Addendum: The Impact of the S. 2(q) Judgment upon the Marital Rape Exception

Previously on this blog, we have discussed the marital rape exception under the Indian Penal Code. Recall that Section 375 of the Indian Penal Code sets out the ingredients of the offence of rape. Exception 2 to Section 375 states that “Sexual intercourse or sexual acts by a man with his own wife, the wife not being under fifteen years of age, is not rape.” I had argued earlier that Exception 2 creates two classes of women – married and unmarried (for the moment, let us ignore the intermediate category of separated women, who fall within S. 376B of the IPC), and accords unequal protection of law to these classes. It does likewise with men, and consequently, infringes Article 14 of the Constitution.

The objection to this line of argument is as follows: the effect of striking down Exception 2 would be to create a new offence altogether: the offence of marital rape. This is, in essence, a legislative task. Consequently, a Court, exercising judicial functions, will be overstepping its jurisdiction if it legislates a new crime.

How persuasive you find this objection depends upon whether you read Exception 2 as classifying acts or classifying persons (a distinction made notorious, of course, in Koushal vs Naz). Yesterday, however, we discussed the judgment of the Supreme Court in Hiral P. Harsora vs Kusum Narottamdas Harsorawhere Justice Nariman, writing for a two-judge bench, struck down S. 2(q) of the Domestic Violence Act on the ground of an Article 14 violation. In my view the Court, in Hiralal P. Harsora, did precisely what it would need to do to strike down the marital rape exception. Recall that S. 2(q) of the DV Act stated:

“….“respondent” means any adult male person who is, or has been, in a domestic relationship with the aggrieved person and against whom the aggrieved person has sought any relief under this Act.

Provided that an aggrieved wife or female living in a relationship in the nature of a marriage may also file a complaint against a relative of the husband or the male partner.”

Just like the marital rape exception, S. 2(q) effectively stipulated that acts of “domestic violence” (defined in S. 3 of the Act) would not fall within the DV Act if they were committed by female and/or persons (who were not relatives of a husband or male partner). The Court reasoned that this classification bore no rational relation with the purpose of the Act, which was to protect women from domestic violence “of all kinds”. The effect of the Court’s judgment was to widen the ambit of the DV Act by ensuring that henceforth females who committed domestic violence could be proceeded against under the Act. Or, if you want to put it another way, the Court legislated a new offence (albeit not a criminal offence): commission of domestic violence by females (who are part of a domestic relationship). Transposing this logic to the marital rape exception the argument is straightforward: the marital rape exception stipulates that husbands who commit rape do not fall within the ambit of S. 375 IPC. This classification bears no rational relation with the purpose of the Section or the IPC (i.e, to prevent and punish crime). The effect of striking down the Exception will be to widen the ambit of S. 375 by ensuring that henceforth husbands who commit rape can be proceeded against under S. 375. This might amount to “legislating a new offence” – but the Court just did that last week.

An immediate objection may be raised: S. 2(q) of the DV Act was part of the definitional section, while S. 375 IPC stipulates the ingredients of the crime of rape. The two are not equivalent, therefore, and it is only in the latter case that constitutional invalidity would result in legislating a new offence. In my view, however, the distinction is only a semantic one. As described above, what the Court is doing in the two cases is the same thing, and the result is the same. In both cases, two things are happening: an act that was not previously an offence is now an offence (commission of domestic violence by females and commission of rape by husbands); and a class of persons that were previously exempted from liability for committing the same act (females committing domestic violence and husbands committing rape). The basis is also the same: the two enactments tackle a certain kind of offence (domestic violence, rape), and consciously leave out a class of persons from liability, even though that class of persons might – in the non-legal sense – commit exactly that offence. Whether this is done through the definitional section or the ingredients section is a question of legal form, and not relevant to an Article 14 enquiry.

However, at this stage, a further objection may be made: the previous argument only deals with the act that constitutes the offence. S. 375, however, does not merely punish sexual intercourse, but punishes sexual intercourse without consent. The marital rape exception is based upon the premise that within marriage, spousal consent to sexual intercourse is presumed. Whatever the validity of this assumption, it is open to the legislature to make it; more importantly, if this is the basis of Exception 2, then it takes the provision out of the ambit of Article 14 altogether, and also dispenses with Hiralal P. Harsora as precedent.

It is no doubt true that the underlying assumption of Exception 2 is the doctrine of presumed consent within marriage. However, that is not what the Exception says. Had the Exception stated that “Sexual intercourse by a man with his own wife is deemed to be with her consent“, then it would have been a different matter.

It might be objected, however, that this is precisely the semantic distinction that we have argued against above. Section 375 is entirely about the question of consent. Consent is built into the definitional clauses. Consequently, the only reasonable way to read Exception 2 is to read it as stipulating deemed consent.

Such a reading, however, would put Exception 2 at odds with the rest of the IPC. Consider Section 87 of the IPC:

“Nothing which is not intended to cause death, or grievous hurt, and which is not known by the doer to be likely to cause death or grievous hurt, is an offence by reason of any harm which it may cause, or be intended by the doer to cause, to any person, above eighteen years of age, who has given consent, whether express or implied, to suffer that harm; or by reason of any harm which it may be known by the doer to be likely to cause to any such person who has consented to take the risk of that harm.”

Now, while Exception 2 states that marital rape is not “rape”, legally defined, it does not exempt marital rape from falling within other sections of the IPC. Prima facie, marital rape could constitute hurt (S. 319), wrongful restraint (S. 339), use of criminal force (S. 350), and sexual harassment (S. 354A). Now, under the doctrine of presumed consent, none of these offences could apply to sexual intercourse between husband and wife, since Section 87 would kick in. Consequently, there ought to have been a marital exception – along the lines of Exception 2 to Section 375 – for each of these provisions. However, there isn’t. Consequently, the doctrine of presumed consent cannot be taken to be the only explanation for the marital rape exception.

In fact, the explanation that is most commonly given – and which was invoked during the recent public debates on marital rape – is that the Exception is necessary in the interests of family unity and integrity. Whatever one may think about the merits of this answer, and the further argument that it family unity is a legitimate legislative purpose under Article 14, it is important to note that this argument is also excluded by Hiralal P. Harsora, since Justice Nariman, in that case, clearly derived the legislative purpose from the statement of objects and reasons and preamble of the DV Act, and did not engage in a roving enquiry about other possibly justified purposes. Under this framework, it is immediately clear that family unity is no part of the legislative purpose underlying the IPC as a whole, or Chapter XVI (“offences affecting the human body”), or even S. 375. The IPC is a criminal statute, and its purpose is the prevention, detection, deterrence, and punishment of crimes. And it is impossible to conceive of a defence of the classification drawn by the marital rape exception that bears a rational relation to these goals.

Consequently, I would argue that the reasoning in Hiralal P. Harsora, if applied consistently, leaves the Supreme Court with no other option but to strike down the marital rape exception as unconstitutional, if and when a challenge was brought before it.


Filed under Article 14, Equality, Marital Rape

The Invalidation of S. 2(q) of the Domestic Violence Act: A Comment

Last week, a two-judge bench of the Supreme Court struck down Section 2(q) of the Protection of Women from Domestic Violence Act, 2005 [“DV Act”], on the basis that it violated Article 14 of the Constitution. Section 2(q), which is part of the definitional clause of the DV Act, read:

“…”respondent” means any adult male person who is, or has been, in a domestic relationship with the aggrieved person and against whom the aggrieved person has sought any relief under this Act:

Provided that an aggrieved wife or female living in a relationship in the nature of a marriage may also file a complaint against a relative of the husband or the male partner.”

