A Promise of Speed, a Record of Delay

When Parliament enacted the Tenth Schedule in 1985, it made a deliberate choice about who would police defections. The question whether a member had incurred disqualification would not go to a court, and it would not go to the President or the Governor acting on the opinion of the Election Commission, which is the route the Constitution prescribes for every other disqualification. It would go to the Chairman or the Speaker of the House, an insider, on the reasoning recalled whenever the choice is defended: the House is master of its own composition, the presiding officer knows its parties and its whips, and an insider can decide fast. The minister who piloted the Constitution (Fifty-second Amendment) Bill told the Lok Sabha that if defection was to be outlawed effectively, a forum had to be chosen which would decide the matter fearlessly and expeditiously, a formulation the Chairman of the Rajya Sabha recalled in 2019 while disposing of petitions with a speed he held up as the standard. Speed was not an incidental hope attached to the design. Speed was the design’s justification.

The record of the design in operation is a record of variance so wide that the variance itself is the finding. At one end of the record, a Speaker of the Himachal Pradesh Assembly decided six petitions in February 2024 within two days of the conduct complained of, in a thirty page order, and six members ceased to sit with immediate effect. At the other end, petitions filed in the Manipur Assembly in 2017 were still undecided three years later, when the Supreme Court recorded that the member complained against had in the meantime sat in the Council of Ministers of the government his crossing had helped to form, and the Court directed a decision within four weeks. Between those poles lie the Maharashtra petitions of 2022, decided in January 2024 after the government they concerned had fallen and been replaced, the Telangana petitions of 2024, which drew their first notices in January 2025 only after the petitioners reached the Supreme Court, and a long tail of petitions in state assemblies that were never decided at all before the Houses that received them were dissolved. The grounds of disqualification, the texts of the whips, the arithmetic of mergers, all of which matter intensely on paper, decided none of these outcomes by themselves. Who held the chair, and how long the chair took, decided them.

That is the test this study applies to the practice of the Tenth Schedule. A reader who works through the record should be able to explain why the outcome of a defection episode is determined less by the grounds pleaded than by who decides and when, and why a provision that disqualifies the defector can leave the defection itself entirely effective: governments formed, budgets passed, ministries staffed, terms served out, all on the strength of votes whose casters a tribunal later holds, or would have held, to have been disqualified from the day they crossed. The law of the Schedule, its grounds, its exemptions and the critique of its constitutional design, is treated in the companion study of the anti defection law explained ground by ground, whose record this study follows exactly: the Fifty-second Amendment received assent on 15 February 1985 and came into force on 1 March 1985, and the leading authority remains Kihoto Hollohan v. Zachillhu, reported at 1992 Supp (2) SCC 651. What follows is the other half of the subject, the Schedule as it is actually administered: the filing of petitions, the pace of the chair, the conduct of parties and members during the long interval in which nothing has been decided, the interventions of the courts, and the remedies that arrive after the political work of a defection is complete.

A word about the record itself, before it is used. The practice of the Tenth Schedule is documented unevenly, and the unevenness is part of the evidence. What survives in citable form are the proceedings that reached courts: reported judgments of the Supreme Court and the High Courts, the orders of presiding officers that were carried in challenge, and the assembly episodes large enough to occupy the national press. Petitions filed in smaller assemblies, against members of smaller parties, in configurations that threatened no government, leave traces mainly in the bulletins of their Houses and in occasional reporting, and the great majority are never the subject of a reported order at all. A study of practice therefore works from the visible stratum: the episodes in which something was at stake, which are also the episodes in which the system’s incentive structure was under load. Conclusions are drawn from named records, attributed as they arise, and where a pattern is asserted it is asserted as a pattern across those records and not as a count of all petitions, because no public source counts all petitions. The absence of a census is itself diagnostic. A legal regime whose proceedings generated a reliable public tally of filings and disposals would be a regime in which delay was at least measured, and the first finding of the practice is that in most Houses it is not even that.

The Machinery the Tenth Schedule Actually Builds

Paragraph 6 of the Tenth Schedule is the whole of the adjudicatory machine, and it is worth pausing on how little machinery it is. If any question arises whether a member of a House has become subject to disqualification under the Schedule, the question shall be referred for the decision of the Chairman or, as the case may be, the Speaker of the House, and the decision shall be final. Where the question concerns the Chairman or the Speaker himself, it is referred to a member elected by the House for the purpose. Paragraph 6(2) adds that proceedings in relation to such a question shall be deemed to be proceedings in the House within the meaning of Articles 122 and 212, the provisions that shield internal proceedings of Parliament and the state legislatures from inquiry by courts on the ground of irregularity. Paragraph 8 completes the scheme by empowering each House to make rules for giving effect to the Schedule, rules that must be laid before the House and that may provide for the committees through which the presiding officer works. Nothing in the Schedule fixes a period within which the question must be decided. Nothing prescribes an appeal. Nothing says what happens to the member, the member’s vote, or the government the member sustains while the question awaits decision. The machine has an entrance, a decision-maker and an exit marked final. It has no timetable, no interim state and no enforcement arm of its own.

The rules made under paragraph 8 supply the steps, and their character repays attention, because they show a procedure carefully specified at every point except the one that governs outcomes. Under the Members of Lok Sabha (Disqualification on Ground of Defection) Rules, 1985, which the corresponding state rules follow in substance, a proceeding begins with a petition in the prescribed form, ordinarily moved by a member of the House and supported by the material relied on; the defection petition is one of several election-connected proceedings whose routes are mapped in the wider guide to election law in India. If the petition complies with the requirements of the rules, the Speaker causes copies of the petition and its annexures to be forwarded to the member complained against and, where that member belongs to a legislature party and the petition has not been made by its leader, to the leader as well. The member or the leader must forward comments in writing within seven days of receipt, or within such further period as the Speaker may for sufficient cause allow. After considering the comments, the Speaker may either proceed to determine the question or refer the petition to the Committee of Privileges for a preliminary inquiry and a report, in which case the determination follows as soon as may be after the report is received. The procedure for the inquiry follows, so far as may be, the procedure for questions of breach of privilege, and no finding of disqualification may be recorded without affording the member a reasonable opportunity to represent the case and to be heard in person. Every step has its form. The only temporal words in the scheme are the seven days for comments and the phrases as soon as may be, which govern the chair’s convenience and bind nothing.

