Civil Liberties and the amended Information Technology Act, 2000
After close to a decade of dealing with English statutes, European directives and pan-European regulations, I was struck anew by the antique style of Indian draftsmanship on my return. Much of the world is moving away from stiff legal speech and towards plain English. Even England has converted to a simpler, more concise legal rhetoric. India, however, has a peculiar genius for imprecision and euphemism that makes the purpose and implications of the law hard to understand and apply. While it may seem quaint, to pepper a law with terms like ‘inconvenience’, ‘nuisance’ or ‘annoyance’, the language fails to convey the seriousness of the offences being defined. A reading of the Information Technology Act, 2008, in its new incarnation incorporating the latest amendments and rules (ITA), is a case in point.
Legal draftsmen inevitably wrestle with the age-old dilemma of the generic versus the specific, the potential dangers of a broad definition versus the built-in obsolescence of a narrow spotlight. The crafters of the ITA, in their admittedly admirable attempts to redress some of the gaps and ambiguity in the original law, appear to have struggled in their efforts to strike a balance between clarity and flexibility. While the new avatar is certainly an improvement in some areas, one can’t help but regret the missed opportunity to make necessary changes. Most importantly is the negative impact of the occasionally sloppy and sometimes overly wide drafting on deeply cherished fundamental rights and personal liberties.
Among other things, the ITA has sought to address and improve aspects such as technology neutrality, data protection, phishing and spam, child pornography, the liability of intermediaries and cyber terrorism. While many of these amendments are a step in the right direction, the actual drafting that implements the high level objectives suffers in many respects. For example, the previous emphasis on ‘digital signatures’ has shifted to the technologically neutral ‘electronic signatures’ but the changes have not been carried out thoroughly enough to expunge the old concept entirely. The current law is a bit of an abnormal document in that it contains elements of both concepts, which some attention to detail could easily have averted. Another example is that the provisions meant to combat spam and phishing end up using the dreaded ‘annoyance’ and ‘inconvenience’ terminology with the effect of casting the net of criminality over far more than is appropriate. For example, mail sent with the purpose of causing ‘annoyance’ or ‘inconvenience’ (not exactly the worst offence in the offline world) could put someone behind bars.
An important set of well intentioned but woefully inadequate provisions are those relating to the protection of data. The absence of a specific law on data protection had, in itself, garnered much criticism both within the country as well as in the context of international transactions and outsourcing. The old Act offered the feeble protection of a single provision (section 43) that dealt with unauthorised access and damage to data. In an attempt to meet industry demands and international market standards, the ITA introduced two sections that address civil and criminal sanctions. While this exercise understandably falls far short of a comprehensive law relating to data (being squeezed into an omnibus piece of technology related legislation, rather than one geared up only to deal with data), there was considerable anticipation of its role in papering over the existing cracks and provide a workable, if temporary, data protection regime.
However, the attempt is such a limited one, and so replete with shortcomings that the need for a ‘proper’ data protection law still stands. Given the proposed initiation of the UID scheme, in particular, there is a compelling need for a robust and intelligent law in this regard. Most other countries’ regimes clearly do at least the following:
- define and classify types of data (for example, in most European countries, ‘personal data’ is any data that identifies an individual, ‘sensitive personal data’ is data that reveals details of ethnicity, religion, health, sexuality, political opinion, etc.),
- fine-tune the nature of protection to the categories of data (i.e., greater standards of care around sensitive personal data),
- apply equally to data stored offline and manually as to data stored on computer systems,
- distinguish between a data controller (i.e., one who takes decisions as to data) and a data processor (i.e., one who processes data on the instructions of the data controller),
- impose clear restrictions on the manner of data collection (for example, must be obtained fairly and lawfully),
- give clear guidelines on the purposes for which that data can be put to and by whom (often involving a consent requirement that gives the individual a great degree of control over their data),
- require certain standards and technical measures around the collection, storage, access to, protection, retention and destruction of data,
- ensure that the use of data is adequate, relevant and not excessive given the purpose for which it was gathered,
- cater for opt-in and opt-out type regimes, again to provide individuals with a measure of control over the use of their data even after the stage of initial collection (which has a huge impact on invasive telemarketing or unsolicited written communication)
- impose a knowledge requirement and procedures for allowing individuals to seek information on what data is held on them, and
- create safeguards and penalties that are well tailored to breaches of any of the above.