To understand what, precisely was at issue, it is also important to set out the definitions of “aggrieved person” and “domestic relationship”. Section 2(a) defined an “aggrieved person” to mean “any woman who is, or has been, in a domestic relationship with the respondent and who alleges to have been subjected to any act of domestic violence by the respondent.” Section 2(f) defined domestic relationship as “a relationship between two persons who live or have, at any point of time, lived together in a shared household, when they are related by consanguinity, marriage, or through a relationship in the nature of marriage, adoption or are family members living together as a joint family.”

The effect of Section 2(q), therefore, was that insofar as a domestic relationship was concerned, an aggrieved woman could proceed only against male perpetrators of domestic violence. However, if the domestic relationship was a marriage or a relationship in the nature of a marriage, the aggrieved woman could file complaints against the relative of her husband/male partner. It is important to note that it is, by now, settled law, that under the proviso to Section 2(q), women could be respondents. Consequently, the distinction drawn by S. 2(q) was between marriages/relationships in the nature of marriage on the one hand, and other domestic relationships on the other. In the former case, female relatives of the husband/male partner could be made respondents, while in the latter, a respondent could only be an “adult male”.

The Supreme Court found that this distinction was irrational, arbitrary, and contrary to Article 14. For the most part, the judgment is a textbook application of Article 14 doctrine, and needs no comment. However, a couple of interesting issues do arise out of the judgment, which deserve to be examined.

The first is the issue of legislative purpose under Article 14. Relying upon Shashikant Laxman Kale vs Union of India and Harbilas Rai Bansal vs State of Punjab, Justice Nariman held that the Statement of Objects and Reasons and the Preamble of the DV Act must be examined to discern its purpose. Reading the two together, he found that the purpose of the Act was to “provide for effective protection of the rights of women who are victims of violence of any kind occurring within the family.” (emphasis his) (paragraph 16) In light of the wide definition of ‘domestic relationship’, which included members of both sexes (paragraph 18), the amendments to the Hindu Succession Act that now made women co-parceners in a joint family (paragraph 18), the gender-neutral definition of “domestic violence” under Section 3 of the DV Act (paragraph 19), and the fact that the remedies under the Act (such as protection and residence orders) could easily be defeated if “respondent” was limited to adult male persons (paragraph 20), he then held that the classification under S. 2(q) failed the rational relation test under Article 14. This was not just true for “male”, but for “adult” as well, since it was easy to envisage 16 and 17- year olds engaging in acts of domestic violence within shared households (paragraph 24). The linchpin of Justice Nariman’s opinion, which he repeated, was that “the classification of “adult male person” clearly subverts the doctrine of equality, by restricting the reach of a social beneficial statute meant to protect women against all forms of domestic violence.” (paragraph 31) The phrase “domestic violence of any kind” was repeated in paragraph 36.

In short, therefore, the Court struck down S. 2(q) on the basis that the distinction it drew between the persons who could be arraigned as respondents in the case of marital relationships, and other kinds of domestic relationships, bore no rational relation with the purpose of the Act, which was to protect women against domestic violence of “any kind”, or of “all forms”. This legislative purpose was drawn from its statement of objects and reasons and the Preamble.

It is important, however, to draw a conceptual distinction between two kinds of “legislative purposes”. In the first sense, “legislative purpose” is what the legislature actually had in mind (to the extent that collective purposes make sense) when enacting the statute, something that a Court can determine by looking at the text and surrounding documents of the law. This is what Justice Nariman did in the present case. Call this the “intended purpose“. In the second sense, “legislative purpose” is a purpose that can be justifiable attributed to a statute, regardless of whether or not it was actually within the contemplation of the legislature while drafting the law. Call this the “justified purpose“. In this case, after having found that the stated purpose of the Act was to protect women against domestic violence of “all kinds”, and that the S. 2(q) classification did not serve this purpose, Justice Nariman did not ask (and indeed, the State did not propose) whether S. 2(q) could be plausibly justified in relation to any other possible legislative purpose. Here is one possible candidate:

The Domestic Violence Act understands “domestic violence” as not simply violent acts committed by one person upon another within a domestic setting, but as a problem that flows from differential, structural power relations between men and woman in the family (which is why only women can be complainants under the Act), and therefore, primarily seeks to prevent male-on-female violence. While we may disagree with this framing, it is within the realm of legislative discretion to make this call. This accounts for S. 2(q). However, the legislature was also cognisant of the fact that the marital relationship is a space where women are specifically vulnerable, in no small part because in many circumstances they must leave their homes and live with their husband’s family. Consequently, the legislature chose to carve out a proviso to S. 2(q) to deal with the heightened vulnerability of women in marital relationships.

I am not arguing that this restated purpose of the Domestic Violence Act would necessarily clear Article 14 scrutiny. It might be argued, for instance, that even if one is to accept the argument that domestic violence is structural and institutional, acts of domestic violence can and are committed by both men and women – and so, even if we were to take the above argument on its own terms, it would fail the test of rational classification (in that case, however, the Court would also have to explain why the legislature’s identification of the specific harm that it was seeking to prevent was irrational and could be overridden in a judicial enquiry). What I am arguing, however, is that principles of judicial deference and the presumption of constitutionality would require the Court to adopt a generous approach towards the determination of legislative purpose, which would include, at times, reconstructing legislative purpose in a manner that would make the strongest case for the constitutionality of the law. If, even then, the law failed Article 14 scrutiny, then of course, it would need to be struck down.

As an aside, it is also interesting to take note of the path that this judgment did not take. Recall that in Yusuf Abdul Aziz vs State of Bombay, the Supreme Court upheld the constitutionality of adultery law against a gender-equality challenge (women are not liable in case of adultery) on the basis that it was saved by Article 15(3) of the Constitution (“special provisions for women and children”). An argument could have been made in this case that exempting a class of women from legal liability was exactly what was done in Yusuf Abdul Aziz, and upheld under Article 15(3). The problems with that approach are too many to list out here, and so, it is certainly a good thing that the Court showed no signs of retracing its steps along that road.



Filed under Article 14, Equality

Preliminary Thoughts on the Triple Talaq Case

The Supreme Court is presently considering the legal validity of triple talaq under Muslim personal law. According to reports, yesterday, the Centre filed its affidavit before the Court, stating that triple talaq violates gender equality and women’s dignity, and also that “no undesirable practice can be elevated to the status of an essential religious practice.”

While I do not have access to the Centre’s affidavit, the phrase “essential religious practice” appears to have been used in response to the Muslim Personal Law Board’s affidavit, which, inter alia, sought protection for triple talaq under Articles 25 and 26 of the Constitution. In my view, this indicates a line of argument which would take the Supreme Court down the wrong path, and ought to be resisted. The use of the phrase “essential religious practice” (as has been discussed many times on this blog) is an integral part of the Supreme Court’s religious freedom jurisprudence under Articles 25 and 26, and acts as a threshold test for according constitutional protection to religious practices. Conceptually, however, triple talaq does not come within the category of practices that fall within the scope of Articles 25 and 26.