Paragraph 8 holds one more possibility that the practice has never taken up, and its neglect is evidence of where responsibility for the missing clock lies. The rules the paragraph empowers are the House’s own rules, laid before the House and open to modification or disapproval by it, and nothing in the paragraph confines them to the forms of petitions and the mechanics of hearings. A House that wished its chair to decide within a stated period could write the period into its defection rules without any constitutional amendment, and a House that wished the petition transferred on default could attempt that too, subject to the argument that the Schedule’s allocation of the question to the chair is exhaustive. In the decades since 1985, so far as the reported record shows, no House has used the power to fix a period for decision. The omission cannot be charged to the Schedule’s text, which left the door open, nor to the courts, which have no warrant to write House rules. It belongs to the Houses themselves, whose majorities are, at any given moment, the actual or prospective beneficiaries of their own chair’s discretion over time. The missing clock is missing because the only hands positioned to install it belong to the side of the room that the clock would inconvenience.

A petition is, in consequence, a documentary enterprise before it is anything else. It must assemble the proof of the original party and the ticket on which the member was elected, the text and the service of any whip relied on, the record of the vote or the conduct from which a voluntary giving up of membership is to be inferred, and the notices and replies exchanged along the way, and it must keep that file alive through hearings spread over months or years, changes of counsel, changes of Speaker and, in the hardest cases, a change of government. Parties that litigate defections maintain such files the way litigants in long civil suits maintain theirs, and the reader-facing services built for organising documentary payloads of petitions, whips, notices and orders in a single protected file answer a need that this procedure itself creates, of the kind described in accounts of how journalists use Vaultbook to protect sources. The file matters because the proceeding is unforgiving about paper: a whip that cannot be proved to have been issued by the authority competent to issue it, or to have been communicated to the member, will not carry a paragraph 2(1)(b) case however plain the defiance, and a course of conduct from which paragraph 2(1)(a) is to be inferred must be documented act by act. Practice therefore rewards the party that papers its defections and its accusations with equal care.

The communication of the whip is the most litigated sheet in that file. Parties issue whips through letters delivered by hand, through messages on the telephones of members, through publication, and through notices affixed or read in the legislature party’s meetings, and members resist disqualification with denials of receipt that the petitioner must then displace with proof. In the Maharashtra proceedings the factions contested the authority by which the rival whips of June and July 2022 had been issued and the manner of their communication to members who were, at the material hours, away from Mumbai in the custody of their own camp’s arrangements; the Constitution Bench’s holding that the whip is the political party’s to appoint, and that the chair must test recognition against the party’s constitution, settled the question of authority for the future without shortening by a day the inquiries such contests generate in a live file. A party that cannot prove its whip’s journey from the competent office to the member’s hands has, for practical purposes, no whip case, however public the defiance was; and a member determined to create the doubt needs only to be unreachable at the right hours, a condition that the logistics of modern defection episodes, with their chartered movements and guarded resorts, are well adapted to supply. The doctrine of the whip ground belongs to the companion study. Its practice is an evidence problem, and evidence problems are the natural habitat of adjournments.

What the presiding officer must do, step by step

The chair must forward the petition to the member and, where apt, to the legislature party leader; receive written comments within seven days, extendable for sufficient cause; hear the member in person before any adverse finding; and record a reasoned order. The rules fix no deadline for that order.

The chair’s mandatory path runs: receive a petition in the form the House rules prescribe; test it against those rules; forward copies to the member and, where apt, to the legislature party leader; allow seven days for written comments, extendable for sufficient cause; hear the member in person before any adverse finding; decide, possibly after a preliminary inquiry by the Committee of Privileges; and record a reasoned order. The steps are duties. Their pace is not.

Two features of this machinery govern everything that follows in the practice. The first is that the Schedule creates no interim disability. A member against whom a petition is pending remains a member in full: entitled to sit, to speak, to vote, to hold ministerial office if appointed, and to be counted in every division, until the moment a decision records disqualification. The second is that disqualification, when it comes, is generally treated as relating back to the conduct that incurred it, so that the member is held to have been disqualified from the date of the defection rather than from the date of the order. The two features fit together in a way that defines the practical subject. Between conduct and decision, the member’s political acts are fully effective. After decision, the member is deemed never to have been entitled, yet the acts done in the interval, the votes cast, the divisions carried, the governments formed and sustained, stand untouched, because the Constitution separately provides that proceedings of a House are valid notwithstanding that it is afterwards discovered that a person not entitled to take part sat or voted. The sanction is real and it is personal. The politics it was meant to undo are irreversible.

The Missing Clock

The absence of a decision period in the Tenth Schedule is not an oversight that practice has had to work around. It is the load-bearing fact of the system, and it can be located precisely: paragraph 6 commits the question to the chair and declares the decision final, paragraph 8 leaves procedure to House rules, and neither the Schedule nor the rules made under it attach a consequence to the passage of time. The point is sharpened by contrast with the neighbouring provisions the framers of the Fifty-second Amendment had before them. The Constitution fixes periods where it means to. A Money Bill’s fate in the Legislative Council is bounded by fourteen days under Article 198. An ordinance lapses at the expiration of six weeks from the reassembly of the legislature under Articles 123 and 213. The Representation of the People Act, 1951 directs, in Section 86(7), that every election petition be tried as expeditiously as possible and that endeavour be made to conclude the trial within six months of presentation. The Tenth Schedule, enacted by a Parliament legislating in full knowledge of this vocabulary of deadlines, contains no comparable words, and the House rules under paragraph 8, which could have supplied a period administratively, instead repeat the open texture of the parent text: comments in seven days, a decision as soon as may be.

The claim this study presses, because the record sustains it at every point, can be named in a sentence: the remedy outlives the House. A defection petition is a proceeding whose practical value depends entirely on its being decided while the configuration it challenges still exists, since its only sanction is the loss of a seat in a House whose remaining life is the outer limit of the remedy’s usefulness. Delay does not merely postpone the remedy; it consumes it, month by month, until what is left to decide is the historical question whether a member who has served out the term, or the government that member sustained, was entitled in law to do what it did in fact. Speakers deciding at leisure are not breaching a deadline, because there is none to breach. Petitioners complaining of delay are not pointing to a violated rule; they are pointing to a missing one. And courts asked to intervene face the same absence from the other side: until the Supreme Court constructed a standard of reasonableness out of the adjudicatory character of the office itself, there was no text against which the pace of a chair could be measured at all.

It follows that the interesting questions about the Tenth Schedule in practice are not questions about its text. They are questions about the incentives of the office that administers it, about what governments and opposition parties do with the interval the missing clock creates, about the points at which courts have been willing to convert an absence of deadline into a duty of promptness, and about reforms that would put a clock, or a different decider, or both, into a scheme that has neither. The sections that follow take those questions in that order.

The seven days for comments deserve a final look, because they disclose the framers’ own sense of the natural timescale of such a proceeding. A member complained against is expected to answer the petition within a week, and the rules treat an extension of that week as a concession to be justified by sufficient cause. The scheme thus contemplates a respondent who replies in days, evidence that is documentary and close at hand, and a tribunal already seized of the House’s own records. Nothing in the nature of the inquiry explains a disposal measured in years: the facts to be found are a closed set, consisting of party membership, the issue and communication of any whip, and specified conduct on specified dates, all of it provable from documents and the record of the House. Delay under the Schedule is therefore not a by-product of complexity, as delay in a land acquisition reference or a tax appeal may be. It is a surplus the system generates over and above everything the inquiry requires, which is why the search for its causes leads away from the proceeding’s demands and towards the proceeding’s politics.

