the same system of machine. But than many people say 'we do work offshore; we have all these means of electronic means of communication, but it is really hard to do a whole project like this. There are certain points in which you need to have this face to face communication'. And people feel this very strongly. So either they organize meetings where they go to the client side or the client comes here and they have a face-to-face meeting to talk it out or in the absence of that they have a videoconference. And to me the videoconference is the most interesting thing. There are the guys sitting around the table, the local Bangalore guys and you have someone sitting on a TV screen over here, talking real time, so I don't now whether you call that presence or absence, to me it was a very unnerving experience, maybe because I have never done it, the manager is sitting in California and talking to the people on the table here.

CN: He was talking down?

CU: Well he was talking down, physically (in the screen) and literally, but also because he is the manager and he is telling people what to do, he is like beaming in, like some sci-fi thing. I know it is not sci-fi to many people, it is very normal but for me it was weird to see this. In a videoconference you are supposed to pretend that you are all there, and I know these technologies are coming where you virtually have all these people sitting around, but it's not the same, anyone knows it is not the same.

CN: Do you know why people feel that they should meet?

CU: Yes I think that they thought that is does help to enhance mutual understanding, that communication is just not complete when you are just doing emails and whatever.

CN: In my dissertation I come to the conclusion that people need to meet in real presence when things of ethical nature are at stake. So when you decide what is good and what is bad. So the moment you make a plan and you have to decide what is good to do and what is bad to do, than you have to meet. Or when you evaluate you have to meet. But when you are producing, you do not have to meet?

CU: It is usually in the beginning and at the end of the project. In the beginning you are planning and designing or you get the requirements, like clients when they give you the specifications that is always face-to-face. And then at the end it is often what they call post mortem, everything that went wrong in the project, you have to sit and figure that out.

CN: And they also do that on real life?

[Top] [Transcript]

Sequence 3

CU: Very often, I have to look at my notes, I never thought about it this way.

CN: If so that would resonate with my conclusions in my dissertation. Let's do a little bit more on the social engineering before we go into the tuning and the witnessing. What I did not read in the article I read from you, is what elements are self-organized in the social engineering? There are a lot of structures how people collaborate, how management is done, but I am sure there will be some self-organization happening as well?

CU: Among the workers for example? It is a good question you are giving me ideas for my writing. I think a lot is self organized in the sense that given a specific piece of work to do and a chain of work to do, and they are given deadlines and goals all the time, within that they are left on their own and go and do it. Like you have so many days to finish this. In that context I see what many people commented to me about on the social environment in the company, that there is a lot of sharing, a lot of interaction among people, if I can't figure out my bit, I go and ask you and say I have a problem can you help me. So there is a lot of reaching out like that in terms of information sharing, knowledge sharing. I would say that is self-organizing. Because nobody is telling me that I should ask you or you or this one. I go to the person I think will help me, who knows or has the information. So how the work is done in terms of the social networking in the company is probably very much self organized, within the structure of course.

CN: So actually within the structure the knowledge production happens informally?

CU: I would think so to a large extent. But again there are also formal mechanisms for that. And this is one if the interesting things about these companies, I want to write about it and haven't done, there is this whole thing called 'knowledge management'. And every company has this knowledge management system that is supposed to capture that knowledge which is created, but it doesn't. A lot of knowledge is circulating in these informal ways and those informal networks reach beyond the company. Very often they will tell me that if I have a problem, how to solve it and how to do this piece of code, they have these email groups and all these net based groups and they go there, they put the questions out there and they get answers also.

CN: Yes that is very nice, there is whole network through the companies of shared work to use.

CU: Which is something the companies may have problems with, with IP and so, but maybe it does not reach the level of IP, at least something are more routine. But still there is a community of people who work in this industry and the divisions of the companies at that level are not that important to them.

CN: But the higher they get ...