There are two reasons for this. The first is based in precedent. In Narasu Appa Mali, the Bombay High Court held that personal laws (which had not been codified under a statute) were not to be tested on the touchstone of Part III of the Constitution. This proposition was affirmed by the Supreme Court in Krishna Singh vs Mathura Ahir. Article 25 of the Constitution expressly states that “subject to public order, morality and health and to the other provisions of this Part, all persons are equally entitled to freedom of conscience and the right freely to profess, practise and propagate religion.” If personal laws fell within the scope of Article 25, therefore, they would be “subject to other provisions of [Part III]”. At the same time, as per Narasu Appa Mali and Krishna Singh, personal laws are exempt from Part III scrutiny. In its counter-affidavit, therefore, the Muslim Personal Law Board rests its case upon two legal prongs that are contradictory. It cannot say that personal laws are exempt from Part III scrutiny, and simultaneously argue that they are protected by Articles 25 and 26.

The proposition that personal laws do not fall within the scope of Articles 25 and 26 is further buttressed by the debates during the framing of the religious freedom clauses. As Ambedkar famously argued, in a speech that we have discussed often on this blog:

“The religious conceptions in this country are so vast that they cover every aspect of life, from birth to death. There is nothing which is not religion and if personal law is to be saved, I am sure about it that in social matters we will come to a standstill. I do not think it is possible to accept a position of that sort. There is nothing extraordinary in saying that we ought to strive hereafter to limit the definition of religion in such a manner that we shall not extend beyond beliefs and such rituals as may be connected with ceremonials which are essentially religious. It is not necessary that the sort of laws, for instance, laws relating to tenancy or laws relating to succession, should be governed by religion.”

I focus on succession, because marriage – like the law of succession – belongs in the domain of personal law (i.e., the law that is based upon the personal status of the parties) which, as Ambedkar pointed out, was never meant to be protected by Articles 25 and 26. This, along with Narasu Appa Mali and Krishna Singh, makes it clear that the issue of whether triple talaq is an “essential religious practice” under Islam is irrelevant to the present enquiry, and the Court should refrain from going into a question that will – yet again – make it the arbiter of religious doctrine (note, in particular, that the Centres affidavit seems to argue that any religious practice that runs counter to constitutional principles cannot, by that reason, be called an “essential religious practice”. This is an interesting legal fiction to press before the Court – and the Court is no stranger to adopting such legal fictions – but it remains a highly problematic one, for reasons that have been extensively discussed before.)

There are, of course, other avenues open to the Court. While issuing notice, the Court called triple talaq a “customary” practice. In Madhu Kishwar vs State of Bihar, the Supreme Court held that customary laws would be subject to Part III (while personal laws remained exempt). However, the distinction is superficial at best, and furthermore, if – as in Narasu Appa Mali – bigamy under Hindu traditions was held to fall within the domain of personal law, then it is unclear how triple talaq will not fall within personal law. Another option would be to import the “essential religious practices” test from Article 25 into the domain of personal law, on the basis that personal law, like claims under Article 25, pertains to religion. This, however, would be a somewhat odd innovation sixty-three years after Narasu Appa Mali, especially in light of the fact that the Bombay High Court, in that case, applied the essential practices test specifically while adjudicating under Articles 25 and 26, and refrained from applying it in holding that personal laws were not subject to Part III.

I would suggest, therefore, that under existing constitutional jurisprudence, a two-judge bench of the Supreme Court, being bound by Krishna Singh, cannot invalidate triple talaq. Of course, that is not dispositive of the issue. The judgment in Narasu Appa Mali was controversial, and there are strong legal and constitutional arguments against its reasoning. Anirudh Krishnan, for instance, persuasively argues that Narasu was incorrectly decided (see, also, the comprehensive debate between Anirudh Krishnan and V. Niranjan, in the comments section of the same post). Perhaps, then, in 2016, it is time a Supreme Court bench of three judges reviewed Narasu and Krishna Singh, and overruled them as incorrectly decided.

In my view, there is no doubt that triple talaq is an unconstitutional practice, and should be judicially invalidated. However, the Supreme Court ought to refrain from the temptation of repeating its mistake in the Make-Up Artists Case, and riding roughshod over existing precedent in order to achieve a progressive outcome. There is a correct and constitutional way of doing this, which is to refer the case to a three-judge bench, which can then re-examine the question of whether personal laws are subject to Part III of the Constitution, and correct its earlier errors on this score.


Filed under Equality, Sex Equality

The Bihar High Court’s Prohibition Judgment: Key Constitutional Issues – IV: (More on) Punishment

(In this guest post, Manish carries forward the conversation on the punishment clauses that was initiated by Abhinav yesterday.)

In a strongly worded and well-reasoned judgment, the Patna High Court on Friday struck down the Bihar government’s attempt to impose total prohibition in the state through amendments to the Bihar Excise Act, 1915 (“the Act”). Other aspects of the judgment have been dealt with elsewhere on this blog, and in this post I will discuss the Court’s reasoning with regard to the penal provisions of the Act, particularly in relation to due process. I argue that in deciding this issue, the Court used its strongest words, but not its strongest reasoning.


As part of the amendments, punishments under the Act were drastically enhanced, ranging from a minimum of 10 years in jail and extending up to life imprisonment, and fines ranging from 1-10 lakh rupees. In the writ petitions before the Court, the enhanced punishment was challenged on three main grounds: first, that the imposition of high minimum punishments under sections 47, 53 and 54 was disproportionate to the offence and took away judicial discretion even where mitigating circumstances might warrant a lesser sentence; secondly, that the confiscatory power provided to the State to seal premises and destroy or forfeit property under sections 68A and 68G was excessive; and thirdly, that the mechanism of collective fine introduced in section 68-I was vague and lacked procedural safeguards. The only response of the State on record was that the legislature possessed “plenary power to legislate and provide for punishment” and that the Court could not interfere with legislative wisdom.

At the outset, it is submitted that the existence of “plenary power to legislate” is hardly an adequate response in a case where it is the exercise of the said power that is being challenged, particularly given that under Article 13 of the Constitution, this power is expressly subject to the provisions of Part III. Nevertheless, the Court proceeded to consider the amendments in some detail, and found all the challenged provisions to be in violation of Articles 14 and 21.

Reverse Onus and procedural due process

Importantly, the court prefaces its analysis with an examination of section 48 of the Act (itself not under challenge), which reverses the burden of proof for all offences under the Act. It explains how the lack of due process in cases where stringent punishment is envisaged can make the burden on the accused more onerous:

“…punishments by itself cannot be seen but have to be seen along with the procedure, for, the procedure may create certain liability, which, coupled with the punishment, would made things worse.” (Para 89.03)

The Court observed that while a reverse onus clause, such as section 48, would not by itself be unconstitutional, the presumption it created against the accused would render the substantive penal provisions subject to a higher standard of scrutiny. In doing so, it made a critical link between substantive and procedural due process i.e. using the absence of procedural safeguards to decide the validity of substantive provisions of the law: a variation on the integrated Article 14-19-21 approach that courts have followed since Maneka Gandhi.

Collective fines and subjective satisfaction

The Court then considered the provision relating to collective fines, under section 68-I, which reads as follows:

If the Collector is of the opinion that a particular village or town or any locality within a village or town or any particular group/community living in that village or town have been repeatedly violating any of the provisions of this Act or are habitually prone to commit an offence under this Act or are obstructing the administration of this Act, then the Collector may impose a suitable collective fine on such group of people living in such area of the town or village and may recover such fine as if they were Public Demands under the Bihar & Orissa Public Demands Recovery Act, 1914.