The Presiding Officer in Operation

The presiding officer who decides a defection petition is, by the time the petition arrives, a person with a political past and, frequently, a political future. The Speaker of a state assembly is elected by the majority of the House from among its members, which in practice means chosen by the ruling party or coalition, and the office carries no convention of severance from the party on whose ticket its holder was elected. The holder of the chair remains a member of the political party, is frequently a senior figure within it, and returns to ordinary politics when the tenure ends, sometimes to ministerial office. The Constitution (Fifty-second Amendment) Act thus entrusted the disqualification of defectors to an adjudicator who is structurally interested in the arithmetic the petition seeks to change, and it did so while proclaiming, through the piloting minister, an expectation that the forum would act fearlessly and expeditiously. Kihoto Hollohan sustained the arrangement, with the majority reasoning that the Speaker’s office, its traditions and the expectation of impartiality attached to it justified the trust, while recording that the presiding officer deciding such a question acts as a tribunal and is bound by the obligations that attach to tribunals. The minority in Kihoto Hollohan would have struck the Schedule down, and one strand of its reasoning was precisely the structural one: that the choice of an insider as adjudicator was itself the infirmity. The practice since 1992 is, in large part, the testing of the majority’s trust against the minority’s warning.

Why the chair keeps the party card

No rule requires an Indian Speaker to resign from a political party on election to the chair. Filled by the ruling side’s majority, held with its confidence and vacated into its ranks, the office expects impartiality as a discipline of its partisan holder, not as a property of the office.

An Indian Speaker, unlike the Speaker of the House of Commons at Westminster, does not resign membership of a political party on election to the chair, and is not expected to stand apart from the party’s counsels for the rest of a career. The chair is won with the ruling side’s votes, kept with its confidence and surrendered with its fall, and nothing in the Constitution or in the rules of any House requires the holder to demonstrate independence by any formal act of separation. The impartiality the system expects is therefore a discipline of conduct laid on a partisan office, not a feature of the office’s design.

The consequences for defection adjudication follow from ordinary incentives rather than from any assumption about individual virtue. A petition against members whose votes sustain the government of the day asks the chair to reduce the majority that elected the chair; a petition against members of the opposition asks the chair to weaken the government’s opponents. In the first case, expedition injures the chair’s own side; in the second, expedition serves it. The record bears the asymmetry out in both directions. Petitions have been decided with striking speed where the members complained against belonged to the ruling party and had crossed against it: the six Himachal Pradesh members in February 2024 were members of the ruling Congress who had voted for the opposition candidate in a Rajya Sabha election and defied the whip on the budget, and their petitions, moved by the parliamentary affairs minister of their own government, were decided against them within days. Petitions have equally been decided with striking slowness where the members complained against had crossed into the ruling side: the Manipur petitions of 2017 lay undecided for about three years while their subject sat as a minister, and the Telangana petitions of 2024 drew notices only when the matter reached the Supreme Court. Attribution discipline matters here: these are patterns visible across the decided and reported record, illustrated by named cases, not a statistical claim about all petitions, most of which are never reported at all. But the pattern is the one the Supreme Court itself named in Keisham Meghachandra Singh, when it observed that the Speaker, a member of a political party, would in most cases fail to act impartially where the defectors sustain the government, and it is the pattern the Court repeated with evident weariness in the Telangana judgment of July 2025, recording that Parliament had trusted the high office of the Speaker to act expeditiously and that the trust, in many cases, had not been honoured.

One further feature of the office compounds the problem of the interested chair: the chair can be placed under a disability of the opposite kind at the moment adjudication matters most. In Nabam Rebia v. Deputy Speaker, decided on 13 July 2016 by a Constitution Bench, the Supreme Court held that it is constitutionally impermissible for a Speaker to adjudicate disqualification petitions after a notice of intention to move a resolution for the Speaker’s own removal has been received, since the Speaker would otherwise be positioned to alter the composition of the very House that is to vote on the removal. The holding grew out of the Arunachal Pradesh crisis of 2015 and 2016, in which the Governor had advanced the Assembly session and directed the order of its business, the Assembly had purported to remove the Speaker at a sitting held away from the Assembly premises, and the Court ultimately restored the position as it stood before the Governor’s intervention. Whatever its merits as a protection against a chair disqualifying its way out of a removal vote, the Nabam Rebia proposition hands a tactical instrument to any group facing disqualification: a notice for the removal of the presiding officer, moved before petitions are decided, is said to freeze the adjudication. The proposition had precisely that bite in Maharashtra, where disqualification notices issued by the Deputy Speaker in June 2022 crossed with a pending notice for the Deputy Speaker’s own removal moved by the breakaway group. In Subhash Desai the Constitution Bench doubted the correctness of the proposition, noting provisions it considered the earlier bench had not examined, and referred the question to a bench of seven judges. That reference, made in May 2023, has not been decided, so the law on the point stands in a state of suspended doubt: a holding of a Constitution Bench, formally operative, whose foundations a later Constitution Bench has questioned and left for a larger bench that has yet to speak.

Alongside the interested chair and the disabled chair stands a third figure in the practice, the successor chair. Assemblies elect new Speakers when governments change, and a change of government mid-term, which is what most large defection episodes produce, routinely means that the petitions arising out of the episode are decided, if they are decided, by a presiding officer elected by the beneficiaries of the defection. The Maharashtra sequence is the cleanest illustration. The petitions of June and July 2022 were filed before one presiding arrangement, a new Speaker was elected on 3 July 2022 by the House as reconstituted under the new government, and it was that Speaker who, eighteen months later, decided the petitions against both factions and disqualified nobody. No rule was broken in that sequence; the Schedule assigns the question to the Speaker of the House, and the House is entitled to change its Speaker. The point is structural, and it is the same point the One Test directs attention to: the identity of the decider is a variable in the system, and the politics of the episode determines the variable.

A fourth figure completes the set: the acting chair of a vacant office. The Maharashtra Speakership stood vacant from February 2021, when the incumbent resigned, until July 2022, and through the opening of the defection episode the functions of the office, including the receipt of petitions and the issue of notices, were discharged by the Deputy Speaker. The Schedule and the rules make no distinction of consequence for this purpose: the question is referred to the Speaker of the House, and whoever lawfully discharges the office holds the question with the same finality and the same absence of a timetable. A vacancy therefore changes nothing about the adjudication except the person, which is the variable this study has identified as decisive. The Maharashtra petitions were noticed by a Deputy Speaker labouring under a removal notice of his own, inherited by a Speaker elected by the defection’s beneficiaries, and decided after a Constitution Bench and two rounds of Supreme Court directions on timing: a complete tour of the office’s possible configurations within a single proceeding, at the end of which the configuration that decided was the one the episode itself had produced.