CU: And in fact the companies don't like it too much, for example in many companies you can't actually access the Internet from the office, it is blocked. Because they want to make a security ring, so this is something people will go home and do.

CN: Did you get any cues that there are also self organizational mechanisms within the network of one contract? Would for example people from India have contact with people in Holland or Belgium or...

CU: Within the project and within the company they would have. How much they would evoke those contacts I do not know. I really did not look at it that way.

CN: Probably only on one level, you don't go one hierarchy up?

CU: Probably not. For one thing you do not want it to be visible to your boss. You see, there is this whole problem...people are not only doing their job, they are also trying to build a career, so the whole question of retention, promotion, salary is always in the background. So one thing is to do your work well (and get a good) performance rating. The other is to impress your boss, working long hours, delivering on time, displaying your knowledge - that is important if you learn something new, next time you meet the boss, those words come out to show that you are always updating yourself. At the same time, and this is where this whole knowledge thing again becomes complex, because you do not want to share too much knowledge with your colleagues, that may give them one-step ahead of you in the promotion, so the competition is also there.

CN: Let's stick for a moment at this crises and celebration, when you do well you go up, if there is a crisis you go down. Maybe it is including people or excluding people?

CU: In terms of the company the performance evaluation in the company are very crucial because part of your salary is linked to that. You have a basic salary and your provisional salary. You get that component only if you reach certain level. And on the other end if your performance level is really low, you can actually be fired. So the pressure is always there. That is why, when I was thinking about the knowledge sharing, you do not want to share too much and at the same time you do help each other. And you don't want you boss to know when you are in trouble; you reach out to other people before you go to your boss with the question.

CN: Did you talk to people about when there is a real crisis in the company?

CU: Specifically I did not talk too much. My observation is that every project faces a crisis at some point. And often it is a problem of time, a time crunch, so what really happens when they go fire fighting is that they just go in day and night for over 7 days or weekend to put in extra hours to get is sorted out. I never heard of a project that really died because of crises, but I am sure it happens sometimes.

CN: Because you focus on the businesses and the entrepreneurs a large part of the witnessing on the work floor and on the market is colleagues and competitors. So you do not see a person, you see a colleague or you see a competitor. Do you have an impression on how these are identified and experienced?

CU: These aspects I did not really look at. But you can see there is this larger world of IT, in which all these players are there, the programmers, the senior managers the CEO's the lower level and where you fit in I suppose, especially because the software labour market in a place like Bangalore is very focused and within that you have to have a certain level of knowledge to be able to operate. So if you are an entrepreneur you have to have the right knowledge to be able to hire the right people, if you are a manager the same thing, and also as an employee you need a certain level of knowledge. So how do you identify that? Again I think that is often a problem solved in social networks, but it is also through the Internet and the websites, job websites and things like that.

CN: How do people behave when they meet competitor?

CU: In the company the competition is very much there among engineers because of the hierarchy and so on. Between companies, I think the bosses socially they are very polite and so. Among the people on the work floor there are certain things they would do to get an edge over each other. On one level they maybe very friendly socially, but when it comes to a question of the job they would do things like withholding knowledge, important information for each other, that kind of thing does go on.

[Top] [Transcript]

Sequence 4

CN: Let's start talking about the July Boys, the film you made in which you were really part if the team and saw the thinking, producing and going to market. When I watched the 10 minutes that are available at YouTube, I was impressed with all the body movements that were visible as an expression of the relations and processes that were going on. All these movements looked like 'a tuning of presences'. Can you elaborate on this 'tuning'?

CU: I don't know whether I can elaborate on this. I also found the whole body language very interesting, but I never got around to interpret that. Because it is a small company and a start up and all that, therefore people would notice each other more face to face than they would in a big company. So I think here it was very important for the leaders of the company to develop this kind of group behaviour and to take people along with him. That is what it seemed like in the meetings. Of course there were a lot of group discussions trying to come to a decision. Concerning the body language for example, you see the person who is the president, the CEO, his body language is very much that of the boss. You can see others around who seem to be subservient to him and others more his equal and then challenging him.