The Court observed that the entire process under section 68-I was dependent on the subjective satisfaction of the Collector. There were no guidelines for the identification of a locality or group within a village or town, no provision for any of the affected persons to be heard prior to imposition of the fine, and no means of appeal against the decision. In these circumstances, it struck down the provision as being in violation of Article 14 and 21 of the Constitution, terming it “draconian, completely vague, uncertain and unlimited”.

It should be noted that the provision of collective fines under the Bihar Excise Act is not a unique phenomenon. Most notably, section 16 of the SC & ST (Prevention of Atrocities) Act, 1989, read with section 10A of the Protection of Civil Rights Act, 1955, empowers the State Government to impose collective fines in cases of atrocities against members of the Scheduled Castes and Scheduled Tribes. However, these provisions contain substantially more procedural safeguards: the satisfaction of the State Government is to be determined on the basis of an inquiry; the apportionment of fine among the inhabitants of the area is based on the means of individuals to pay; and an appellate process is provided for by means of filing a petition before the State Government, which is to be disposed of only after providing the appellant with a hearing. It is submitted that the Bihar government would do well to emulate these safeguards, should it deem it necessary to continue with the mechanism of collective fines.

Life, liberty and property

The Court finally dealt with the reasonableness of the provisions relating to imprisonment, fine, confiscation and destruction of property. The ground for review was drawn from the requirement under Maneka Gandhi that procedure established by law for deprivation of a person’s life or liberty under Article 21 was required to be just, fair and reasonable. Using this due process requirement, the Court constructs the beginning of a case against excessively stringent or draconian penal provisions without procedural safeguards.

Unfortunately, it does not do so convincingly: while the reasoning is logical, the use of precedent is the shakiest in this part of the judgment. Two of the four cases it cites (Mithu and Dalbir Singh) were situations where the Supreme Court struck down a mandatory death sentence as being in violation of Articles 14 and 21, partly because judicial discretion in determining punishment and taking into account mitigating circumstances was taken away. In fact, one of the other cases cited (Vikram Singh) categorically upheld the validity of section 364A of the Indian Penal Code, holding that where even one alternative was provided, the punishment could not be challenged as being unreasonable or taking away judicial discretion. Relying only on these grounds, the punishments imposed under the impugned sections could be argued to allow for sufficient judicial discretion, and the decision to that extent stands on shaky ground. It is submitted that the court’s initial observations with respect to the reverse onus clause, and the lack of procedural safeguards, form a stronger ground for making a case for violation of due process requirements under Article 21.

Proportionality and substantive due process

This brings us to the final link in the argument – that of proportionality – which the Court does make to some extent. This thread of reasoning is as follows: in light of a reverse onus clause, a higher burden is already placed on the accused by the procedural provisions of the Act. Therefore, the substantive provisions must not be so onerous so as to take away all elements of due process from the accused. For a comparative standard of fairness, the Court looks at the NDPS Act, which also deals with punishments for possession and consumption of prohibited substances. In that Act, the Court observes, the punishment is graded, varying with the quantity of the prohibited substance in question. (Incidentally, the NDPS Act also contains reverse onus clauses, and the Supreme Court while upholding the validity of these clauses had invoked a standard of ‘heightened scrutiny’ while dealing with prosecutions thereunder.)

The Court also makes reference to the 47th Report of the Law Commission of India (1972), dealing with the trial and punishment of socio-economic offences. The suggestions of the report in relation to sentencing, particularly the disapproval of “mathematically identical sentences”, remain relevant even today and the Court takes note of them in relation to the offences under the impugned sections 47, 53 and 54, all of which provide for almost identical sentences of long imprisonment and heavy fines for manufacture, transport, possession and consumption of alcohol. The Court observes that in all three sections, the punishment imposed is disproportionate to the offence, and there is no scope for the Court to award a lesser punishment even when there are mitigating circumstances present.

Finally, in regard to sections 68A and 68G, which makes premises and properties used for committing offences liable to confiscation and sealing, the Court holds that the effect of these provisions is to virtually convert the state into a police state. It observes that the reverse onus clause read with these sections would lead to situations of an innocent owner of the premises being punished for acts committed by tenants, even when she had no knowledge of the same. It holds that the provisions are “clearly draconian and in excess of the balance need to be maintained [for achieving the social objectives of the legislation].” In conclusion, it strikes down all the impugned sections as being unreasonable, draconian and ultra vires Articles 14 and 21 of the Constitution.

The Court’s emphasis on individual due process is rather heartening, especially given the tendency of courts to side with the State in cases where reverse onus clauses and stringent punishments are provided for by law. Elsewhere, Mrinal Satish and Aparna Chandra have argued that in cases of anti-terror legislation, the Supreme Court has adopted a minimalist approach while adjudicating their constitutional validity:

The Court articulates its role in terms of balancing competing interests of national security and civil liberties; it provides broad deference to the legislature, not only to its policy, and its understanding of what is required and permissible to implement the policy, but also by engaging in a fair amount of legal gymnastics to uphold constitutionality of provisions. It evaluates legislative provisions not for their impact on Fundamental Rights of citizens, but to examine whether the provisions further the purpose of the Act on the one hand, and whether there are enough procedural safeguards to prevent misuse on the other. Where in spite of this curtailed review, a provision does not pass muster, the Court takes over the role of the legislative drafter and provides a procedural framework to prevent misuse, or recommends measures for the Parliament’s consideration. Very rarely, if at all, does it invalidate a provision.

In this context, the High Court’s decision to strike down the provisions in their entirety for being in violation of substantive due process norms under Article 21 is a promising step forward. One only wishes that its jurisprudential reasoning had been stronger.


The last word is yet to be said on prohibition in Bihar. In August 2016, the legislature passed a revised Prohibition and Excise Act, and reports have suggested that the State Government is considering notifying the new Act. An analysis of its provisions suggests that several of its penalties are similar to the ones that have been struck down, and it remains to be seen whether the Government makes appropriate changes to ensure adequate procedural safeguards.


Filed under Article 21 and the Right to Life, proportionality, Punishment

The Bihar High Court’s Prohibition Judgment: Key Constitutional Issues – III: Proportionality in Punishment

(This is a guest post by Abhinav Sekhri, cross-posted with permission from his Proof of Guilt blog.)



The separate opinion of Justice Singh covers seven issues and he saves the best for last with his analysis of punishment clauses coming right at the end at Paragraph 89. Effectively, the law criminalised anything one did with alcohol (i.e. acts/omissions associated with its production, possession and consumption) contrary to the Act and rules. Punishments for these offences are neatly summarised into three categories in the judgment: (i) incarceration and fine (Sections 47, 53), (ii) confiscation of property (Sections 68-A, 68-G), and (iii) collective fines (Section 68-I) [There are other provisions on punishments in the Act as well that developed on these basic offences]. These were supplemented by the now-standard provision on reversing the burden of proof and presuming the accused is guilty (Section 48). The High Court unanimously held that these penal clauses were unconstitutional.

The Opinion

Out of these three categories, take categories (i) and (ii) separately from category (iii) which is collective fines. The collective fine provision is set aside for obvious violations of Article 21 of the Constitution as it imposed punishments contrary to a procedure established by law as no right of hearing or right of appeal was provided by the clause. Similar provisions for collective fines are present elsewhere that have these procedural safeguards [See, Section 16 of the Scheduled Castes and Scheduled Tribes (Prevention of Atrocities) Act 1989].