What Happens While a Petition Sleeps

The interval between a defection and its adjudication is not a pause in the political process. It is the political process, conducted under a suspended legal question, and the parties to a defection episode understand its rules precisely. The member complained against keeps the seat and the vote; the government the member sustains keeps the majority those votes compose; ministries can be and are formed with the participation of members under petition, as the Manipur record shows, where the member crossed within days of the notification of election results in March 2017 and held ministerial office while the petitions against him slept; and every division held in the interval is carried by the arithmetic of the House as it stands, not as a tribunal may later declare it ought to have stood. A government facing a confidence motion is entitled, as a matter of the House’s own procedure, to the votes of members under petition, and no rule of the Schedule or of any House permits those votes to be set aside, discounted or held in escrow pending the chair’s decision. The floor test, which is the Constitution’s own instrument for resolving doubt about a government’s majority, therefore runs on numbers that a pending petition calls in question but does not touch.

Opposition parties conduct themselves accordingly. The petition is filed, and it is a real instrument: it creates a public record of accusation, it forces the member to answer in comments and at hearings, and it preserves the legal question for the day the chair decides or a court intervenes. But no competent opposition treats the petition as the remedy for the defection itself, because the tally that matters, the tally on the floor, is unaffected by it. The responses that actually contest a defection are political: counter-mobilisation of members, approaches to the Governor, litigation aimed not at the merits but at the clock, and the organisation of the by-election campaigns that follow if seats do fall vacant. Petitioners who go to court during the pendency of a petition are, in case after case, asking for one thing only: a direction that the chair decide by a date. The merits are the chair’s business; the calendar becomes the courts’ business, and the struggle over the calendar is the real litigation of the Tenth Schedule.

Governments, for their part, manage the interval as a parliamentary resource. A ministry whose majority includes members under petition has every reason to order its legislative business so that the interval is consumed by ordinary sittings rather than by confidence tests it does not need, to distribute office in a way that binds the petitioned members to the arrangement their continuation sustains, and to treat the chair’s calendar as a component of its own survival plan, since each adjourned hearing is, in effect, a division won without a vote. The interval also disciplines the members under petition themselves: a member who has crossed once and whose seat awaits a chair’s pleasure is poorly placed to cross back, because the pending petition, which the receiving side’s chair controls the pace of, can be expedited as readily as it can be slept on. Pendency is thus not merely a shield for the defection that has happened. It is an instrument of governance over the defectors, held by the same hands that hold the decision, and the possibility that an undecided petition functions as a leash explains a feature of the record that pure delay does not: petitions against members of the governing side’s new allies are not always decided slowly. They are, in the episodes on record, decided at the speed the governing configuration requires, which for a leash is not a speed at all but a standing capability.

The burden of the interval falls asymmetrically, and the asymmetry explains a good deal of the episode-level behaviour the record shows. For the defector and the receiving party, time is an asset: every month of pendency consolidates the new arrangement, distributes its offices, passes its budgets and brings the end of the term, at which the petition lapses, closer. For the petitioner and the abandoned party, time is a cost paid in a currency that cannot be recovered, since even a complete victory at the end yields only the vacancy of a seat whose occupant has already done the work the defection was organised to do. This is why the practice shows petitioners pressing for expedition and respondents perfecting the arts of the adjournment, the extension of time for comments, the challenge to the petition’s form and the objection to the chair’s jurisdiction, each unimpeachable in itself and cumulatively decisive. It is also why episodes cluster as they do: a defection timed early in a term maximises the interval available to be consumed, and a defection timed near a term’s end may need no adjudication strategy at all, since the House will dissolve before any tribunal’s order can matter.

Two thirds of the legislature party, counted in the room

Paragraph 4 protects a member when the original party merges with another and not less than two thirds of the legislature party agree to the merger. The chair receives the claim, counts the legislature party as it stands, and accepts or rejects the fraction; the Goa crossing of July 2019 turned on exactly that count, ten of fifteen.

The merger exemption of paragraph 4 is where counting, rather than time, does the work. The exemption protects a member where the original political party merges with another party and the member joins the merged party, and the Schedule deems a merger of the original party to have taken place where not less than two thirds of the members of the legislature party concerned agree to it. In practice the provision is operated as a headcount of the legislature party at the moment of crossing, certified to the chair, and the chair’s acceptance of the arithmetic ends the legal question in most cases. The Goa episode of July 2019 is the standard illustration: ten of the fifteen Congress members of the Assembly crossed to the Bharatiya Janata Party claiming the shelter of paragraph 4, ten being exactly two thirds of fifteen, the Speaker accepted the claim, and challenges to that acceptance occupied the courts for years without displacing a single member. The exemption converts a question of loyalty into a question of arithmetic, and arithmetic, unlike loyalty, can be assembled in advance.

The merger route also displays, in a quieter register, the same dependence on the chair that governs the rest of the practice. The Schedule does not itself count the two thirds; the presiding officer receives the claim, examines whether the statutory fraction of the legislature party has agreed, and accepts or rejects it, and the acceptance is a decision of the same tribunal whose finality paragraph 6 proclaims and whose slowness the rest of this study records. Where the fraction is comfortably above the line and the receiving party holds the chair’s confidence, acceptance in practice follows the claim closely. Where the fraction is contested, as where rival factions each assert that they command the party, the merger question fuses with the symbol and leadership questions being fought before the Election Commission, and the chair’s treatment of it acquires the same temporal character as any petition. A defection organised as a merger is, in the taxonomy of the practice, a defection that arrives with its defence pre-decided in its favour, provided the numbers have been gathered before the announcement. The lesson organisers draw from the record is not that the law is weak but that it is specific: it punishes the lone crosser and the small group, and it prices the large group out of its reach.

The litigation tail of the Goa episode completes its lesson. The disqualification petitions filed against the ten members were dismissed by the Speaker in 2021, and the Bombay High Court at Goa upheld that dismissal in February 2022, in the same month in which the State voted for a new Assembly and the term to which the petitions belonged expired. Two and a half years of proceedings thus produced a final judicial word at the moment the question ceased to be capable of affecting anything but the record. The members concerned had served in the government they crossed to, several of them as ministers, throughout the interval. Merger cases are often described as the Schedule’s loophole; the practice suggests the description is incomplete in an instructive way. The loophole is not only that paragraph 4 shelters the sufficiently numerous. It is that every route of challenge to the shelter, petition, chair and court, runs on the same calendar that governs the rest of the Schedule, so that even a meritorious challenge must win twice, once on the law and once against the clock.

