Holy cow of the Constitution?


back to issue

Article 82, which provides for the setting up of the Delimitation Commission, is a thoughtful and crucial part of the Constitution. The sine qua non of Indian democracy is the autonomy and independence that the Election Commission of India (ECI) has demonstrated.

Delimitation of constituencies is the first step for holding elections because candidates cannot file nominations unless the constituencies reserved for SC (Scheduled Castes) and ST (Scheduled Tribes), and those left for general candidates, are identified and notified. This task is entrusted to the Delimitation Commission. The first delimitation in India was done in 1956, and subsequently in 1966 and 1976.

The importance of the Delimitation Commission can be better appreciated if we compare the independence of the ECI with the State Election Commissions (SECs). It was the intention of Parliament to make SECs as independent as the ECI. Hence, the wording of Article 243K, which created the SECs, was copied, word-to-word, from Article 324 under which the ECI was created. But the officials who prepared the draft bill obviously forgot to frame a provision equivalent to Article 82 or give necessary powers to the SECs to delimit the local body constituencies.

The Panchayat Raj and Municipal Bodies Acts of different state governments do not provide for the setting up of a Delimitation Commission, but vests the power of delimitation with state governments. The SEC cannot hold elections unless state governments identify and notify the SC, ST and general constituencies. More than one SEC had to file Writs of Mandamus against their state government for not notifying SC, ST and general constituencies in time. In contrast, the ECI does not have to depend on the Government of India for notifying constituencies for the Lok Sabha and state assemblies, thanks to the Delimitation Act under Article 82.


The second crucial aspect of the Delimitation Commission is that constituencies remain in force for 15 to 20 years. A third critical aspect is that Article 329 totally bars judicial review of the recommendations of the commission, a provision upheld by the Supreme Court. Another distinct feature of the Delimitation Commission is that unlike every other legislation wherein the detailed instructions for its implementation are laid down in the ‘Rules’ framed by the government under that act, the Delimitation Act empowers the commission to frame its own rules.

The composition of the present Delimitation Commission too provides a departure. In the past it consisted of a chairman, a member, and third person drawn from one of the Election Commissioners as an ex officio member. On this occasion, the SECs of all states and union territories were made ex officio members, the idea being that their knowledge of their respective state and vast experience in the sub-division, district, division, Board of Revenue, and in the secretariat would be made available to the commission.

The point at issue is whether the Delimitation Commission, appointed under the Delimitation Act 2002, has adequately understood the criticality of the delimitation process and is aware of the need for adopting both principles of natural justice and due process. This is particularly important since the government did not frame rules under the act and because judicial review is totally barred.


The Delimitation Act lays down broad parameters or principles for delimitation in sections 7, 8 and 9 of the act. In section 3 it also lays down who are to be the ex officio members and associate members and their respective roles.

It appears (to quote a mole in the commission) that the chairman and his advisors compared the composition of the current commission with the previous one and felt that the presence of SECs was an avoidable nuisance. Worse was the decision of the commission to treat SECs as members only for the purpose of discussing proposals specific to their own state.

The SEC’s ire at not being associated with the commission, as contemplated in the Delimitation Act, was made clear at the All India SECs Conference held in Bangalore on 25 June 2003. The ‘sense of the conference’ appears to have been communicated to the chairman and the EC member. Their failure to respond speaks volumes about the intention of the commission to downgrade the role of SECs. In addition, some SECs wrote individual letters objecting to the chairman’s style, suggesting a further meeting of SECs to work out a methodology. Though after concerted pressure from SECs, the commission circulated a draft of the methodology, but the SECs feel that this was a case of a fig leaf formality, because their suggestions were totally ignored.

A suggestion seems to have been made to the commission by some SECs that MLAs and MPs, who are associate members of the commission, should also be consulted before finalizing the methodology. It was pointed out that section 5 of the Delimitation Act clearly mandates that the commission ‘shall associate the associate members with itself for the purpose of assisting it in its duties in respect of each state though they do not have voting rights.’ Since the methodology, to a large extent, predetermines the delimitation process, any reasonable person would have expected the commission to consult associate members before finalizing the methodology, which the commission unfortunately failed to do. Some of the SECs seem to have cautioned that they would disassociate themselves with the working of the commission if the associate members of all states were not consulted in working out a methodology. Even this well meant caution fell on deaf ears.


