Data in the Time of Pandemic - A Marxist Perspective

B. Bipin

The biggest strength of the left front government in Kerala is the opposition! Led by the Congress in decay, the opposition in Kerala drags down every social and political issue to its own intellectual level and thus absolves the left from addressing them in any meaningful manner. The opposition makes every issue into a personal mud-slinging contest which the left enjoy as it spares them from conducting a dialogue on the political-ideological level. The latest issue of transfer of personal data of large number of people under treatment and isolation for Covid to the US data analytics firm ‘Sprinklr’ is no exception. The sociopolitical discussions that followed immediately got dragged down to the lowest level possible with political leaders and self-appointed data experts of both sides going at each other’s throats with ridiculous and contorted arguments. The media absolutely loves this as it keeps people glued on to their TV screens, while making sure that the social, political and ideological dimensions involved stay securely wrapped up. Hence there is an urgent need to address the real issues here which are getting drowned in the vulgarities hurled at each other by rival politicians.

The Kerala government initiated the transfer of personal data, including health data, of infected and suspected persons to a US data analytics form Sprinklr on March 24. The government claims that Sprinklr, whose CEO is by birth a Keralite (but a US citizen now), had come forward with an offer to help without any financial obligations to the state. The news on the initiation of data transfer came out only on April 10 and as per media reports, the agreement with the company was only signed subsequently on April 12 in a haphazard manner. Following the opposition’s accusations of irregularities in the deal, the left have advanced three arguments in support of the deal. First, that even though data privacy is a very important thing, human lives are much more important and hence in times of a pandemic there is nothing wrong in giving sensitive data to corporate entities for analysis. This argument has also been endorsed by the CPM Secretariat. Second, all legal steps were taken to make sure that the data is secure. Time was insufficient for drawing up a detailed contract with the firm, but enough guarantees have been obtained from the firm to assure safety of the data. Third, the deal was concluded because there were no alternatives; the state agencies does not possess the infrastructure or technical competence to handle such data and analyse it and hence it was necessary to involve corporate firms for this. It is for the greater common good as data analysis can help fight the spread of infection better. The merits of these arguments can only be looked at based on the correct understanding of the emerging political economy of data.

Data as a commodity

Many peddle the phrase ‘data has become a commodity’ very irresponsibly now. So it is very important to first understand what it means to say that data is a commodity and why it has become a commodity. The commodity nature of data has to be viewed at different levels. First, raw data collected from people with or without their permission is similar to any raw material. And as Marx makes clear, raw materials may or may not have value (the term ‘value’ is used in the Marxist sense here and not in its literal sense. Those who are not familiar with the Marxist concept of value may think of it as the price of the commodity even though value and price are not the same for Marx). If the raw material is freely available and can be picked up without expending any substantial labour, then it does not have any value. But if labour needs to be expended in extracting the raw material, then it becomes valuable. In our case, raw data is collected either directly though questionnaires or forms (this is what was done by the Kerala government) or indirectly through social media or search platforms. For example, google may collect data about one’s search records and browsing history. This needs appropriate software tools to be developed which needs labour to be expended. Hence, raw data has value. Raw data, which is similar to raw materials and which has value, becomes a commodity when it is brought to the market and sold at its value. But raw data is unlike any other raw material, it does not get expended when used! This is an important dimension to which we shall return later.

This sale of raw data will be realised only if someone has some use for this data. Here is where data analytics firms come in, raw data is indeed raw material for them. For instance, from this raw data they may be interested in extracting behavioural patterns, food habits or health condition of those concerned. Sometimes, from data pertaining to a particular region, data analytics firms may want to extract patterns regarding the dominant food habits of the people in this region. This needs the data to be processed using advanced techniques and obviously needs the expenditure of specialised labour. Thus, the processed data acquires value. This processed data is the second form that data takes as a commodity. When the processed data and the patterns and conclusions that can be drawn from it is sold, it becomes a commodity.