Which brings me to the more interesting part of the decision concerning categories (i) and (ii). The primary plank on which these penal clauses are set aside is an argument on proportionality, i.e. the sentence imposed for an offence must be proportionate to the harm purportedly caused by the offence. This is a rather crude summation of what is a highly technical concept in sentencing theory [See, Andrew Ashworth Sentencing and Criminal Justice (5th edn., 2010); Andrew Von Hirsch Censure and Sanctions(1993)]. The High Court located the concept of proportionality in criminal statutes in Article 21 of the Constitution by relying upon the Supreme Court decisions in Mithu Singh v. State of Punjab [(1983) 2 SCC 277] and Vikram Singh v. Union of India[(2015) 9 SCC (502)]. The three judges in Vikram Singh framed the proportionality test as follows: “Courts, however, have the jurisdiction to interfere when the punishment prescribed is so outrageously disproportionate to the offence or so inhuman or brutal that the same cannot be accepted by any standard of decency.

In this case, the harm as per the State of Bihar was the various evils associated with excessive alcohol consumption that are afflicting its people. The response was to introduce penal provisions that had these notable features:

  • A presumption of guilt which the accused must rebut without specifying whether the State had any burden to prove objective facts displaying commission of the offence;
  • Broad constructive liability attracting persons whose property was allegedly involved in the commission of offences. These persons would be subjected to the similar presumption of guilt which they must rebut;
  • Mandatory minimum punishments – warranting that a sentence of at least ten years’ imprisonment be awarded for possessing alcohol and at least five years for consuming it in a public place;
  • Sealing of properties allegedly involved in commission of offences and their confiscation upon conviction;

The Court posed to itself a question – was this response proportionate to the harm – and answered it in the negative as it found the penal clauses were notoriously overbroad and unspecific. There was no gradation of punishment to differentiate minor and major violations. Nor did the punishment clauses show any sensitivity to how different acts posed different harms. The NDPS Act was cited as an example of how such clauses can be provided within constitutional limits (to think that the NDPS Act would appear as a measure for principled criminal legislation!). Notably, Justice Singh did not restrict his opinion to the badly drafted punishment provisions for his proportionality analysis. It also looked at the various procedural conveniences incorporated by the 2016 Amendments, i.e. the reversed burdens of proof and coercive procedures of confiscation and sealing of property. Although persons could plead ignorance as a defence, broad constructive liability meant far too many people would be unfairly accused and subjected to these coercive procedures while having to prove their innocence. These dangers were not set aside by the stringent punishments imposed on vexatious actions by overzealous state agents.

Some Comments

The absence of any sentencing policy or guidelines has been a common lament for several criminal law scholars in India over the years. This decision puts the problems starkly into perspective. Imposing a minimum ten-year imprisonment for possessing alcohol and confiscating premises where nobody knew alcohol was present there seems to satisfy the outrageously disproportionate standard of the decision in Vikram Singh. However, the Supreme Court in both Mithu Singh as well Vikram Singh was at pains to stress the different position that the death penalty occupied from a sentence of imprisonment. In Mithu Singh, one may recall, Section 303 IPC was set aside because it only allowed for the death sentence without any alternative. The Supreme Court in Vikram Singh put it bluntly: “there are very few and rare cases of sentences of imprisonment being held disproportionate.” The Patna High Court decision is therefore remarkable, for it places the penal clauses in the Bihar Excise (Amendment) Act 2016 in this category of very few and rare cases. To my knowledge, this is the first decision to use the proportionality doctrine to set aside penal clauses despite no death sentence being involved.

There are deeper problems though. The decision in Vikram Singh provided a neat ‘sum up’ of principles on proportionality at Paragraph 52. It was a summing up of principles drawn from Canadian and American jurisprudence – both have specific constitutional prohibitions on certain kinds of punishment. India does not. In fact, the Supreme Court on an earlier occasion in Jagmohan Singh v. State of U.P. [AIR 1973 SC 947] specifically noted that American jurisprudence on prohibiting cruel and unusual punishment [flowing from the Eighth Amendment to the United States Constitution] could not be imported in India because the Indian Constitution did not have similar clauses. Of course, Maneka Gandhi came after Jagmohan Singh and expanded the boundaries of Article 21 allowing us to possibly incorporate the Eighth Amendment jurisprudence. The problems of having such an indeterminate constitutional jurisprudence have been seen with the tussles over the right to privacy and its scope. Therefore, it should be clear that proportionality analysis of punishment clauses does not have a sure footing in Indian constitutional law and the expansion by the Patna High Court may well be considered beyond its current scope.



For now, though, one can only be happy with how proportionality is invoked to set aside statutes inflicting persons with disproportionate terms of imprisonment. This experience is similar to what happened in Maharashtra with the beef ban, where again a proportionality argument was raised to set aside the reverse burden clauses. It shows us that High Courts are stressing on a more principled brand of criminalisation, something that has not been the hallmark of how criminal law developed in India over time. A refusal to budge especially when prohibition was such an integral scheme of the ruling government reflects a commitment to principles that must be lauded. The decision is kind, for it tells the legislature what can be done to fix the law. If the State of Bihar chooses to challenge the decision rather than amend the statute, the Supreme Court will be able to consider these important questions. One hopes it upholds how the Patna High Court applied proportionality to restrain the legislature from abusing its near-plenary powers in sending people away for a decade for having a pint.

1 Comment

Filed under proportionality, Punishment

The Bihar High Court’s Prohibition Judgment: Key Constitutional Issues – II: The Fundamental Right to Privacy

The final, substantive ground of attack upon amended Section 19(4) of the Bihar Excise Act, and its notification of total prohibition, was that of fundamental rights violations. It was on this point that Justice Singh and the Chief Justice differed. Let us first consider the Chief Justice’s reasoning. The Chief Justice pointed to Article 47 of the Constitution, one of the Directive Principles of State Policy, which stated that:

“The State shall regard the raising of the level of nutrition and the standard of living of its people and the improvement of public health as among its primary duties and, in particular, the State shall endeavour to bring about prohibition of the consumption except for medicinal purposes of intoxicating drinks and of drugs which are injurious to health.”

The Chief Justice then observed that, in accordance with well-established precedent, Parts III and IV of the Constitution had to be harmoniously construed. Accordingly, the Chief Justice held:

“When the State has been, asked by the Constitution to make endeavor to bring about prohibition of the consumption except for medicinal purposes of, intoxicating drinks and/or drugs, which are injurious to health, it, undoubtedly, means that making of serious and sincere efforts to take the society to a situation, where it accepts prohibition as a constitutionally declared obligation of the State. When and how it would be done is a question, which needs to be answered by the State depending upon the manner in which the endeavour may be made to bring about prohibition. If it is made in the manner, as has been done in the present case, the endeavour may not survive the test of constitutionality; but if it is done in accordance with law and the constitutional scheme of governance, one cannot be heard to say that his fundamental rights are violated.” (paragraph 39)

In other words, since Article 47 expressly contemplated the State taking legislative measures towards prohibition, the imposition of prohibition (if done in the correct manner) must, by definition, be in conformity with Part III, because any other conclusion would imply a direct clash between fundamental rights and Directive Principles.