Unfortunately, and perhaps understandably, the ITA barely begins to scratch the surface of what a good data protection regime entails. The provisions that it does introduce (sections 43-A and 72-A) have glaring inadequacies. Briefly:
- the term ‘sensitive personal data or information’ is used indiscriminately without any definition,
- the provisions only cover electronic data and records, not data stored in non-electronic systems or media,
- they offer no guidance on most of the principles set out above such as in relation to accuracy, adequacy, consent, purpose, etc.,
- in the absence of the controller-processor distinction, liability is imposed on persons, who are not necessarily in a position to control data, even if it is in their possession,
- civil liability for data breaches only arises where ‘negligence’ is involved (i.e., failure to have security procedures or failure to implement them correctly will not automatically result in damages unless negligence is proven),
- similarly, criminal liability only applies to cases of information obtained in the context of a service contract, and requires an element of ‘wilfulness’, or a disclosure without consent or in breach of a lawful contract – this is a very limited remit aimed largely at preventing disgruntled or unscrupulous employees from dealing in company/customer data.
For these broad reasons, we can see that even the amended ITA disappoints those who expected a greatly improved regime in relation to data. It is widely anticipated that the UID scheme, which poses so many potential data protection issues, will serve as a catalyst for a standalone law that is on par with the more sophisticated regimes that function very well in other countries. One great feature common to most of those regimes is that they are consumer/individual focused. The freedom and privacy of the individual is the central concern of protection. Our ITA seems far more concerned with providing corporates with a stick to beat errant employees with, and with catering to the needs of the outsourcing and IT industries. It remains to be seen whether the UID scheme will merely galvanise some targeted legal action covering UIDs rather than generating a broad based piece of legislation.
In addition to the criticisms levelled at the data protection provisions, the other large subset of concerns has been in relation to the civil liberties implications of the ITA. There has been some horror expressed in various forums and media about the ITA contributing to the growth of a police state, to severe curtailment of the freedom of speech and expression, to the invasion of privacy, and to the disproportionate severity of penalisation for offences that are placed on crimes committed in cyberspace compared to crimes committed in the hear and now. Sadly, this is true to a large extent given the clunky treatment of ‘cyber terrorism’, the intolerable pre-censorship that is enabled by the blocking of websites, the broad approach to the monitoring and collection of data, and the demanding obligations of intermediaries to cooperate with interception, monitoring and decryption of data for poorly defined reasons.
While our Constitution’s fundamental rights chapter, which enshrines certain basic, democratic, and profound rights, might not have the same vocabulary of due process as we see in the US, it nevertheless requires restrictions to be reasonable. Precedents and the wider jurisprudence in the field have further developed the concepts of checks and balances, procedural safeguards and legitimacy of restraints that a functioning democracy like India must accord to its people. It can be argued that several provisions of the ITA cause significant tension with the right to freedom of speech and expression, the right against self-incrimination, the right to equality before the law, and the right to practice a trade or profession. To briefly deal with the worst offenders in the IT Act, I have divided them into some broader topics:
Some of the most excessive provisions relate to the free hand with which public access to websites can be blocked. Previously, there was some hope that the rules yet to be formulated in connection with section 69-A would offer some procedural safeguards. The recently notified rules do contain details – in the bureaucratese that we have come to expect – of the process to be followed by the designated functionaries. They also permit the concerned person or intermediary to submit a reply and clarifications to the committee before the decision to block access is taken.
These rules are to a large extent undermined by rule 9 (“Blocking of information in cases of emergency”), which provides that, “…in any case of an emergency nature, for which no delay is acceptable…”, the process will turn into an internal escalation within the department of IT and interim directions relating to blocking access may be issued without giving (him) an opportunity of hearing. There are those who think that, given the events of 26/11, this is wholly justified but the prospect of abuse fills others with dread. The rules may offer detailed time-frames within which orders are made and approved, require reasons to be recorded in writing, provide that emergency orders may be revoked and information unblocked, etc. Regardless, the nature of the process (executive rather than judicial), the ease with which it can be abused, and the fact that the review committee will only meet once in two months to check for compliance, set aside incorrect orders and unblock information, does not offer much comfort. If a site is incorrectly blocked, it could take up to two months for this to be rectified, which could cause a great damage to the owner of the site, and indeed to the wider public that has an interest in uncensored, free speech.