CN: Did you not find a dynamical hierarchy? Like when they discuss marketing, other voices are more important than when they discuss software?

CU: Yes, definitely. When it is a marketing question than the marketing guys would come in and would be dominant. And if it is a technical problem than the engineers would come in. That is certainly true.

CN: Which is nice of a small company, this dynamical hierarchy, everyone gets to be on top some time?

CU: Some time, indeed. But this is the only company of this type that I could observe that closely. That is why they kept talking about the Silicon Valley model, because the whole idea is that this kind of an organization, that free open space and all, is what leads to innovation etc etc etc etc. But, and you would know more than me, but to me it seems that a lot of it I very engineered. It is not just you and me and all together we are doing something; there is an engineering behind that freedom, that so called freedom.

CN: Can you talk about this?

CU: If you look at the three films together for example you see a contrast. The way people behave in these three types of companies, it is very different. But the people themselves basically come from the same kinds of backgrounds. So that leads me to think that it is the context of the organization that creates certain types of behaviour or certain types of interaction. So here they are told we are a high-end company, we are a product company, we are innovation driven and we are venture capital funded, therefore we are like this. And therefore all the people in the company start becoming like that. There is a kind of a studied casualness, informality, the jeans, the T-shirts, the way in which people talk to each other, very informal, very we all together.

CN: It is like a theatre school? They are witnessing each other's performance?

CU: Very much so. I found it very performative and even more so than in the more conventional company. There the performance seems more muted. Here it was very much about being this kind of person that will fit into that environment. That was my impression.

CN: We spoke to Sunil Abraham, who emphasized the performance aspects in the open source community. He argues that the performance aspect is a major drive for contributing to Open Source movement. They need an audience to perform. Do you feel that in the performance space of the working environment are your senses important?

CU: That is a good question. You know what happened to me in this research is that as an anthropologist you are always thinking that you have a way of getting close to how people think and feel in most contexts. This is the first time I studied a corporate context where you have organizations with a particular goal and a structure and all that. And here I found it very hard to get a sense for people as they are. Because I always felt that anyone is playing out some given role. And of course everyone is doing performance in [ordinary] social life, but here it seemed so much more as a predetermined structure. So it was really hard to find someone who would talk to me as a person of some type. Otherwise people were all the time giving me what they felt they were supposed to be saying or representing themselves in a particular way, doing the performance.

CN: Would that be connected to that they work with technology?

CU: I don't know. I would imagine it would be the same in almost any kind of company. For example if I were to go to an airline industry or hotel industry, I think it would be about the same. But I am guessing that.

CN: So part of professional life is performing in public space, opposite to being in a personal reality. I ask myself the question whether there will be in the end a community of people and machines. Whether machines will be accepted in their own right fully in a community. In such a case they will have an influence on the community. And the companies where people spend so much time behind the machines, nevertheless working together, from personal experience I find the atmosphere in working environments where people are all the time online, very different from environments where people are not online?

CU: But different how?

CN: At least half the time everybody is somewhere else with their attention. When they 'come back', than something needs to fall on the floor, something needs to break that sphere and bring them into this sphere and when you are in the machine you have to perform your presence as well. There is a lot of performance happening in the machine, you have to adapt yourself to the machine to be able to operate it, as you know because you also work with computers. Maybe, I am just exploring now, maybe this sense of performance also influences how people interact, or maybe one can argue this is the cause why people do not interact because they are all the time with their attention somewhere else.