Arunachal Pradesh in 2016 supplies the merger logic at its largest scale, at the level of a whole government. After the constitutional crisis that produced Nabam Rebia, the restored Congress government of Nabam Tuki gave way to a government led by Pema Khandu, and in September 2016 forty three of the forty four Congress members of the Assembly, the Chief Minister among them, joined the People’s Party of Arunachal Pradesh, leaving the party of their election with a single member; before the year was out, the Chief Minister led most of the same members into the Bharatiya Janata Party. No disqualifications followed, and none were seriously expected: a crossing organised at the scale of an entire legislature party presents the chair and the Commission with a change of party identity rather than a defection, and the Schedule’s provisions, drafted against the retail defector of the decades before 1985, have no tractable purchase on a wholesale migration executed in a single step. The episode is the limiting case of the arithmetic this section describes, and it fixes the practical boundary of the law more clearly than any commentary: the Schedule governs defections that are small enough to be counted against a fraction, and falls silent, by design or by exhaustion, before defections that carry the fraction with them.

The Anatomy of an Undecided Petition

Between filing and decision, a defection petition leads a life of its own in the secretariat of the House, and the stages of that life are where outcomes are quietly determined. The petition is scrutinised for compliance with the rules, and defects of form, a missing annexure, an unverified copy, an imprecise pleading of the whip, are the first occasion of correspondence and the first consumption of weeks. Copies are forwarded and comments invited; the respondent member seeks an extension for sufficient cause, pleading the volume of the record or the press of the session, and receives it, because a chair who refused a reasonable extension would hand the eventual order a natural justice challenge on a plate. Hearings are fixed and adjourned around the session calendar, which in most state assemblies occupies a minority of the year, and around the availability of counsel, of the petitioner and of the chair itself, whose diary is that of the House’s presiding officer and not of a full time tribunal. Where the chair refers the petition to the Committee of Privileges for a preliminary inquiry, a second body with its own calendar enters the chain, and the Schedule’s rules make the determination wait upon its report. Each of these stages is defensible, most are necessary to a fair proceeding, and their sum, repeated across a dozen petitions in a large episode, is a disposal time that no single participant chose and every participant on one side had reason to welcome.

The content of the hearing stage explains some, but only some, of the consumption. A whip case under paragraph 2(1)(b) requires the petitioner to prove the direction, its issue by the authority the party’s constitution empowers, its communication to the member, and the absence of prior permission or of condonation within fifteen days; a conduct case under paragraph 2(1)(a) requires the tribunal to assemble speeches, meetings, letters and public acts into an inference of voluntary abandonment, an exercise the courts since Ravi S. Naik have permitted but which invites contest at every link. Members are entitled to be heard in person, and in a multi-member episode each is entitled to a separate consideration, since the conduct and the whip’s service may differ member by member. A diligent chair in a twenty member episode has real work to do. But the arithmetic of the decided cases refuses to let diligence carry the explanatory weight: the Himachal Pradesh chair in February 2024 performed the same tasks, for six members, on evidence of the same documentary character, in two days, because the evidence was two days old and the will to conclude was present; and the Maharashtra chair’s eighteen months were not, on the Supreme Court’s own observation in October 2023, a function of the inquiry’s demands, since the grounds of delay recited were procedural wrangles of the kind the Court directed should not be permitted to hold the petitions up.

Two further features of the undecided petition belong to the anatomy. The first is succession: a proceeding that outlasts a change of government or a change of Speaker passes to a new holder of the chair, who inherits the file but not the familiarity, and whose first acts in the matter are liable to be a fresh round of hearings so that the decision, when written, is the decision of the officer who signs it. The Maharashtra petitions were filed into one presiding arrangement and decided by another; the Telangana petitions were filed before a Speaker elected with the government the petitioners opposed and were still pending when the Supreme Court intervened to set the pace. The second is pairing: in a split, petitions run in both directions, each faction accusing the other, and the chair holds a portfolio of proceedings whose combined effect, if all were decided against the petitioners’ opponents, would be to disqualify much of the House. A chair facing paired petitions has an additional, presentable reason for deliberation, the interdependence of the files, and an additional, unspoken one: deciding either set first confers a first mover’s legitimacy on one faction’s account of the party. The anatomy, taken whole, is of a proceeding engineered, by the accumulation of individually reasonable features, to be slow exactly when it matters, and to place the responsibility for the slowness nowhere in particular. That diffusion of responsibility is the missing clock’s true mechanism: there is no deadline to miss, no default to trigger, no officer answerable for the interval, and so the interval is answerable to no one.

Why a Defection Can Succeed, Step by Step

The argument of the preceding sections can be compressed into the sequence through which a contested defection actually travels, with the factor that determines the outcome marked at each step and the point at which the Schedule’s silence operates identified. The table is the map of the practice; the sections around it are the evidence for its rows.

Step in the sequence	What happens at this step	What determines the outcome
Alleged defection	A member crosses, defies a whip, or joins a rival formation; governments fall or are formed on the new arithmetic at once	The act itself; the Schedule attaches no consequence until a tribunal decides
The petition	The abandoned party assembles its papers and files a petition in the form the House rules prescribe	The quality of the file: proof of party, ticket, whip, service and conduct
The chair’s procedure	Copies are forwarded, comments are invited within seven days or an extended period, hearings are held, a committee inquiry is possible	The rules govern the steps; no rule governs the pace between them
The pendency period	The member sits, votes and may hold office; the new majority legislates and budgets	The missing clock: the Schedule fixes no period for decision
Votes during pendency	Divisions and confidence motions are carried by the House as it stands	Article 100(2) and Article 189(2): proceedings stand even if a participant is later held disentitled
The decision	The chair allows or dismisses the petition, often many months or years after the conduct	The identity and incentives of the chair, and the political configuration at the date of decision
Judicial review	Courts test the order on the narrow Kihoto Hollohan grounds, or direct the chair to decide by a date	Review reaches legality, not the lost interval; directions to decide depend on litigants reaching a court
Remaining life of the House	The term runs out; undecided petitions lapse with dissolution	Time itself: a remedy unexercised within the term expires with the term

Read vertically, the table makes the study’s claim concrete. At no step does the merit of the underlying accusation control the result. The accusation’s strength matters at the hearing, if a hearing is reached in time; everywhere else, the outcome is governed by paper, by arithmetic, by the chair’s incentives and by the calendar. A defection succeeds, in the practical sense that it achieves the political object for which it was organised, when the sequence is traversed slowly enough for the new arrangement to become the settled fact of the House. It fails when the sequence is traversed fast, which is why every serious contest under the Schedule is, at bottom, a contest about speed.

