A Police Officer in an Assistant Public Prosecutor’s Clothing

I came across a strange provision in the Code of Criminal Procedure, 1973. Part of Section 25, which concerns the appointment of Assistant Public Prosecutors (APPs), reads as follows:

…(2) Save as otherwise provided in sub-section (3), no police officer shall be eligible to be appointed as an Assistant Public Prosecutor.

(3) Where no Assistant Public Prosecutor is available for the purposes of any particular case, the District Magistrate may appoint any other person to be the Assistant Public Prosecutor in charge of that case: Provided that a police officer shall not be so appointed—

(a) if he has taken any part in the investigation into the offence with respect to which the accused is being prosecuted; or

(b) if he is below the rank of Inspector.

Yes, the appointment of a police official as an APP is an exception rather than the rule and yes, an officer involved in a case is not eligible to be a prosecutor as well. Even then, it seems stretching the bounds of propriety to have a provision such as this.

So many questions come to mind about this provision. How is it consistent with the principle of separation of powers? Can it be said with a high degree of certainty that an individual will not be conflicted between her loyalty to the police force to which she belongs and the responsibility she has been entrusted with as an APP? Is there such a lack of qualified advocates who can be appointed as APPs that this provision needs to exist? How often has this been utilised over the years?

On the last question, it would be a relief to know if, like several legal provisions, this one is seldom applied in practice. But even if it is never used, we need to take a long hard look at the continued existence of such language in the statute.

A Case for Unpredictability

Researchers in Germany have created a machine learning tool that has predicted the winner of the ongoing FIFA World Cup. This tool says that Spain has a higher chance at the outset but if the Germans make it to the quarter-finals, the odds tilt in their favour.

The creation of this tool does not come as a surprise. Machine learning tools thrive at making predictions and this is just another example of a technology that has become adept at doing a task far better than humans.

However, impressive as a technology might be, there is always room to ask if it should be applied in a particular field. These were my thoughts as I read the article. It goes without saying that better predictions would be beneficial in a lot of fields, including ones like medicine and weather. But would they add value to a sport like football?

I must admit I pose this question from a philosophical bent of mind. Isn’t part of the thrill of a sport its inherent unpredictability, of the unexpected happening? Bookmakers and gamblers might beg to disagree but there is a reason why seeing an unfancied team win against all odds is deeply satisfying. Is it possible then that the invention of a machine learning tool that accurately predicts the result of every match might lessen the enjoyment of the game itself? Worse, can it contribute to a self-fulfilling prophecy where players contrive to fit the results of the prediction (of course, this means having a particularly low opinion of the free will of the human beings involved)? I do not have any concrete answers at present but this is a line of questioning worth pursuing, both for this particular application and for machine learning in general.

Hope for the Future

The NFL (the National Football League) in the US has approved a policy that requires players to stand while the national anthem is played before matches. This effectively prevents the act of kneeling that some players, beginning with Colin Kaepernick, used to draw attention to racial tensions within the country. I found a fantastic headline in the Politico on this:

The Arc of History Bends a Knee Toward Kaepernick

This is, of course, a tribute to the iconic lines uttered by Dr. Martin Luther King Jr. during the march to Montgomery. Dr. King said that the arc of the moral universe is long but that it bends towards justice. These lines, borrowed from a nineteenth century preacher named Theodore Parker, are a cause for optimism. They emphasise that if you are on the path of righteousness, the world will eventually come around to your way of thinking, even if the odds seem stacked against you in the present.

Ta-Nehisi Coates highlighted this in a piece last year when he drew a parallel between the anthem protests and the civil rights movement. He said that the point of such protests and movements is not so much to convince your contemporaries, who are likely to have set opinions, but to reach out to future generations and make them understand what is right.

This is an idea worth holding close in the field of public policy. When a lot of what you advocate for goes against the grain of conventional thinking, it is reassuring to believe that things will take a turn for the better in the future.

The Omnipotence of Advertising

Warning: Mild spoilers for the second season of Westworld to follow.

Westworld is a TV series that is synonymous with heavy conversations around artificial intelligence and what it means to be human in a world where the distinction between man and machine is blurry. It captures the zeitgeist well, tapping into the elemental fear of robots taking over and juxtaposing it with the ethical dilemma of treating the same robots as equals if they were to attain consciousness. But in the second episode of the ongoing season, a conversation between two characters, one old and one young, shows something else at the core of the fictional theme park: advertising.

The older character is deliberating on whether to invest in Westworld. According to him, the technology is great and it is all very well to create an immersive world that is fantastical, but these are not sufficient reasons to spend his money. The younger man interjects and says that what the theme park offers to the people who run it is far more valuable than what it offers to its guests. And this, he says, is the ability to peek into the lives of people, see what they desire, and use it for advertising and marketing. His argument is accepted.

This is a clever nod to the real world that already exists today, a world in which advertising is all pervasive and is made possible by the proliferation of data and the ways in which it can be processed. It is sobering to reflect on the impact this might have on the development of new technologies.

I’m looking forward now to see if the upcoming instalment of Jurassic World, set in another theme park, also manages to throw in a reference to advertising!

The Politics of Last Resort

The recent assembly elections in Karnataka presented yet another display of resort politics, of elected politicians being herded off to swanky resorts to isolate them from the attention of opposing parties and prevent them from defecting. This charade, though common, never fails to make for a viewing that is both amusing and cringe-inducing. It also struck me as a uniquely Indian phenomenon. Until, that is, I came across this incident that happened in the US a few years ago.