What was particularly interesting was how the Chief Justice distinguished Minerva Mills vs Union of India, which had clearly held that the goals under Part IV could not be pursued at the cost of infringing fundamental rights. The Chief Justice held that in Minerva Mills – and in all other similar cases besides – there had already been an existing infringement of a fundamental right, which was sought to be justified by recourse to Part IV (and which was held to be impermissible) (paragraph 52). Here, however, given the text of Article 47, it was clear that the framers had never intended the right to drink alcohol to fall within any of the provisions of Part III, because if they had, they would not have laid out prohibition as a Part IV goal. In other words, the Chief Justice did not undertake an independent analysis of whether or not the right to drink alcohol was part of any fundamental right, but held that as a matter of construction – i.e., harmonising Parts III and IV – it could not be:

That the right to privacy is integral and inseverable facet of fundamental right can no longer be in dispute; but the question of all questions is : whether one’s desire to consume alcohol is a fundamental right? If consumption of alcohol by one is regarded as a fundamental right, then, infringement thereof would, undoubtedly, amount to intrusion into one’s right and would be struck down. When, however, the Constitution obliges the State to make endeavour to bring complete prohibition in respect of consumption of intoxicating drink, consumption of intoxicating drink cannot be treated as a fundamental right.” (paragraph 62)

The Chief Justice finally buttressed this conclusion by citing the Supreme Court judgments in Khoday Distilleries and Kerala Bar Hotels Association, both of which had agreed that the imposition of various stringent controls over the production and sale of alcohol amounted to a reasonable restriction upon Article 19(1)(g), and passed Article 14 scrutiny.

This is a powerful argument, but not entirely persuasive. The relationship between Parts III and IV of the Constitution has had a long and fraught history (one that I have examined in detail elsewhere). It is true that from the mid-1960s, the Supreme Court has consistently held that fundamental rights and directive principles must be interpreted “harmoniously”. For the most part, this harmony has taken three forms: first, directive principles have helped determine the scope of “public interest” under Articles 19(4) – (6); secondly, they have been used to determine whether restrictions upon fundamental rights pass the test of reasonableness; and thirdly, they have been invoked in situations where the abstract wording of a Part III right is open to more than one competing conception. The last one is particularly important for our purposes here, and ought to be understood with an example. In State of Kerala vs N.M. Thomas, the Supreme Court made its famous turn from understanding Article 16(4) as an exception to Article 16(1), to holding it to be a facet of 16(1). This turn was predicated upon a deeper shift: from viewing the equality code under Articles 14-15-16 as espousing formal equality to interpreting it as endorsing substantive equality. In their concurring opinions, Justices Mathew and Krishna Iyer invoked the Directive Principles (in particular, Article 46), to make this shift. The Directive Principles, then, were used to decide which conception of equality was more faithful to the overall constitutional scheme, which included both the abstract text of Articles 14, 15 and 16, as well as Part IV.

It is important to distinguish this use of the Directive Principles as structuring values that may be invoked to infuse concrete content into the fundamental rights provisions from what the Chief Justice did, which was to limit the scope of fundamental rights through the Directive Principles. If, on an independent analysis, it was found that the right to personal liberty under Article 21 included the right to choose one’s way of life (including one’s choice of food and drink) independent of State coercion, then Article 47 could not (in accordance with precedent) be invoked to eliminate that right as far as alcohol was concerned (in fact, it is important to note that Article 47 does not provide a carte blanche for prohibition, but is limited to intoxicating drinks and drugs which are “injurious to health).

The flaw in the Chief Justice’s argument is evident from a perusal of Article 31C of the Constitution, which states that:

“Notwithstanding anything contained in article 13, no law giving effect to the policy of the State towards securing  the principles specified in clause (b) or clause (c) of article 39  shall be deemed to be void on the ground that it is inconsistent with, or takes away or abridges any of the rights conferred by [article 14 or article 19].”

In other words, Article 31C expressly contemplates the possibility that a law enacted in accordance with Articles 39(b) or (c) may, on its merits, infringe Articles 14 or 19, and provides a constitutional override for situations of that kind. If the Chief Justice’s argument was to be accepted, then this override would be entirely redundant.

Justice Singh, in his opinion, began by addressing the Chief Justice’s argument. He – and in my view, correctly – noted that although the Chief Justice had framed his argument in terms of “harmonising” Parts III and IV, in effect, such an interpretation (which, recall, allowed the scope of rights to be curtailed by the DPSPs) would end up making Part III subordinate to Part IV, which was entirely at odds with Minerva Mills (paragraphs 88.02 – 88.03).

Having dealt with this argument, Justice Singh then moved on to considering the scope of the right itself. After noting that prior precedents upholding prohibition had done so by finding it to be a reasonable restriction upon the freedom of trade and commerce, and had not considered the issue of the individual right of consumption of alcohol, he observed that:

“Similarly, with expanding interpretation of the right to privacy, as contained in Article 21 of the Constitution, a citizen has a right to choose how he lives, so long as he is not a nuisance to the society. State cannot dictate what he will eat and what he will drink. We have to view this concept in changing times, where international barriers are vanishing. Any restriction by a State, on the right to choose what to eat and what to drink, apart from being invasion of right of privacy under Article 21, would prejudicially affect free movement and free residence, in any part of territory of India, for the citizens. Keeping in view these factors, a citizen cannot be prohibited from his choice, within the confines of his house, subject to orderly behaviour, of enjoying his drink, which he can procure from any other part of the country, where prohibition is not in force.” (paragraph 88.04)

There are two distinct arguments here that need to be examined separately. The first is the argument from privacy. Justice Singh went through the now well-known series of judgments on the right to privacy, and focused, in particular, on those aspects of the judgments that endorsed the right to privacy as encompassing a right to decisional autonomy. Consequently, he held, in paragraph 88.15, that:

“Thus seen, in my view, the right to decide as to what to eat and drink within the confines of once house, by an individual citizen, would come within the matter of right of privacy, within Article 21 of the Constitution. It is not the case of the State nor any material placed on record that drinking alcohol per se as a responsible citizen is bad or injurious to health. It is abuse thereof that is injurious. On the plea of mere possibility of abuse by some persons, the right of others cannot be abrogated. In my view, if the State starts dictating a citizen what to drink or what not to drink, though the same is not per se injurious to health, it would be a direct intrusion on personal liberty affecting meaningful life. It would be violation of personal liberty guaranteed by the Constitution.” 

In other words, absent a specific showing of harm (which, indeed, was what Article 47 required), which the government had not undertaken in the present case, the denial of choice in matters of private consumption infringed the fundamental right to privacy.

The second aspect of the argument in paragraph 88.04 was based upon the freedom of movement, and framed by Justice Singh as an argument from unconstitutional conditions. In paragraph 88.06, he held:

“Just to illustrate the unreasonableness of this, consider a case of a person born and brought to the Metropolis like Bombay or Delhi, educated there and serving there. Consumption of liquor to him is a part of his life and part of his relaxation, he is accustomed to it. If he has to move to this State and has an option, he would not do so, because he would have to give up his life style. That would infringe not only Article 21 but also Article 19 (1) (d) and Article 19 (1) (e) of the Constitution. He would be inhibited from coming to this State. India is one country.”

While I am normally a strong supporter of the doctrine of unconstitutional conditions – i.e., that the State cannot make a benefit or a penalty conditional upon my waiver of a fundamental right (in this case, my freedom of movement), this argument appears to prove too much. Under this logic, it would mean that state amendments to the IPC that provided stricter or heavier punishments were unconstitutional, since the end result would be that a person would be deterred from moving to those states (in violation of his Article 19(1)(d) rights). And taken to its conclusion, this logic would entail that any issue which even peripherally touches upon fundamental rights can only be legislated by the union Parliament. This clearly cannot be correct.*

In any event, even sans the freedom of movement argument, Justice Singh’s privacy argument remained, which he concluded by noting:

“Thus, in my view, a citizen has a right to enjoy his liquor within the confines of his house in an orderly fashion and that right would be a part of right of privacy, a fundamental right, under Article 21 of the Constitution and, any deprivation thereof would have to withstand the test of Articles 14 and 19 of the Constitution as well.”