CU: There are many aspects. One of the things I found peculiar, although I have not put it together, they have these big floors, these big open offices where everyone in their own cubicle is stuck on their own computer, doing whatever they are doing online. When they want to communicate even with the person next to their cubicle, they will send an email or do a chat. They hardly ever get up and say -- You know, hi should we have coffee or whatever. I found this very odd in the beginning. This is one way peopleinteract; even their social interaction gets re-structured in certain ways. We are so used to being online and having that type of interaction, that maybe they find it easier. Then I thought if I get up and talk to my neighbour physically, maybe that does disturb him too much, whereas if I send a message then they are free to see that message when they are ready. This way you are not interfering with the other's work. But then I discovered that especially when it is work-related communication they always do it through email and through the machine, because everything they do is recorded. And the recording is a very important part of the work process; actually they need to do it. Of course there are many meetings, talk face-to-face, but the actual mechanics of the work has to be routed through the machine. And this is not just computer programming; it is in almost all forms of work and many people have written about this, the work that people perform now is pre-programmed. If you are a cashier in a bank or standing at the checkout in the supermarket, the way you do your work is nearly completely determined by the machine. So that certainly will affect the way you think about your work and I don't know whether it affects the rest of your life.

CN: What is your impression?

CU: I would like to think that people, at least in India, have a feeling that there is still a lot of compartmentalization between what happens at work and what happens elsewhere. People have a tendency to shed that that part of your life when they go home, when they go to their personal life. That maybe less true in Europe where technology has penetrated so many spheres of life, which is precisely why the kind of research you are doing makes a lot of sense in that context. But in India somehow it has not penetrated that deeply into the society, so therefore the kind of questions you are asking don't occur to many people here. Even for the software engineers.

[Top] [Transcript]

Sequence 5

CN: Lets go back to the recording of the workflow. You don't talk, but chat or email or so. The essential of a witness is that my action that you see, becomes a deed. An action that is not witnessed can also not exist. The moment it is seen it is a deed. You can judge it, you can intervene, you can change it, so something can be done. While the witnessing that you describe people don't intervene, and don't judge each other's work, or do they? Do they witness each other work?

CU: The engineers? I think they do because they are working together. They do collectively get something out, so they are doing that in fact. Very much so. Exactly how they do it I am not sure.

CN: How does the manager witness?

CU: This is what is interesting about it, because the manager is first of all responsible for overseeing everything and making sure it comes together, that the project is coming together. And then there are all these online mechanisms too, which is being recorded, and it is been recorded in different places. There is a quality control person and there are all these statistics being collected all the time through the machine and these are in packages fed back to the manager. This is where the performance, this is where the project stands, the measurement are done and it is all highly technical quantifiable way they do all that. But what I thought was very interesting was the fact that very often on the client side this stuff is also being monitored. How the workflow is going, every single day, the person who is responsible for this at the client side is seeing what is happening. They do not just send an order to India and say come back in 6 months and show me what you have done. Every single day, almost every hour there is continuous monitoring of this person.

CN: So this is low trust?

CU: Very low trust, I would say. Because they have to continually justify. They will actually say, "How many lines of code have you written today? Why not a hundred and only eighty? And how many test cases have been run?" From every level, the team leader, the project manager, the guy above him to the client sitting over there, continually.

CN: So you're actually arguing that it's a really low trust industry? So the more outsourcingis there, the less trust?

CU: You know it's interesting, I never thought about this thing about trust. But when I used to ask this question to people about this whole thing about the Indian industry specialising in the low-end of the work, the simple routine stuff -- that is what gets sent over here and that is what we call the cyber-coolies and all. So now the industry is all the time trying to claim that "No, we're moving up the value chain. We're doing more high-end work. We're doing more consulting. We're doing more products." All that is true at some level, but managers or people who developed the Indian industry, who have been here for a long time, say what's the crucial difference as to whether a company is sending more challenging, more high-end or more responsible work to India? It's when they trust the company enough. So let's say you have a centre here, say Philips has a centre here or Siemens, when it's been here for a certain number of years and the parent company starts to trust this centre enough, slowly they build up their trust and then gradually they start giving better work. That's what happens.