Some public-spirited individuals and civil society organizations wrote to the chairman, offering to assist the commission in finalizing the methodology. Their reasoning was simple: the methodology has a critical bearing on the final product. Some SECs also made a similar suggestion, but without success. Even organizations such as the Reserve Bank of India invariably place its draft policy on any issue on their website for a month before issuing orders, despite the fact that it is not a requirement under the Banking Regulation Act or RBI Act.

Why do we claim that the methodology to a large extent predetermines the final product? The process of delimitation can begin from North to South or South to North, or from West to East or vice versa; it may be done clockwise or anti-clockwise or in a zigzag manner; it can be done in an alphabetical or numerical order. It can also proceed in the order followed by the previous commission.


The final product will differ considerably depending on the order adopted by the commission. This issue was brought to the notice of the commission by some SECs. The commission in its wisdom, however, chose not to receive inputs from SECs, associate members or civil society. It also deviated from the order adopted by the previous commission. The commission originally planned to start work North to Northwest and then proceed in a zigzag manner to end at the southern side. This was later changed by the commission which proceeded ‘from North to Northeast’, for reasons never known to anyone.

Section 9 stipulates that constituencies be delimited according to the 1991 Census population. The commission decided that there could be a variation of 10%. However, if a 20% variation were fixed, the shape of the constituencies would be different. But who decides whether it should be 20% or 10% or any other percentage? The 10% variance was obviously based on the precedence of the 1976 delimitation exercise. But the 1976 delimitation done keeping in mind changes in population between the 1961 and 1971 censuses, also had the benefit of increasing the number of seats to accommodate the changes in population.

So when we apply the 10% variance of the 1976 Delimitation Order to the present exercise, are we comparing apples with apples? We are now considering changes in population over three decades, and with the number of Lok Sabha seats being frozen for each state. It is, therefore, logical that we should have allowed for at least a 20% variance in population changes over three decades, and made some allowance for not increasing the number of seats. If we keep the variance at 20 or 25%, the number of changes would be considerably reduced. This begs the question: was the 10% variation adopted in order to destabilize the empires built by MLAs and MPs by nurturing their constituencies over many years (as voiced by a ‘Deep Throat’ in the commission) or to follow the precedent of the previous commission?


The Supreme Court had occasion to pronounce on the scope of plenary powers enjoyed by the Election Commission of India under Article 324. The three significant judgments on the subject are Mohinder Singh Gill vs. Chief Election Commissioner (AIR 1978 SC 851), A.C. Jose vs. Sivan Pillai (AIR 1984 SC 921), and Union of India vs. Association for Democratic Reforms (civil appeal no. 7178 of 2001 – judgment dated 2 May 2002). The substance of the above judgments is:

(a) In the absence of any parliamentary legislation or rule made under the said legislation, the commission is free to pass any orders in respect of the conduct of elections.

(b) When an act exists and rules are expressly made thereunder, the commission cannot override the act or the rules and pass orders contrary to the mandate of the act or rules. In other words, the powers of the commission are meant to supplement rather than supplant the law (both statute and rules) in the manner of superintendence, direction and control as provided by Article 324.

(c) Where the act or the rules are silent, the commission is empowered under Article 324 to give any direction in respect of the conduct of elections.

The Delimitation Commission is not a constitutional body like the ECI, enjoying powers similar to those contained in Article 324. Since it is only a statutory body, it must confine itself to the principles contained in the Delimitation Act 2002, as it has no plenary powers. The Delimitation Commission cannot, therefore, go beyond what is contained in section 9 or any other section of the Delimitation Act 2002.

On the face of it, limiting MLA and even MP constituencies to a district sounds reasonable. But it must be appreciated that state governments form new districts and new panchayat samitis from time to time. It would be naïve to think they do so without any gerrymandering. The decision about whether to restrict a constituency to a district or not will again affect the final product. Hence, adopting existing boundaries blindly without consulting the SECs and associate members who have sound knowledge of the ground realities in the states is not particularly wise.