There are interesting parallels between Justice Singh’s privacy argument, and the Bombay High Court’s invocation of privacy to strike down the possession section (Section 5D) of the Maharashtra beef ban. In that case, I had noted:

“The Court endorses two different (and complementary) conceptions of privacy. The first is a spatial vision: “the State cannot make an intrusion into his home… the citizen has a right to lead a meaningful life within the four corners of his house…” In other words, invasion of an individual’s “private space” in order to discover whether or not he is eating beef violates privacy (readers will not that this argument applies exactly to Section 5C as well). The second is a vision of privacy as decisional autonomy – “what one eats is one’s personal affair, and it is part of privacy… [Section 5D] violates the right to be let alone.” At first glance, it might not seem that dietary choices hardly implicate those kinds of fundamental life decisions that are normally associated with individual autonomy. This is perhaps why it might be more helpful to think of this not in terms of how central dietary choice is to individual autonomy, but in terms of something that Jed Rubenfeld has called the “anti-totalitarian principle” (previously discussed here): State power ought not to be used for “forcing of lives into well-defined and highly confined institutional layers.” Control over diet is one instance of State shaping lives into a rigid pattern (often justified by ideological considerations).”

This twin conception of privacy as decisional autonomy and as the sanctity of the home was at the basis of Justice Singh’s opinion as well. It is interesting to note that, even as the Supreme Court reference continues to remain unanswered, the High Courts appear to be crafting the beginnings of a tradition of constitutional privacy.

Lastly, Justice Singh also held that the Notification violated Article 14. This was because the purpose of the New Excise Policy was primarily to combat alcohol addiction in poor and rural areas (hence, in the Policy (until it was overriden by the Act) the proposed ban was on country liquor, while leaving open sale of foreign liquor in urban areas). However, the Notification, while banning all alcohol (including foreign liquor) said nothing about banning toddy:

“Curious to note that toddy (Tari), which is available in abundance and tapped freely without any licence or permit and sold freely not only in the rural areas but urban areas and which has alcohol content, undisputedly matching or above beer, has not been prohibited. It is freely available even today. There is no notification barring it. Then to say, that on one hand toddy can be freely tapped and sold unchecked, but foreign liquor or IMFL including beer cannot be sold or consumed does not stand to reason, if the true object of the State was to implement Article 47 of the Constitution.” (paragraph 88.10).

This argument, however, is not entirely persuasive. Ever since Ram Prasad Seth vs State of UP, it is settled law that the State can choose to initiate reform in a phased, sector-wise manner, and Article 14 is not attracted if the State has taken the first of many progressive steps towards reform (what reform is, of course, is a matter for the State to determine, subject to the Constitution). Consequently, it was incumbent upon Justice Singh to refute the logic of phased or segmented reform (which, as he himself observed, was the point of the New Excise Policy). However, he did not do so.

Consequently, while I agree with Justice Singh’s holding that the Notification violated the fundamental right to privacy – and his formulation of the right to privacy – I disagree with him on the issue of Article 14. That said, the substantive part of the judgment, as a whole, is well-crafted, tightly reasoned, and rigorously argued (even though one may disagree with parts of it). It is a judgment that repays close study, and is a worthy addition to our constitutional tradition.

(In the next post, Justice Singh’s analysis of the punishment provisions will be discussed in a guest essay)

* This was pointed out to me in conversation by Jawahar Raja.


Filed under Decisional Autonomy, Directive Principles of State Policy, Privacy

The Bihar High Court’s Prohibition Judgment: Key Constitutional Issues – I: Excessive Delegation and Ultra Vires

In a landmark judgment handed down yesterday (Confederation of Indian Alcoholic Beverage Companies vs State of Bihar, Civil Writ No. 6675/2016), a division bench of the Patna High Court struck down the Bihar government’s notification imposing total prohibition in the state, as well as Section 19(4) of the Bihar Excise Act of 1915, which was the statutory provision under which the notification was passed. Justice Navaniti Prasad Singh wrote a detailed judgment, with which the Chief Justice concurred on all points apart from whether the right to drink alcohol was protected under Article 21 of the Constitution, as a facet of the right to privacy. The case itself threw up a host of fascinating constitutional questions, which merit close scrutiny. In Part I, I will examine the Justice Singh’s reasoning striking down the Act and Notification on grounds of excessive delegation, conflict with policy, and ultra vires. In Part II, I will analyse the disagreement between Justice Singh and the Chief Justice on the question of fundamental rights. Part III will be a guest post on the issue of punishment.

Section 19(4) of the Bihar Excise Act, as it originally stood, read as follows:

“Notwithstanding anything contained in the foregoing Sub-sections, the Local Government may, by notification, prohibit the possession of any person or class of persons, either in the Province of Bihar and Orissa or in any specified local area, of any intoxicant , either absolutely, or subject to such conditions as it may prescribe.”

On 31st March 2016, Section 19(4) was amended to read:

Notwithstanding anything contained in this Act and the Narcotic Drugs and Psychotropic Substances Act, 1985 (61 of 1985), the State Government may by notification, absolutely prohibit the manufacture, bottling, distribution, sale, possession or consumption by any manufactory, bottling plant, license holder or any person in the whole State of Bihar or in any specified local area in respect of all or any of the intoxicant s either totally or subject to such conditions as it may prescribe.”

The same day, the Government issued a notification banning distribution, sale, consumption etc. of country liquor. A few days later, it issued another notification [“the impugned notification”], which extended the ban to foreign liquor (and thus made prohibition absolute):

“Notification No. 11/Nai Utpad Niti-01.03/2016-1485, dated 5th April, 2016. – In exercise of the powers conferred under Section 19(4) of the Bihar Excise Act, 1915 (as amended by Bihar Excise (Amendment) Act, 2016), the State Government hereby imposes ban on wholesale or retail trade and consumption of foreign liquor by any license holder or any person in the whole of the State of Bihar with immediate effect.”

The petitioners’ first prong of attack was that the phrase “any person” under Section 19(4) of the Excise Act did not contemplate “all persons“, and that therefore, the Notification was ultra vires. To substantiate this, they relied upon judgments of the pre-Independence Federal Court, which had interpreted the pre-amended Section 19(4) in this manner. This contention was swiftly rejected by Justice Singh, since those judgments had expressly held that because the term “any person” was immediately qualified by “class of persons“, it was not intended to include “all persons” within its ambit. The 2016 amendment to Section 19(4), however, had removed the term “class of persons” altogether. Consequently, it could no longer be held that on a plain reading, “any person” could not include “all persons“. This aspect of the judgment is self-evidently correct, and needs no comment.