The Courts at the Door of the Chair

Judicial involvement in defection adjudication has a fixed constitutional frame, and the frame is narrower than the disappointment it generates. Kihoto Hollohan held the presiding officer, when deciding under paragraph 6, to be a tribunal whose decision is open to judicial review on the grounds on which the orders of tribunals are reviewed: jurisdictional error, violation of natural justice, mala fides and perversity of the order on its own terms, while the bar on interlocutory interference was preserved by the striking down of paragraph 7 and the reasoning that accompanied it. What the decision withheld is as important as what it granted. A court does not sit in appeal over the chair’s assessment of evidence; a court will not ordinarily entertain a challenge while the proceeding before the chair is pending, because intervention at an interlocutory stage was the very mischief paragraph 7 had aimed at and the Court’s quia timet reasoning condemned; and the preservation of review at all was itself justified by the basic structure doctrine, which places judicial review beyond the reach of the ouster the Schedule attempted. The doctrine of the matter is treated in the companion study; the practice is that courts are available at the end of the process, on narrow grounds, and are reluctant at every earlier point, which is to say they are least available at the only point where intervention could change the political outcome, while the interval is still running.

Within that frame, the courts have developed one working instrument, and it is aimed at the calendar rather than at the merits: the direction to decide. The instrument’s genealogy begins with Rajendra Singh Rana v. Swami Prasad Maurya, (2007) 4 SCC 270, where petitions arising out of the Bahujan Samaj Party split in the Uttar Pradesh Assembly had been left in a posture the Supreme Court found indefensible, the Speaker having recognised a split that the Schedule no longer permitted while the disqualification question remained unanswered. The Court not only held that a failure to decide was itself reviewable as a violation of constitutional duty rather than a mere irregularity; it decided the disqualification question itself on the material before it, observing that persons who have incurred disqualification do not deserve to be members of the legislature even for a single day. Rana is the high-water mark of judicial self-help in this field, and later benches have been careful not to treat it as a general licence. The Telangana judgment of July 2025 recounts the line of authority from Kihoto Hollohan through Rana to Keisham Meghachandra Singh and Subhash Desai and then declines, expressly, to decide the petitions before it, remitting them to the Speaker with a three month direction instead. The pattern across four decades is consistent: courts will compel a decision, sometimes by a date, and will review a decision once made; they will almost never supply the decision, because supplying it would collapse the constitutional allocation that Kihoto Hollohan sustained.

The afterlife of Rana repays a sentence of its own, because it shows the system metabolising an exception into a boundary. The circumstances of that case were extreme in combination: a Speaker who had recognised a split the amended Schedule had abolished, petitions whose decision he had simply declined to make, and a term within which the refusal, if indulged, would have decided the matter by expiry. Later benches citing Rana have fastened on its holding that failure to decide is reviewable, and have let its further step, deciding the question in the court itself, recede into the category of the exceptional; the Telangana bench recounts the case at length and then remits, the Maharashtra Constitution Bench remits, and the Keisham bench itself, for all the sharpness of its facts, directs the Speaker rather than deciding. The direction to decide is thus the settled remedy, and its character should be understood: it vindicates the petitioner’s right to an answer while leaving the answer, and the answer’s timing within the directed window, to the very office whose conduct made the litigation necessary. A petitioner who has won everything a court will give has won a date. The contrast with the political side of the episode, where the same courts will order a floor test for the next day, measures the distance between the Constitution’s treatment of governments and its treatment of the adjudication that might unmake them.

Review after decision runs on the same narrow track, and the decided cases show its width in practice. The Karnataka Speaker’s order of July 2019 was challenged on every ground Kihoto Hollohan leaves open, and in Patil the Supreme Court sustained the disqualifications entire, intervening only against the portion of the order that had exceeded the tribunal’s powers by purporting to bar the members from elections; the Maharashtra Speaker’s order of January 2024 was carried in challenge to the Bombay High Court by the faction it had disappointed, on the corresponding grounds that the findings on the party’s constitution and leadership were perverse and contrary to the Constitution Bench’s directions. A challenger must therefore show not that the chair read a party’s documents unwisely, but that the reading was one no tribunal could reach, a standard that protects reasoned orders and exposes only the indefensible. The practical consequence completes the courts’ portion of the map: expedition is litigated before the decision, error is litigated after it, and the decision itself, whenever it arrives, is presumptively the chair’s to have made.

The second front on which courts meet the practice is the floor test, and here the judicial role is the mirror image of the first. Where a government’s majority is in doubt, courts have directed that the doubt be resolved on the floor of the House, promptly and sometimes by a named hour, as the Supreme Court did in March 2020 when it directed the Madhya Pradesh Assembly to hold a floor test the following day in the matter arising from the Congress resignations, and as it did in Maharashtra in June 2022 when it refused to stay the floor test the Governor had called. The floor test doctrine treats the Assembly floor as the sole forum for testing majority, which is constitutionally orthodox; but its interaction with the Tenth Schedule deserves plain statement. A floor test counts the members as they stand, including members under undecided petitions and members whose resignations the chair has not yet accepted, so that a judicially expedited floor test can conclusively settle the political question days or weeks into an episode whose legal question will take months or years. The two judicial instruments thus pull in opposite temporal directions: the direction to decide asks the chair to hurry over months; the direction to test on the floor settles the government in a day. Governments in defection episodes are made by the second instrument and unmade, if at all, by the first, long afterwards.

The third front is the boundary between the chair’s jurisdiction and the Election Commission’s, which Subhash Desai settled for practice in May 2023. A split in a political party generates two proceedings with two different questions: the Speaker asks whether named members have incurred disqualification, a question about the past conduct of individuals, while the Commission, under paragraph 15 of the Election Symbols (Reservation and Allotment) Order, asks which faction is the party for the purposes of the symbol, a question about the organisation’s future. The Constitution Bench held that both authorities may proceed concurrently, that neither must await the other, and that the Commission’s decision operates prospectively and does not disturb the disqualification proceedings or their subject matter. The practical consequence is a division of spoil that the record of Maharashtra illustrates exactly: the Commission, which had frozen the party’s name and symbol by an interim order in October 2022 while the rival claims were heard, recognised the breakaway faction as the Shiv Sena in February 2023, settling the name and the symbol for the elections to come, while the Speaker’s proceedings, concluded in January 2024, settled the past and disqualified nobody. A defection can therefore win the party and keep the seats at the same time, in two forums, on two timetables, neither of which is answerable to the other. The contrast in pace deserves its own emphasis: the Commission, deciding an organisational question of comparable political weight on a record of comparable bulk, traversed its proceeding in about half a year; the chair took eighteen months over the narrower question the Schedule assigns, and the difference cannot be laid to the questions’ relative difficulty, since the Commission’s inquiry ranges over a party’s entire membership and the chair’s over named members and specified dates.