There could be many reasons why the commission chose not to adequately associate SECs, and not consult with associate members and civil society in finalizing the methodology:

1. The general perception of SECs, associate members, and civil society organizations who interacted with the commission is that the chairman seems to have forgotten that he has retired from the Supreme Court!

2. Section 7(4) states: ‘If there are differences of opinion among the members the opinion of majority shall prevail.’ The operative word here is ‘members’. Since associate members have no voting rights, members mean the members mentioned in section 3, under the heading, ‘Composition of the Commission’. They are (i) chairman (ii) ex officio, EC member and (iii) SECs of the concerned state. But the legal wing of the commission (to quote a mole in the commission) seems to have propounded that commission meetings can be held with only the chairman and EC member since these two constitute a majority in a total of three members! The question of a majority only arises after convening the meeting and not for convening the meeting. It seems that a member in the form of SEC, in their view, need be counted as such only for the proposals relating to their respective states and not for the purpose of evolving procedure/methodology under section 7(1). But surely, for deciding on the methodology, the majority means the chairman, EC member and all SECs present.

3. There is a feeling in the commission (quoted by a Deep Throat in the commission) that since many of the sitting MLAs and MPs have cultivated their constituencies for many years and converted them into their pocket boroughs, it would be in public interest to destabilize them. This is clearly outside the purview of the mandate given to the commission by the Delimitation Act or Article 82.

Last, there exists a reckless confidence that since Article 329 bars judicial review, the commission can do whatever it wishes! And this is precisely what it seems to have done in deciding on the methodology and draft delimitation proposals.


It is interesting to note that the mighty Delimitation Commission that plays such a critical role in our democratic process is nowhere mentioned in the Constitution. It is appointed through an ad hoc legislation under Article 82, read with Article 327.

Article 329(a) bars judicial review in respect of delimitation of constituencies, and Article 329(b) similarly bars judicial review of election to either House of Parliament or state legislature. The preparation of electoral rolls has a direct bearing on the result of elections. If we read Article 329 with Article 327 and 328, electoral rolls should also be outside judicial review. But the High Courts freely admit petitions relating to electoral rolls and directly order inclusion of the petitioner in the rolls.

It is really the Supreme Court judgment in AIR 1967 SC P669 (1) in Meghraj Kothari vs. Delimitation Commission and Others, which gave birth to an unconditional protection to the decisions of the commission from judicial review.

A number of acts passed by Parliament and state legislatures have been included under the 9th Schedule to exclude them from judicial review. Literally, every one of them was questioned in the Supreme Court and the court either upheld or struck them down, partly or in full. But the Supreme Court chose to impose self-restriction on itself by this judgment only with regard to the Delimitation Commission.


How faulty or beautiful is the baby that the commission gave birth to is a fitting subject for several doctoral theses. Let us examine some features of this exercise by taking the example of Andhra Pradesh.

1. Two contiguous parliamentary constituencies, Chittoor and Tirupathi in the same district, were reserved for SCs, notwithstanding a clear mandate in section 9(1)(C) of the act that reserved constituencies should be distributed in different parts of the state. Remember that the state has three distinct regions, with a demand for a separate state of Telangana (since 1969) now raging at its peak, about which the commission was made aware before its final order. But the commission chose to reserve two constituencies in the coastal region with 10 districts, two in Telangana region with nine districts and two in Rayalaseema with four districts.

2. The names of a number of MLA and some MP constituencies were changed for no apparent reason.

3. Despite the number of constituencies being frozen by the act, 44 existing assembly constituencies were deleted and 64 new ones created. No valid reasons were given for the deletion and creation.

4. There is not a single MP/MLA constituency that did not undergo change!

5. Out of a total of 19 assembly constituencies reserved for STs, nine are newly created and only 10 retained with changed jurisdiction.

6. Out of a total of 48 assembly constituencies reserved for SCs, 32 are new and only 16 continue, but with changed jurisdiction.

7. A study of 18 out of 36 Lok Sabha constituencies shows that except Adilabad (which remains unchanged in terms of assembly constituencies included in it), the remaining 17 have undergone many changes.

* In Peddapalli constituency, only 3 out of the 7 assembly segments are retained and the remaining 4 are new constituencies.

* In Karimnagar, 3 are old and 4 are new; In Nizamabad 3 out of 7 are new.