But who is interested in buying these patterns and conclusions? If, for instance, the international fast food giant McDonald’s wants to start their outlets in Kerala, they may want to have an idea about the prominent food habits of Keralites. They may also want to know about the cities which will be the most welcoming to McDonald’s outlets. Data analytics may be able to process raw data regarding food habits, search queries, social media posts, life style, income distribution etc. to arrive at certain pointers which may help the fast food firm to locate their stores optimally for maximum profits. Thus, corporate entities become the customers of processed data and its conclusions which they then put to use to sell their actual product. There are direct uses for processed data too, but this is the main use. So, raw data is a commodity which is sold by the data acquiring firm to the data analysing one and processed data as a commodity is sold by the analysing firm to corporate entities. Interestingly, sometimes a corporate entity combines all these roles in itself: Google acquires data through their tools, search engines, apps and the like, process it with their own advanced work force and use it for myriad purposes like targeted advertisements etc.

Why is decaying capitalism desperate for data?

Of course data acquisition and processing have their important social uses too. If, for instance, the health related data of an area which is voluntarily provided upon analysis yields a pattern of respiratory diseases, then its cause can be traced to some kind of severe air pollution source and action can be taken to control it. But capitalism does not put data to its proper social use, but to corporate use for increasing their profit. Why does data become a commodity in capitalism? Because there is a need for more planned and pointed selling strategies and targeted, vigorous advertising methods to ensure profits. Such a course of action is necessary because of the nature of the capitalist system. In capitalism, production is never aimed primarily towards consumption but towards realising a profit. It is not demand that drives production but profitability. So, production is carried out at full capacity is search of profit. This most often calls for aggressive advertisement strategies to boost up the demand. Especially in the last two or three decades which were characterised by steady fall in the rate of profit, realising profit from maximum sale has become a necessity to firms. This is where data analytics becomes very important not just for individual corporate houses but for the capitalist system as a whole.

On one level, data analysis helps to recognise behavioural patterns on individual and social levels which enables capitalism to anticipate and predict individual and social behaviour and to orient their selling and advertisement strategies. But, of late, the political economy of data has taken one step further. From predicting the behaviour, it has gone to controlling it. As Shoshana Zuboff writes in her new work ‘The age of surveillance capitalism’, “…. automated machine processes not only know our behaviour but also shape our behaviour at scale. With this reorientation from knowledge to power, it is no longer enough to automate information flows about us; the goal is now to automate us”. Capitalism is no longer satisfied with analysing our private data to predict and anticipate our behaviour, it wants to use this data to shape our behaviour to suit them. A simple example will illustrate this. Let’s say I browse an online store and look at the smart watch models that they have. After going through the specs, uses and price, I conclude that it’s not worth and decides against buying a smart watch. I go for lunch, comes back and opens the page of an online newspaper and out pops an ad for smart watch on the right hand side. I ignore it. In the evening I open my FB profile on my mobile and start browsing through the posts and I keep getting sponsored posts indicating the benefits of smart watches. I ignore them too. Next morning, I get ‘reading suggestions’ on my mobile and 3 out of 5 of them are regarding smart watches! The data mafia ensnares me in a web of smart watch hyperbole; if I am not stubborn enough, I may easily fall prey to this barrage and become convinced about the utility of smart watches. Of course, this is a very simple example but it shows us how capitalism has moved from using data to predict behaviour to using data in an extremely invasive manner to shape behaviour.

This tells us one very important thing, that invasive and ubiquitous use of data neither changes the basic characteristics of capitalism nor takes it beyond the realm of Marxian analysis, as many have claimed. The different commodity forms of data can be understood in Marxian terms and the use of data by corporate firms exposes irrational nature of capitalist production which Marx had theorised 150 years ago! Capitalism uses data analytics and data surveillance to anticipate, predict and manufacture behavioural patterns and to thereby survive.