The facts are reminiscent of a Hollywood caper. In 2003, after Republicans in Texas tried to take gerrymandering too far, more than fifty Democrats decided to abscond to ensure that a quorum would not be formed in the House. A hunt for the missing legislators soon followed, with a toll-free number set up and the Texas Rangers roped in. Finally, in a hilarious passage sprinkled with mundane details and deadpan in its delivery, the report says:

On Monday night, the delinquent Democrats were found at a Denny’s restaurant in Ardmore, Okla., 30 miles north of the Texas border. They were holed up at a nearby Holiday Inn, where they said they were discussing strategy.

It is fascinating to see political actors displaying such similar behaviour, of bunching together and jumping ship, in two very different jurisdictions. The US case also shows that this may even be necessary at times, if it is done for a just cause and where no other alternatives are available. Finding a case like that in India would go some way towards tempering the disappointment brought by the current version of resort politics in the country.

Wanted – A Solicitor General for India

I found out today that India has not had a Solicitor General for the past seven months. The last Solicitor General, Ranjit Kumar, resigned from the post in October of last year.

The Solicitor General is the second highest ranking legal advisor to the government, only below the Attorney General. While the post of a Solicitor General is not a constitutional one like that of the Attorney General, a person appointed to it still performs a very important role. It is necessary for the government to be well-represented so that the strictest legal scrutiny and defence is made possible for its own policies. At present, the Additional Solicitor Generals are bearing the responsibility of representing the government in important cases. This status quo is not ideal.

Such dithering over an appointment to the role does not reflect well on the Appointments Committee of the Cabinet (AAC), the body responsible for filling the vacancy. One can only hope that an appointment to this important position is made sooner rather than later.

Categories Law

Free Speech Then and Now

The right to freedom of speech and expression under the Indian Constitution is not an absolute right. A cursory look at Article 19 (2) reveals there are no less than eight broad restrictions that can be imposed on a citizen’s right to free speech and expression. It states that a law will continue to be valid

…in so far as such law imposes reasonable restrictions on the exercise of the right conferred by the said sub-clause in the interests of the sovereignty and integrity of India, the security of the State, friendly relations with foreign States, public order, decency or morality, or in relation to contempt of court, defamation or incitement to an offence.

I revisited this article recently after encountering a fascinating passage in the Law Commission Report on hate speech. This passage notes the wording of one of the very first versions of this provision, which said that the right would be subject to the following proviso (this proviso was, in fact, dropped when the freedom clause was introduced during the Constituent Assembly debates):

Provision may be made by law to make the publication or utterance of seditious, obscene, blasphemous, slanderous, libellous or defamatory matter actionable or punishable.

Here are a few thoughts I had on this:

One, we seem to have more restrictions on this right at present than were originally proposed. The Constituent Assembly debates and a few subsequent cases and amendments hold the key to understanding the circumstances behind such expansion. It would be fascinating to further study this evolution.

Two, it is encouraging to see that obscenity and blasphemy did not make the final cut. But this is tempered by the existence of decency and morality in the article as it exists today. We could do with thicker skins in this country.

Three, should defamation have a place as a reasonable restriction, particularly when recourse in the form of civil damages is available? The continued existence of criminal defamation in the rule books points towards the shortcoming of this status quo.

A Four-Fold Classification of Data

One of the vexing issues while talking about data protection is the various ways in which data can be classified. Public or private, identified or identifiable, data or metadata; the glut of classifications makes it difficult to achieve consistency.

In The Master Algorithm, Pedro Domingos classifies data into four types:

  • Data that is shared with the world at large. This includes blogs, tweets, reviews on websites, etc.
  • Data that is shared with a limited circle of people. The simplest example under this category would be interactions on social media platforms.
  • Data that is shared with companies. This can often happen without the individual knowing that a company has data belonging to him.
  • Data that individuals do not wish to share with anyone, which is quite self-explanatory.

This is a simple enough classification. That said, there is scope for an overlap between the four types of data. As Domingos is quick to point out, the data from the second category could well fit the third as well, since modes of communication are often controlled by a few companies.

Despite this caveat, the classification is useful. Its simplicity works in its favour by helping more people be a part of the conversation around data protection. And one can hope that more conversations will act as a springboard for developing fresh insights into how we look at our data.

The Rising Spectre of Populism

The Guardian ran this headline last week:

‘He’s not a populist, he’s popular’: Nikol Pashinyan becomes Armenian PM.

For some context, the article is about the change of leadership in Armenia and the quote within the headline comes from one of the new PM’s allies.

Notice the defensive edge to the declaration. Would such a statement have made it to the headlines a few years ago? I’m not so sure. It would have been safe to assume that a person elected to lead a country enjoyed significant support from the general populace. This is no longer the case. Perhaps we have Donald Trump to thank for this. After all, he managed to get elected despite riding a wave of divisive politics and losing the popular vote.

The more pertinent question would be to ask if this trend is here to stay. If, going forward, being popular would no longer be enough for politicians to stake a claim to widespread support and legitimacy; they would also have to dissociate themselves from the spectre of populism.

A Quid Pro Quo Life

While reading Pedro Domingos’ The Master Algorithm, a book about machine learning, I found the following extract that talks about life in the digital age:

Every transaction works on two levels: what it accomplishes for you and what it teaches the system you just interacted with. Being aware of this is the first step to a happy life in the twenty-first century.

To reassure us that this is a positive development, Domingos goes on to say that it is better to think of a computer as a tool to serve us rather than as an adversary. And what it learns from us helps it to serve us better.

I have three thoughts about this.

One, this element of reciprocity is already present in transactions between humans. A shopkeeper I buy something from, is learning about me and, by extension, his customer base. The presence of machine learning amplifies this part of our lives.

Two, this logic can be applied to the products offered by companies like Google and Facebook. This would be a more nuanced way of looking at our relationship with these enterprises than the simple, and now over-used, notion of data being the new oil.

Three, in the event that Domingos’ optimism is misplaced and a system happens to be malign, we should have the ability to walk away from a transaction.