Keisham and the Three Month Direction

The case that converted judicial impatience into a standard is Keisham Meghachandra Singh v. Hon’ble Speaker, Manipur Legislative Assembly, decided on 21 January 2020 and reported at [2020] 2 S.C.R. 132, in a judgment authored by Nariman J. The facts were of a simplicity that damaged every institutional defence of delay. Thounaojam Shyamkumar Singh had been elected to the Manipur Assembly on a Congress ticket in the elections whose results were notified in March 2017; within days he had crossed to the Bharatiya Janata Party and taken office as a minister in the government it formed. Congress members filed disqualification petitions. The Speaker did not decide them. Years passed, applications multiplied, and the petitions remained pending while the member continued as minister, until the petitioners carried the matter through the High Court to the Supreme Court, which was asked, in substance, to answer the question the Schedule leaves open: how long may a chair take?

The Court’s answer has two limbs, and both have governed practice since. The first limb is interpretive: drawing on Kihoto Hollohan, the Court held that the Speaker, acting as a tribunal under paragraph 6, is bound to decide within a reasonable period, and that judicial review which Kihoto Hollohan withheld from interlocutory stages is not withheld from intervention in aid of a prompt decision, the quia timet bar applying to attempts to injunct a decision, not to attempts to obtain one. The second limb quantifies the reasonableness: in the ordinary run of cases, the Court stated, a period of three months from the date of filing should be the outer limit within which the Speaker decides, absent exceptional circumstances, and in the case before it the direction was sharper still, four weeks. The Court then went beyond the case to the design. Noting that the Speaker is a member of a political party and that disqualification petitions against government-sustaining defectors would predictably not be decided impartially by such a hand, it stated that Parliament should seriously consider substituting the Speaker with a permanent tribunal headed by a retired Supreme Court judge or a retired Chief Justice of a High Court, or some other outside independent mechanism, so that disputes are decided swiftly and impartially. No amendment followed. The direction stands as the operative standard; the recommendation stands as the authoritative diagnosis, unacted upon.

The sequel in Manipur supplies the standard’s measure. The four week direction was not met with a decision; an extension was sought and the matter returned to the Court, which on 18 March 2020, exercising its powers under Article 142, restrained the member from entering the Assembly and recorded that he had ceased to be a minister with immediate effect, an extraordinary interim outcome in which the political consequences of disqualification were imposed by judicial order before any tribunal had found the disqualification. The Speaker decided on 28 March 2020, disqualifying the member and rejecting the argument that he had never been a Congress member of the Assembly because he had crossed before taking his oath, membership being traced to the Election Commission’s notification of the elected members under Article 170 and Section 73 of the Representation of the People Act, 1951 rather than to the oath. The petition had thus taken about three years, a Constitution-cited direction, a further extension and an Article 142 restraint to produce. When the Manipur Assembly was dissolved ahead of the 2022 elections, other petitions from that term which the Speaker had still not decided lapsed with the House, unrescued by the standard Keisham had announced, because a direction to decide binds a chair in office over a House in being, and dissolution ends both.

What Keisham changed, and what it did not, should be stated with care. It changed the enforceability of expedition: after January 2020, a petitioner anywhere in India could go to a constitutional court with a citable outer limit and expect, in a plain case, a direction. The Maharashtra and Telangana proceedings both ran through the channel Keisham cut. It did not change the identity of the decider, which its own reasoning doubted; it did not create a consequence for a chair who outruns even a directed deadline other than the escalating embarrassment of further orders; and its limit remains a judicial standard of reasonableness, not a rule of the Schedule, so that each fresh episode must still be litigated to the point of a direction before the clock starts. A standard that must be invoked is a standard that operates only for petitioners with the resources and the standing to invoke it, in the episodes prominent enough to carry to a constitutional court. Below the waterline of reported litigation, in the assemblies where petitions sleep unreported, the missing clock is missing still.

The standard’s career since 2020 also shows the judiciary negotiating with itself about how far the construction should go. When the Telangana petitions first reached the courts, a division bench of the High Court fixed outer limits for the Speaker’s decision, converting Keisham’s standard into a local timetable. At the Speaker’s instance, the Supreme Court set those limits aside in November 2024, holding that a High Court could not prescribe a timeline breaching the discipline of Kihoto Hollohan, while reiterating that the decision must come within a reasonable time; within eight months the same institution, confronted with the sequel, itself directed a decision within three months and spoke of the trust reposed in the office not having been honoured. The oscillation is intelligible: a fixed judicial timetable in every case would be legislation by another name, and a reasonableness standard without dates is a sermon. The courts have therefore settled, in practice, on an escalatory ladder: exhortation first, a Keisham direction on proof of sleep, sharper language and consequences on default. A ladder, however, must be climbed case by case, petitioner by petitioner, and the climbing takes the very resource whose scarcity created the problem, which is time within the term of a House.

Maharashtra: A Defection Decided After Its Work Was Done

The Maharashtra episode of 2022 to 2024 is the most fully documented traverse of the sequence this study has mapped, and it rewards a continuous telling from the record. In June 2022 a large group of Shiv Sena members of the Legislative Assembly, led by Eknath Shinde, withdrew from the party’s legislature ranks and aligned with the Bharatiya Janata Party, depriving the government of Uddhav Thackeray of its majority. Disqualification petitions followed at once, from both directions: the Thackeray side petitioned against the breakaway members, and the breakaway side petitioned against members who had remained. On 25 June 2022 the Deputy Speaker issued notices on the petitions against sixteen members of the breakaway group, at a time when a notice for the Deputy Speaker’s own removal, moved by that group, was pending, which raised the Nabam Rebia objection in a live case. The Governor called a floor test for 30 June 2022; the Supreme Court refused to stay it; Uddhav Thackeray resigned on 29 June 2022 without facing the test, and Eknath Shinde was sworn in as Chief Minister on 30 June 2022. The political question raised by the defection was thus settled, completely and irreversibly, within days, by the ordinary operation of majority rule on the floor. The legal question had barely been filed.

The legal proceedings then ran their long course in both available forums. In the Election Commission, the rival factions contested the party itself, and by an order in February 2023 the Commission recognised the breakaway faction as the Shiv Sena and allotted it the party’s symbol, a decision the Constitution Bench in Subhash Desai later confirmed operates prospectively and does not touch the disqualification question. In the Supreme Court, the petitions and the challenges to the Governor’s floor test and to the new Speaker’s recognition of the breakaway faction’s whip were heard by a Constitution Bench, which delivered judgment on 11 May 2023 in Subhash Desai v. Principal Secretary, Governor of Maharashtra, reported at 2023 INSC 441. The Court held that the Governor had lacked objective material to call the floor test and could not use a floor test to resolve an intra-party dispute; that it could not restore the Thackeray government, since the Chief Minister had resigned voluntarily; that the whip of a party is appointed by the political party and not by the legislature party, that directions to vote or abstain issue from the political party, and that the Speaker must recognise only a whip and leader duly authorised by the political party, determined with reference to the party’s constitution; that the new Speaker’s recognition of the breakaway faction’s whip in July 2022 was contrary to law; that disqualification is not automatic, dissent being insufficient by itself to establish a voluntary giving up of membership; that members remain entitled to participate in the proceedings of the House until the Speaker decides; and that the disqualification petitions were for the Speaker to decide, not for the Court to decide in the first instance. Every institutional question was answered. The petitions went back to the chair.