Left’s arguments and evasions

Now that we have seen how data functions as a commodity and why capitalism is desperate for it, we can revisit the arguments put forward by the left in support of the transfer of private data of individuals to the US firm. The first one is simple, lives matter more than data. Of course data privacy is an important thing but in the middle of a global pandemic there is no importance for it. Hence, there is nothing wrong in data transfer. Interestingly, this argument was the one that was put forward by the CPM state secretariat also in support of the deal. Capitalism and bourgeois state have always loved disasters and pandemics, they make everything ‘legal’. To circumvent all due democratic procedures in the name of a crisis is an old trick. It continues to be used around the world in the time of Covid-19 too. We saw how the Hungarian far right prime minister gave himself the power to rule by decree citing the danger posed by the pandemic. The reason offered was chillingly similar: democracy is a wonderful thing but lives matter more, and to save lives absolute powers are needed! This step drew sharp responses on social media from left sympathisers even in Kerala who were professing the inalienability of fundamental rights. They have obviously forgotten now that privacy is a fundamental right too! We also saw how the Bombay and Rajasthan High Courts passed verdicts saying that bail pleas will not be heard during lockdown. Progressive groups across the country objected and the Supreme Court subsequently stayed the verdict. We also saw how the Delhi police, in the name of the pandemic, circumvented due process to selectively arrest and sent to police custody activists and people from the minority community in the Delhi riots case. This shows how a health crisis can be made an excuse for political and communal witch hunting.

Marxists like crises too, but in a way diagonally different from capitalists and the state. State and capitalism look to use crises as an excuse to attack the people, to take away whatever little rights and guarantee they have and to consolidate their rule. Marxists look to use a crises to rob the ruling class of its privileges. Capitalism uses crises to obscure its systemic causes; Marxism uses it to expose how capitalism produces crises (including pandemics). Capitalism’s only priority during a crises is to scrape through it unscathed and to emerge from it with even more power over the people than before (this is termed ‘going back to normality’). Marxism’s priority is to use the crises to expose the system and to make sure that, even if capitalism survives the crises, it is battered, bruised and vulnerable and that the toiling masses have started questioning the system. The common thread that connects all the examples that we saw in the previous paragraph and the decision of Kerala government to hand over data is that they all contribute towards consolidating capitalist state power during the crisis. To say that the right to privacy is not important during the crisis and that it may be suspended by the state without informing the people is a thoroughly bourgeois use of the crisis. It makes the state omnipotent, sets precedence that can be used and consolidates the bourgeois hold on private data. So the class essence of this argument is thoroughly bourgeois. But the fact that the CPM secretariat thinks the way bourgeoisie does is hardly news to any one, is it?

The second argument is that all possible legal and technical precautions have been taken to make sure that the data is safe. Guarantees have been obtained from the US firm regarding the safety of the data. This argument has been repeated by the Kerala government in its submission to the High court of Kerala. The left sympathisers ensure us that the data is safe on two counts; legal and technical. The argument goes that legal conditions regarding data security is much stronger in the US than in India and hence the firm will make sure about it. Furthermore, any data breach, they say, will have negative impact on the credibility of the firm and hence can be ruled out. Technically, they ensure us that the security protocols are strong enough to keep the data safe in the cloud. Another clinching argument is that Sprinklr is a ‘reputed’ company run by a Malayali and will never do ‘ugly’ things like leaking data. That settles it then!

Any data analytics firm lives on data, data is its life breath. Of course it will not leak data to third party, not because it is illegal to do so (corporates do not give a damn about legality) but because it is their priced possession. As we have seen, data is a special type of raw material which is not expended upon use! No amount of legal guarantees and no amount of encryption are going to prevent the firm from using this data for purposes other than pandemic control and contact tracing. Various fast food chains and pharmaceutical giants are the clients of Sprinklr, nothing prevents them from doing consultancy for these giants regarding their best selling and advertisement strategies in Kerala using THIS data later. Nothing prevents them, as we saw before, from using the same data to launch targeted ads for these giants with the aim of tinkering the behavioural patterns of the people of Kerala to suit their needs. Sprinklr has made sure that there is high visibility to their association with the state of Kerala and the corporate entities with plans in Kerala would not have missed this. So they will come to Sprinklr sooner or later for ‘advice’.  