CN: But they build the trust through a sort of rigorous monitory witnessing every little step?

CU: At least in the third party centres I'm supposing that's what happens, I can't say it so much for these captive centres what happens. I'm sure it must be more or less the same. Precisely, you know, because they don't trust them and therefore they have to make sure that they're on target all the time, that things are happening correctly.

[Top] [Transcript]

Sequence 6

CN: But does that not connect also to your cultural observation? That there are all these clichÈs, sort of reproduced. If there is low trust, you cannot say to someone "how are you? And what do you mean?" You don't say how do you feel and how are you when you're in a low trust interaction. You just analyse and judge.

CU: I mean you're talking about the distance between the manager and everybody else, that kind of relationship?

CN: Well I mean that when you're in a situation like you describe, where people are constantly monitored, by the hour from a distrusting perspective and one person is Indian and the is Dutch or whatever. How will you start to understand each other? Because if you do not trust, you will not ask an open question, which you need to get cultural communication.

CU: Well, it's interesting. I mean you're right... I never thought of it in terms of trust. I have thought of it in terms of hierarchy, between the client and the service provider. One thing is the hierarchy and I guess the lack of trust, the fact that they are sending a certain kind\ of work to, you know, a place where.. They seem to imply that there is a certain level of incompetence and therefore it needs to be managed, but how these cultural stereotypes get worked and reworked from different actors, you know. So there is this whole bunch of things that always get said about Indian engineers, especially from the European perspective, which is where I did the work. One is that (see, again it's like a no-win situation) they are too subservient, they don't assert themselves, they don't take ownership, they're not taking initiative in doing, they are very passive. But on the other hand in a sense that's 'good' because they are 'very good at taking directions'. So you can give them this work and then just go on and let them do it, but you cannot expect innovation from them. So that is the way they are seen, but at the same time if that is the way they are seen they are almost expected to behave that way, right? Because they're in a situation where they're not given any responsibility. And if you talk with the Indian engineers one on one, they will never see themselves like this. They see themselves as being very intelligent, very very capable and very much capable of doing much better work than what they're getting. But they're not getting that kind of work because there is this low trust situation.

CN: And also toward responsibility it may be crucial what you say there? That if you have no responsibility, that you can also not change the monitor. So if you are a witness, you also have the responsibility of being a witness. The moment you become a witness you can also start to lie about what you saw. So you take responsibility for being a witness or for being a participant. And if you don't get that wrong, then how can you..... I'm just puzzling right now with the fact of being a witness and having a task. You know there is a very big difference between whether you delegate responsibility or whether you delegate tasks as a manager. I've been a manager for many years, I was very good at delegating responsibility because it means that I can do my things as well. But if you start delegating tasks, you have to monitor all the time to check whether the task is carried out well. And you say now that in the IT here it's been a lot of monitoring, task monitoring, before responsibility is delegated.

CU: In fact another thing that they often say --again these are all comments about the Indian engineers - is that they need 'micro management'. See this is the implicit comparison with my Dutch example. That, you know, there they are very self-driven, you just give them the job, delegate the job and they will come back with a finished product and it will be on time and it'll be of a certain quality. But the Indians, they all the time need directions. You can't just give them a task and say, "come back". You have to all the time see what they're doing, they all the time have questions. Indeed I feel that this has nothing to do with being Indian or something. It really has something to do with the work context, precisely the fact that they are not being given this kind of responsibility and the hierarchy.

CN: But do you also find, for example, with the IT companies who are based in Bangalore, like take a Dutch example of the Philips group here, they are also not trusted?