Given that any person can submit a request, it is not unreasonable to anticipate a certain level of frivolous and malicious requests for blocking sites, especially given that the grounds for blocking are very wide (the often repeated set that we are familiar with, namely, in the interest of sovereignty and integrity of India; relating to defence of India/ security of State/ friendly relations with foreign states/ public order and for preventing incitement to commission of any cognizable offences). Without a review committee constantly monitoring and policing the unbridled use of the provisions, the backlog of blocking decisions that may need to be reversed can become a mountain very quickly. The dangers of pre-censorship and the curtailment of dialogue, debate and free speech are even greater in a country with an increasingly thin-skinned populace. Faced with a volatile backdrop of great diversity of religion, political opinions, views on sexuality, morality, obscenity and other highly subjective values and beliefs, there is immense extra-legal pressure on free speech. Thus, there is now a need for greater vigilance so that the thought police do not wield the stick of harsh penalties under the ITA without reason and due process.
Privacy and surveillance
This topic pulls together concerns around the blanket monitoring and collecting of traffic data or information, the interception and decryption (under duress) by intermediaries (now a large superset of ISPs, search engines, cyber cafes, online auction sites, online market places, etc.) and the wide definition of ‘cyber terrorism’ (which ludicrously even casts defamation as a terrorist activity).
Some of the broad concerns in relation to interception, monitoring and decryption in (section 69) are that:
- there is no provision for a clear nexus between an intermediary and the information or resource sought to be monitored or intercepted,
- the usual internationally recognised exception to liability where an intermediary operates purely as a conduit and has no control over data flowing through its network is not clearly spelt out,
- the penalties for non-cooperation are extremely harsh, especially given the absence of a) and b) above,
- these onerous penalties can be said to be in violation of Article 14 as they seem entirely disproportionate. Similar offences and remedies in the Code of Criminal Procedure or the Indian Penal Code prescribe less severe penalties, by an order of magnitude in fact. When the only difference between the offences is the medium in which information is contained, it seems arbitrary to impose a much harsher punishment on an online intermediary than on a member of the public who, for example, furnishes false information to the police in connection with a trial or enquiry.
- the rules made in relation to monitoring, interception and decryption, offer some procedural safeguards, in that they impose a time limit on how long a directive for interception or monitoring can remain in force, a ceiling on how long data can be kept before it is required to be destroyed, etc. However, the effect of these is greatly diluted by exceptions “for functional requirements”, etc. The astonishing irony is that rule 20 requires the intermediary to maintain “…extreme secrecy…” and “…utmost care and precaution…” in the matter of interception, monitoring or decryption of information “…as it affects the privacy of citizens…”!!!!
In a similar vein, there are concerns around the monitoring and collection of traffic data (section 69B) as the section contains an unreasonably long list of grounds for monitoring. These include such extreme excesses as “forecasting of imminent cyber incidents”, “monitoring network application with traffic data or information on computer resource”, “identification and determination of viruses/computer contaminant”, and the catch-all “any other matter relating to cyber security”.
Finally, the main criticism of the ITA approach to ‘cyber terrorism’ is the very wide net that it seeks to cast, looking for a game that has little or nothing to do with the named offence. Amongst the cast of creatures unwittingly caught during this fishing expedition, we find some unlikely victims. In addition to the usual grounds of offence against sovereignty, national security, defence of India, etc., which we have seen in relation to other sections, the ITA considers the following as acts of cyber terrorism – broadly speaking, unauthorised access to information that is likely to cause:
- injury to decency,
- injury to morality,
- injury in relation to contempt of court, and
- injury in relation to defamation.
This would almost be laughable if these grounds were not enacted unto law, posing a threat to civil liberties by their very existence. Other countries have some notion of political ideology, religious case, etc. in their view of terrorism. That (a) to (d) above have been shoehorned into a clause that imposes the stiffest penalty within the entire ITA (life imprisonment) gives even more cause for concern.
In closing, I should reiterate that the ITA includes other deficiencies and worthwhile improvements alike, but an article focusing largely on the data protection and civil liberties aspects cannot reference them all.