This brought Justice Singh to the second argument: that of excessive delegation. He relied extensively upon the classic judgment of the Supreme Court in Hamdard Dawakhana vs Union of India to draw a distinction between conditional legislation (which left the determination of the time, place, and manner of the legislation to become effective upon the delegated authority), and delegated legislation (in which the legislature provided broad policy guidelines, and allowed the delegated authority to fill in the gaps). The key test for the validity of delegated legislation – as observed by the Court in Hamdard Dawakhana – was that it should not amount to abdication of the legislative function. In Dawakhana, the Court had invalidated Section 3(d) of the Drugs and Magical Remedies Act, which used the term “or any other disease or condition which may be specified in rules made under this Act”, on the basis that there was no legislative guidance on how these “diseases” were to be selected. Armed with this, the Justice Singh was then able to find:

“The first thing to be noticed is that except for Section 19 (4) of the Act, there is no legislative enactment in the Act, which would indicate the circumstances and the conditions, whereunder total prohibition could be notified by the State Government as a delegatee. A reading of Section 19 (4) of the Act would show that it authorizes the State Government to issue notification absolutely prohibiting various activities, either in the whole of the State or in specified local areas, in respect of all or any of the intoxicants, either totally or subject to such conditions as may be prescribed. There is no legislative guideline at all discernible from the Act in relation to the aforesaid delegation.” (paragraph 85.11)

What particularly weighed with Justice Singh was that Section 19(4) belonged to the Bihar Excise Act of 1915, which was primarily about “permitting, facilitating and regulating manufacture, storage and sale of intoxicants and for collection of excise revenue” (paragraph 85.14), and not about imposing prohibition. Indeed, there was a separate Bihar prohibition act of 1937, which had never been used, indicating clearly that prohibition as a subject was meant to be dealt with by a separate law altogether. This allowed the Court to hold that Section 19(4), in its present form, amounted to impermissible delegation, and was therefore void – along with the notification issued under it.

Justice Singh then considered the interaction between Section 19(4), the notification, and the Bihar government’s “New Excise Policy” of 2015. The New Excise Policy, which had been notified on 21.12.2015, had introduced the goal of total prohibition in Bihar, and it was following this that Section 19(4) was amended. The Court’s analysis of the interaction between the three legal instruments is a little curious. After finding that the NEP’s purpose was to achieve prohibition in a phased manner starting with the prohibition only of country liquor, Justice Singh then held that consequently, the notification was ulta vires and arbitrary because it extended the prohibition to foreign liquor as well (paragraph 86.21). In doing so, he followed the judgments of the Supreme Court in State of Bihar vs Suprabhat Steels Ltd and Secretary, Ministry of Chemicals and Fertilisers vs Union of India, both of which had held that a government policy, once approved by the cabinet and notified, acts as a subsequent self-limitation of powers – i.e., it is not open to the government – on the pain of arbitrariness – to entirely disregard an existing policy that it had already committed itself to, via notification.

However, a close reading of both these judgments raises some doubts about their applicability to the present case. The Supreme Court’s judgment in State of Bihar vs Suprabhat Steels is laconic and confusing; however, a perusal of the High Court judgment (which was appealed by the State of Bihar to the SC) reveals that a key issue was never argued. Paragraph 14 of the High Court’s judgment states:

“In course of his submission the learned Advocate General submitted that the Notification of 4th April, 1994 amounted to a change of policy by the Government. This was objected to by the petitioners, and they challenged the State to produce material to show that the Government had consciously changed its policy by incorporating such a condition. However, ultimately the learned Advocate General did not proceed on the basis that the Government had changed its policy, and submitted that the Notification dated 4th April, 1994 was not inconsistent with the policy decision.”

In other words, the entire argument before the High Court and Supreme Court in Suprabhat Steels was conducted on the question of whether the impugned Notification was or was not consistent with prior policy. Similarly, in Cipla, the debate was about whether the placement of certain drugs in the Schedule to the Drug Pricing Order was consistent with the government’s own Drug Pricing policy.

In the present case, however, circumstances were different. After the New Excise Policy of 2015, which planned to introduce prohibition in a phased manner, starting with the banning of country liquor, the Bihar assembly amended Section 19(4) of the Excise Act that – as we saw above – allowed for complete prohibition, and drew no distinction between country and foreign liquor. What stronger evidence could there be for a change in policy? Justice Singh attempted to get around this by holding that the amended Section 19(4) was intended to give effect to the New Excise Policy, and that consequently, Notifications issued under it must also conform to the Policy. That argument, however, is simply belied by the text of 19(4), which is substantially broader than the Policy. And when the text is clear, is it open to the Court to go behind the language and to the intent of the framers, in order to derive an implied limitation upon its language?

However, Justice Singh then also held the notification to be void on the much more straightforward ground of ulta vires. Before we consider the argument, it is important to dispel a confusion that might arise on a reading of paragraphs 87 – 87.08. At various points, Justice Singh seems to switch between holding that the Notification is ultra vires, and that Section 19(4) itself is ultra vires. Particularly confusing is paragraph 87.03, where he observes:

“I may also refer to certain judgments, where Courts have held that a legislation has an object and if it is sought to be extended beyond the object, the legislation would be open to challenge. The first decision, I would refer to is the case of M.C.V.S. Arunachala Nadar and others Vs. State of Madras and others, (AIR 1959 Supreme Court 300), wherein in paragraph 5 of the reports, this is what their Lordships have noted, the relevant part whereof is quoted hereunder: “5. ……..In order to be reasonable, a restriction must have a rational relation to the object, which the legislature seeks to achieve and must not go in excess of that object…….””

As a statement of law, the first sentence is plainly incorrect, since the legislative object is to be derived from the text of the legislation itself. Invalidating one provision of an Act because it goes against the legislative object would be tantamount to invalidating a provision of the original Constitution because it violates the basic structure. Arunachala Nadar does nothing to support Justice Singh’s case, since it was on a different issue entirely – that of the reasonableness of restrictions under Article 19.

Eschewing Paragraph 87.03, however, the core of Justice Singh’s argument was that the Notification was ultra vires because it went beyond the legislative object. Going into the history of the Act, it noted that the purpose of the Act had always been to “for regulating the manufacturing, trade and business of intoxicating liquor and for augmenting State revenue”, and never for imposing prohibition. After surveying case law on the point, he then held that:

“Section 19 (4) of the Bihar Excise Act, 1915, could only be used for furtherance of the object of the said Act and not to abrogate the Act itself. In other words, the power, conferred on the executive under Section 19 (4), cannot be used by the executive to be a self destruct switch to kill the legislation itself.”

The problem with this argument, however, is that if the Notification was valid in terms of the amended Section 19(4), then it clearly could not be held to be ultra vires because, in a broader sense, it went beyond the legislative object. In other words, Justice Singh had to make an argument that despite its clear textual wording – allowing the government to prohibit any person from undertaking “manufacture, bottling, distribution, sale, possession or consumption” of intoxicating substances – the amended Section 19(4) – in light of the Act as a whole – was subject to an implied limitation that did not permit the imposition of absolute prohibition. This he did not do; and none of the cases that he cited endorsed the proposition that, in case of a seeming conflict between overall legislative object and a single section of the legislation, delegated action that conformed to the single section could nonetheless be invalidated on the ground of ultra vires. This is, of course, subject to the assumption that there could be a conflict between legislative provisions. I would submit, rather, that provisions like Section 19(4) are better understood as carving out exceptions to the overall legislative object (excise) for certain, specified substances. If the Excise Act was to be understood in this way, then the argument on ultra vires would no longer hold.

Consequently, and with respect, it is my view that while Justice Singh was correct on his reading of the amended Section 19(4), as well as his holding that the Section suffered from the vice of excessive delegation, he was incorrect in holding the Notification void for conflict with the New Excise Policy, as well as for holding it to be ultra vires. The outcome, of course, remains correct.

(In the next post, we shall discuss the Court’s approach towards fundamental rights.)


Filed under excessive delegation, Ultra Vires