* Zaheerabad is a new Lok Sabha constituency, with 3 assembly segments from the old Medak constituency, 1 from Nizamabad constituency and 3 are new constituencies.

* In Medak only 2 assembly constituencies are old.

* Malkajgiri is a newly constituted Lok Sabha constituency taking 3 assembly segments from the abolished Siddipeta constituency of Medak district. The remaining 4 are new; but this is justified because of explosive growth of population in the suburban areas of Hyderabad city.

* In Secundrabad 4 are old while 3 are new, which again is justified due to the same reason as in Malkajgiri.

* In Hyderabad 4 are old constituencies and 3 new, again justified due to the same reason.

* In Mahabubnagar 2 are old, 1 is taken from Nalgonda constituency and 4 are new due to migration in Hyderabad suburbs.

* In Nagar Kurnool 5 are old, 2 are taken from Mahabubnagar with little justification.

* In Nalgonda 3 are old 2 are from Mehbubnagar and 2 are new, again with doubtful justification.

* Boongir is a newly formed Lok Sabha constituency, taking 4 from Nalgonda, 1 from the abolished Miryalaguda constituency and 2 new constituencies. This seems to be an avoidable juggling.

* Warangal: 2 constituencies were taken from old Hanamaknonda constituency, which is now abolished, 2 from the old Warangal constituency and 3 are new constituencies. This sea change is of doubtful validity.

* Mahaboobabad is a newly created constituency, taking 4 from the Warangal constituency and 1 from Khammam constituency, 1 from Bhadrachalam constituency and 1 newly created constituency, with little justification.

* Khammam: 5 are old constituencies and 2 are new due to the creation of Mahaboobabad constituency.

* Araku: This again is a newly created constituency, taking 3 from Parvathipuram and 1 from Bobbilli and 3 new constituencies, for no good reason.

Name of District

% of Average Variation

% Range of Variation



50 to 90



25 to 100



25 to 100



75 to 100



60 to 100



25 to 100

(The average variation is misleading because the denominator, which is the number of constituencies in the district, happens to be 15. But actually 5 out of 15 constituencies are new, with 100% variation.)



50 to 100



40 to 100



25 to 100



50 to 100



25 to 100



25 to 100



40 to 100

E. Godavari


40 to 100

W. Godavari


25 to 100



40 to 100



25 to 100



25 to 100



40 to 100



40 to 100



25 to 100



50 to 100



25 to 100


Thus, 5 out of 18 are new constituencies, with good justification for four constituencies, due to urbanization of Hyderabad city. The remaining 13 constituencies have seen far too many changes because of the zigzag criteria and the draft proposals made by the office of the commission, without prior consultation with SECs and associate members.

8. There is not a single assembly constituency that has not undergone considerable change. It’s a question of degree.

(See accompanying Table for district wise changes.)

Since a group of seven assembly constituencies forms a Lok Sabha constituency, the variation in Lok Sabha constituencies is also duly reflected in assembly segments.

I have no doubt whatsoever that similar comprehensive changes would have taken place in the composition of Lok Sabha and state assembly constituencies in other states also as the same methodology was applied to all states.


The voters, candidates, political parties and the nation must learn to cohabit with this arrangement for the next 20 years, since that is the life of the delimitation prefixed under the Delimitation Act 2002. Even assuming that the final product is acceptable, the commission’s gross failure to follow due process is unjustifiable; otherwise, surely it will be a case of the ends justifying the means. Seen in another way, would a scrupulous adherence to due process have in any way adversely affected the quality of the end product? In the view of this author, absolutely not.

If there was a judicial review, would the High Courts or Supreme Court uphold the action of the commission in not following the due process? Not a day passes without the courts granting an interim stay or absolute stay, and subsequently setting aside executive orders only on the ground that due process was not followed, even when the order of the executive subserves public good. Even the slightest deviation from due process is not condoned; the only relief they may give is to permit re-examination.

The entire rationale for the Supreme Court’s decision to bar judicial review is to ensure that elections are held on time and not embroiled in litigation. Can we not find a way to hold elections in time even while following due process?

After reviewing the performance of the present commission, a national debate on the issue of judicial review is certainly called for. And where else but in the pages of Seminar?