It is amusing to see left sympathisers placing complete faith on the legal system in the US and Sprinklr’s commitment towards it. Data giants like Facebook and Google break almost every law in the book on a daily basis; laws have never stopped the bourgeois firms from going to any extent for maximising their profits and never will. It is ironic to see self-declared ‘Marxists’ placing such absolute trust on the US legal system after what Marx had to say to the bourgeois about their laws: “…. your jurisprudence is but the will of your class made into a law for all, a will whose essential character and direction are determined by the economic conditions of existence of your class”. Shoshanna Zuboff in her book has this to say regarding the status of laws regulating health and fitness apps in the US: “In the US, most health and fitness applications are not subject to health privacy laws, and the laws that do exist do not adequately take into account either new digital capabilities or the ferocity of surveillance capitalist operations”. She also documents how, when the state of Illinois passed the Illinois Biometric Privacy Act, the data giants worked “…. feverishly to prevent other states from enacting a law like the one in Illinois”. The political muscle of data giants in the US is unparalleled and to naively believe that US laws will protect our data is foolishness. At least, we should consider the fact that Facebook’s lobbying budget for 2017 was of $11.5 million!

A large number of people, especially those working in the IT field, wrongly assume that concerns about data security are technical in nature. Naïve left sympathisers among them, through videos and write-ups, assure us how safe the cloud servers are and how encryption will keep the data safe. The data with Sprinklr is unsafe not because cloud servers and encryption are found wanting, but because the system in which the firm operates is called capitalism! No law, no security protocol and no encryption can keep the data safe from capitalism’s greed for profit. As Zuboff writes in her book, ".... the digital can take many forms depending upon the social and economic logics that bring it to life. It is capitalism that assigns the price tag of subjugation and helplessness, not the technology."

The third argument is a little more persuasive than the first two. It concedes the dangers of handing over private health data to a US corporate firm. But the deal, it says, was still concluded because there was no other option. It was felt that data analytic tools were essential for effectively tackling the spread of the pandemic, especially with lakhs of people expected to come back to Kerala in the coming months. Government IT concerns does not have the infrastructural facility to host this amount of data nor do they have the needed technical competence in data analysis. Hence, in spite of understanding the pitfalls involved in it, the deal with the corporate firm had to be concluded. For the sake of argument, let us assume that this is correct. Such a deal was concluded out of desperation as there was no other option, let us say. Because of the urgency of the situation, there was no time to frame an extensive legal contract or to follow due process, says the state. Let us grant that too. In that case, how should the government have gone about it?

Whatever the situation, whatever the urgency and whatever the desperation, before concluding the deal and before initiating data collection and transfer, the people of the state has to be informed in black and white. That is non-negotiable. A democratically elected government, especially when it is led by a party which calls itself Marxist-Leninist, is answerable to the people and people alone. Tell the people of Kerala as early as possible and as clear as possible that the state is entering into an agreement for data transfer and analysis with this company. Explain to them the government’s desperation, its helplessness. Clearly inform them about the dangers involved, put it before the people that in the current climate there is a chance that the data shared may be misused later. Tell them the steps that are being taken to minimise the chances of misuse. While collecting data from an individual make arrangements to inform him/her that this data will be shared for analysis. Privacy is a fundamental right and everyone has a right to know what the state does with his/her private data. Make the whole issue as transparent as possible. 