CU: That's why engineers prefer to work in the multi-nationals, because they get better work, I think they get more responsibility, they get more challenging work. So the competition is always to join these companies. But jobs, of course, that everybody can join, it's mostly the third party outsourcing. The other context is very different. But if they have a choice, they'll definitely go to Siemens of Philips or Texas Instruments. There also there's some amount of routine work. By and by I think there is a bit more of a feeling of being given a bit more responsibility and getting a bit more trust. For example, we tried to find out from these people whether they feel that the Dutch employees are treated differently from the Indian employees. And mostly they say "no"; they didn't feel that they were being somehow put on a different track or whatever. The only difference is what kind of work is being given to the Indian software development centre here, as opposed to what happens in Philips Eindhoven or whatever. There still will be some difference. But more and more it's tilting in the other direction. In fact, then it creates so much resentmenton the part of the Dutch employees, because they see their projects being sent out to India much more.

CN: Since we are working on the idea of a community of autonomous systems and people, where is the machine in the hierarchy? The log file is probably really high of everybody --I say, "no, I didn't sent that mail" and the log says " yes, you did send the mail".

CU: Yeah, I think the machine is very central to how these companies work, but it's not something that I've looked at closely. But certainly it is a fact that everything has to be done through the machine and there are systems in place in the machine that you have to use to perform. So I would imagine that it's absolutely central.

CN: Last question: And is the machine a kind machine in your perception? Do they like the machine?

CU: I don't know, I can't answer that question. I think they probably like it when work is going well. When it's giving glitches, they probably hate it. I think there are also ways in which people try and domesticate the machine. I mean, if you see like the screen saver and other things that you put around the machine. The other thing I noticed, which is probably true of IT workers around the world, especially if it's a call centre or even in software, is that I don't think anybody feels overwhelmed. I mean the machine is in many ways controlling the workflow and controlling the work process, but people have ways of tricking the machine and they enjoy tricking the machine, getting around the system. Not logging what you're supposed to log or finding ways to fiddle the system so it doesn't log what you were doing. And these are computer engineers, so they can do stuff like that. There's a kind of subversion that happens also sometimes. People don't talk about it much, but I think it's very common.

CN: So is there something that you would like to add?

CU: No, I don't think so.

CN: I forgot to ask one important question. The difference between men and women. Are there any women?

CU: Oh yes. In fact I was going to mention this gender dimension also. Because one of the things I think for you would be very striking is that in the 'July Boys' film it seems like a very masculine work environment and there is a lot of very masculine body language going on around. And in that company it was a fact that there were much fewer women among the people that were following around. But overall there are about 25% women in the workforce, so it's not that they're absent. Among software engineers I think it's about 25% now. See again, there are a whole lot of issues around gender here. A lot of people say that there is no difference, women treated equally, they do as well and so on. But I mean there are very significant differences, partly having to do with the women's responsibilities to the home and the whole kind of social context in which they still live in India, where they pretty much have to bear all responsibility on the domestic front and then they have a full time job. We did collect some information about how sometimes men resent the presence of women in the workplace, because in the team for example they feel that women don't give an equal amount of work into the project. Because they have other responsibilities, they leave earlier; they may not put in enough working hours and things like that. And there is this male camaraderie that happens a lot in these work places. As soon as you put a woman into the mix that gets broken down, right? So that gender divide is still there. On the surface it all looks very nice, there are women bosses and thing like that. But there is an underlying gender asymmetry and rivalry, I think that happens. For example, the open source movement. I have been to two of the big open source conventions here in Bangalore. You know I could hardly spot any women. It's such a masculine environment. And the open source movement is more the sort of high-end of the IT here, the more intellectual IT people. Very, very little involvement of women.

CN: So women are witnessed differently?

CU: Yeah, sure. I mean, they're there. And IT is seen as a good job for a woman... Now women are getting much more into engineering education than they used to. They get into the software sector through that.

CN: And do they witness differently?

CU: I'm not sure. Again, a lot of the same discourse we find in the west now. Women make better managers, right? They're soft, better interpersonal skills and so on. Whether it's true?

CN: Thank you

CU: Thank you

This website is work in progress. You can use the content of this website under the Creative Commons license Attribution-NonCommercial-NoDerivs.