The irony is that all this could have been easily accomplished. The Chief Minister of the state has been conducting daily press conferences and have been talking even about the need to feed stray dogs and monkeys! It is shocking that he never thought it important to let his people know about this deal. The data transfer was initiated, as per available reports, on March 24 and the news broke on media only on April 10. He had 17 full days in between and on almost all these days he did meet the press. Yet not a single word was uttered about it!! Might it have been an oversight from the part of the state that they failed to tell the people on time? Unlikely, but even if that was the case they should have immediately corrected the mistake when the details of data transfer came out. They should have unconditionally apologised to the people and should have laid out every detail of the deal on the table for everyone to see. This is what any government which considers itself responsible to the people would have done, let alone a Marxist one. That none of this was done, that the deal is still shrouded in mystery, that instead of reaching out to the people and coming clean the state is still trying to convince them how great Sprinklr is and how strict US laws are, gives the people enough reason and more to be very suspicious about the intentions behind this deal.

How would a real Marxist ‘government’ have gone about it?

Let us suppose that a genuine Marxist-Leninist party comes to power in a state like Kerala through elections (very unlikely in the era of decaying capitalism, but still!). So it has to work inside the coordinates of bourgeois democracy. And it is faced with a situation in which it desperately wants to have data analysis support for fighting the spread of the pandemic. What would it do? First, of course, it will pool all its IT resources, all the specialised manpower available to it and stretch every one of its IT sinews to see if the task can be accomplished in-house. If it is convinced that it cannot be done within the necessary time frame, then it is on the lookout for data analytic firms inside its juridical limits which have the infrastructure and tools to do this. If there is such a firm, it never bothers to enter into an agreement with it, but the state just takes over the firm. The state politely asks the owners of the company to take a break and chill out at home (we got this, pals!) and retains all its specialised workforce. It brings in more technically competent workforce from across the state who can be of help and starts using the firm’s resources and tools for data analysis immediately. Of course, the owners will cry foul and the whole bourgeois class will back them, their media will bark that the takeover is unlawful. As long as the whole weight of the people is behind the state and the party, nobody cares what they say. They will move the courts. But a Marxist government would always prefer to fight in court the bourgeois complaining that their ‘private property’ have been taken away from them rather than the people complaining that their fundamental right has been frozen. ‘Desperate times, My Lord, and hence desperate measures’ is the plea of the state. That is, a Marxist government would disregard and circumvent rights to tackle the pandemic, but those will be the rights of the ruling class and not the rights of the toiling masses. That the Kerala government chose to do exactly the opposite tells us about its class affiliation.

If it cannot be done in-house using government resources and if there is no firm at hand which has tools and resources, the government may have to approach a corporate firm. In that case, the state first puts this proposal in front of the people in black and white and explains to them why this is necessary. It explains to the masses the desperation and helplessness of the state regarding data analysis. Then and only then will it go ahead with data collection and transfer. Everyone from which data is collected will be informed personally that this data will be shared with a corporate firm for analysis. All available legal measures will be taken to ensure that the data is safe and they will be explained to the masses as well. The loopholes in these legal measures will also be explained and the masses will be alerted about the possibility of misuse of data.

Further, a Marxist government will invite the attention of the people to one particularly ironic thing. Infrastructure for data storage exists, strong tools for data analysis exists and specialised workforce for doing the analysis is also ready at hand. This mammoth infrastructure as well as these specialised tools are the result of human labour and the workforce consists of IT workers. But when we need to use these resources for saving the lives of the masses, we are forced to enter into a contract with the ‘owners’ of the company who in fact are neither interested in the lives of the people nor have played any part in the labour that created all these tools and facilities. It is because the control of the infrastructure, tools and workforce is in the hands of the ‘owners’ who are profit mongers that our data is not safe. Because as far as the owners are concerned all these facilities, tools and workforce are not meant to save lives but to produce profits. This is the crux of the whole problem here and it directly arises out of one of the fundamental contradictions of capitalism. Any Marxist government ought to put this in front of the masses for them to see and understand.

In Marxian terms, raw and processed data have a use value now in that it can be used to save lives. But the owners of the data analytic firms are least interested in this use value, they are only interested in the exchange value of data, both raw and processed. They are only interested in maximising the profit that can be generated using this data. This is why our data is not safe; the owners will use them sooner or later for making profit. The irony is that these owners played no part whatsoever in the labour that created either the cloud storage facilities or the data analysis tools. Nor are they going to play a part in the analysis, which will be done by their workforce. So, why on earth do we need an owner?? As far as I can see, what is really going on here is an exchange between people on both sides. On this side is the people of Kerala who would like their data to be analysed and on the other side is a group of specialised labourers who can do that. With them are facilities and tools made by our fellow workers. If it was a direct deal between these two parties, the question of data security never arises! It is the parasite in between that creates all type of problems because it is the one who brings the profit motive into the equation. Of course, the owner’s search for exchange value in something intended to save lives of people does not owe its origin to his/her character but to the character of the capitalist system. Thus, a Marxist government will explain to the masses the real systemic contradiction of capitalism which only sees profit in data when the people and the IT workforce see in them nothing but a possibility of saving lives. This is the root of the matter and as Marx suggested, Marxists ought to get to the root of the matter and explain it to the masses.

Naïve left sympathisers, who imagine themselves to be Marxists, but are actually nothing but liberal democrats, will object: all that you say may be correct but what difference will it make to the deal? It makes no difference to the deal, but it makes a hell of a difference to the consciousness of the working masses, my friend! What we are going through is a global health crisis which is caused by capitalism (large number of studies have documented how novel viruses are being released from the bosom of forests because of capital’s forays into them in search of profit). In that sense, it is nothing but a capitalist crisis. And in capitalist crisis, Marxists are duty bound to put before the masses the systemic contradictions of capitalism that leads to such crises. The present issue of data security is a particularly convincing instance for Marxists to explain to the people how capital sees profit in something which is intended to save lives. This is the root of all current concerns on data privacy. Class consciousness of the masses cannot be raised to the proletarian level without explaining to them the systemic contradiction behind every concrete difficulty that they face. Then and only then will they appreciate the need and the unavoidability of the revolutionary solution. Of course, the left government doesn’t care about any of these things, nor do its sympathisers, because they are Marxists only in name and petty bourgeois social democrats in practice.

Data and Imperialism

The connection between data analytics and surveillance and imperialism needs much deeper analysis, but let me say a few words here. Many have criticised the Sprinklr deal as caving in to imperialism. They seem to be troubled by the fact that our data has been handed over to an ‘American’ company which they say is ‘imperialist’. This arises from the wrong notion prevalent among majority on the left concerning imperialism. They view imperialism as primarily a relation between countries. This is what prompted CPM PB member M. A. Baby to rejoice when the US national football team was defeated in the world cup by Argentina. He said that imperialism was defeated! This wrong notion is also the one that makes Walmart bad and Patanjali good. Imperialism, in reality, is a stage of capitalism (its highest stage, as Lenin said) and is characterised by the predominance of monopolies. Hence, big data and surveillance based on it are connected to imperialism in two ways. One, there is monopolisation in data firms themselves. It is known that 90% of data acquisition and analytics is done by 5 companies: Google, Facebook, Amazon, Microsoft and Apple. Two, data analytics and surveillance makes the playing field even more uneven in favour of bigger firms. An international fast food brand will be able to take assistance from one of these big data giants for targeted advertisements and for tinkering the behavioural patterns of individuals to suit them. But a small fast food joint in a provincial town will not be able to afford it. This increases the competitive edge of the international giant and makes the already asymmetric playing field more skewed. Thus data analytics increases the drive towards concentration of capital by busting the smaller firms which leads to monopolisation. This is the connection between data giants and imperialism. Hence, transfer of private data does not suddenly become ‘nationalistic’ if it were given to an Indian corporate entity.