Eighteen months, measured against a five year term

The Maharashtra petitions were filed in June and July 2022 and decided on 10 January 2024: roughly eighteen months inside a term that ran from October 2019 to November 2024. The decision therefore covered a ministry already formed, budgets already passed and a party already recognised elsewhere, and it altered none of them.

The petitions were filed in June and July 2022. The Speaker’s decision is dated 10 January 2024. The interval is roughly eighteen months, within a House elected in October 2019 whose term ran to November 2024: the decision consumed a third of the Assembly’s life from the date of the conduct it judged, and it arrived about ten months before the House ceased to exist. Measured against the tenure of the arrangement the petitions challenged, the interval covers the entire ministry formation, two budget cycles and the consolidation of a new party structure recognised by the Election Commission. The arithmetic, and not any defect of drafting in the petitions, is why the decision changed nothing it decided about.

The interval was not spent in inactivity, and its management is instructive. After the Constitution Bench remitted the petitions, the Speaker proposed a schedule that would have concluded the proceedings by the end of February 2024. The petitioners returned to the Supreme Court in Sunil Prabhu’s plea, and on 30 October 2023 a bench of Chandrachud CJI, Pardiwala and Misra JJ directed that the Shiv Sena proceedings be concluded by 31 December 2023 and the parallel Nationalist Congress Party proceedings, arising from that party’s split of July 2023, by 31 January 2024, observing that the procedural wrangles the Speaker had recited must not be permitted to delay the petitions and that the sanctity of the Tenth Schedule must be maintained. The Sena decision, when it came on 10 January 2024, was thus delivered past the directed date, and the NCP decision followed in February 2024, recognising the Ajit Pawar faction as the real Nationalist Congress Party and likewise disqualifying no member. The two decisions, delivered weeks apart by the same chair out of the same political convulsion, were immediately observed in the public debate to rest on different measures of party identity, the party’s constitution and leadership structure carrying the Sena question while the weight of legislative numbers carried the Nationalist Congress Party question, a divergence the losing factions pressed in challenge. This study records the divergence for a narrower reason than the controversy assigned to it. A tribunal deciding eighteen months after the event, in a House the event had transformed, is not merely deciding late; it is deciding a different question from the one the petitions posed, because the party whose identity it must find has itself been remade in the interval by recognition, by organisation and by the simple accretion of time. Delay does not preserve the subject matter of defection adjudication. It replaces it. The decision itself, by Speaker Rahul Narwekar, held the breakaway faction to have been the real Shiv Sena when rival factions emerged on 21 June 2022, determined by reference to the party constitution of 1999 and the leadership structure on the Election Commission’s record rather than the structure of 2018, recognised the breakaway faction’s chief whip as validly appointed and the other faction’s whip as having ceased to hold that office from 21 June 2022, and on those findings dismissed the disqualification petitions of both factions, disqualifying no member on either side. A proceeding begun to expel the defectors ended by certifying the defection’s own account of the party. That outcome was available on the merits as the Speaker found them; the study’s point is the temporal one. Had the same findings been recorded in July 2022, they would have decided who governed Maharashtra. Recorded in January 2024, they decided only how the episode would be described.

Telangana, running a few years behind on the same rails, confirms that Maharashtra was not an aberration. In March and April 2024 the Bharat Rashtra Samithi petitioned against ten of its members who had crossed to the ruling Congress. The petitions were resisted procedurally, a division bench of the Telangana High Court fixed outer limits for the Speaker which a bench presided over by Khanna CJI set aside at the Speaker’s instance in November 2024, the Court holding that the High Court could not prescribe a timetable though the decision must come within a reasonable time, and notices in the proceedings were first issued by the Speaker’s office on 16 January 2025, the day after the matter had been moved in the Supreme Court. On 31 July 2025, in Padi Kaushik Reddy v. State of Telangana, a bench of Gavai CJI and Masih J traced the whole line of authority, held that the Speaker deciding under the Schedule acts as a tribunal and enjoys no constitutional immunity as such, directed decision within three months, and recorded the sentence that will be quoted whenever this subject is taught: Parliament had trusted the high office of the Speaker to act expeditiously, and that trust, in many cases, had not been honoured. The sequel ran into court again within months, the Supreme Court reacting with fury to the non-compliance reported to it in November 2025, and into 2026, when the Speaker’s dismissal of one of the petitions was set aside by the High Court, which itself declared that member disqualified, and the member carried the matter upward. An episode that began with crossings in 2024 was therefore still generating appellate litigation in 2026, within the life of the same Assembly, with the crossings’ political objects long since achieved.

Resignation: The Door the Schedule Cannot Close

The second route out of a defection’s legal consequences is older than the Schedule and is built into the Constitution itself: the member resigns the seat. Articles 101(3)(b) and 190(3)(b) permit a member to resign by writing addressed to the presiding officer, and the seat falls vacant when the resignation is accepted, subject to the proviso, inserted by the Constitution (Thirty-third Amendment) Act, 1974, that the presiding officer shall not accept a resignation if satisfied, after such inquiry as the officer thinks fit, that it is not voluntary or genuine. Resignation changes the legal posture of an episode completely. The defection, if any, was complete in conduct; but a disqualification proceeding decides whether a member has become subject to disqualification, and once the seat is vacated there is no membership left for the proceeding to bite on in the practical sense that matters, while the resigned member, never having been disqualified, carries no incapacity into the by-election for the vacant seat and may contest it at once, typically on the ticket of the receiving party. Patil confirms the asymmetry from the other side: it is disqualification that is impotent to bar a member’s return, since the Court held the Speaker has no power to bar a disqualified member from contesting; resignation achieves the same return without even the stigma of an adverse order, provided the chair accepts the letter.

Karnataka in 2019 is the worked example. Through July 2019, seventeen members of the ruling Congress and Janata Dal (Secular) coalition, and one independent member supporting it, resigned their seats or were proceeded against, in a sequence that stripped the government of H. D. Kumaraswamy of its majority; the Chief Minister lost the confidence of the House later that month and a Bharatiya Janata Party government followed. The resignations were contested as instruments: the Speaker, K. R. Ramesh Kumar, held the letters for inquiry into their voluntariness and genuineness, granted time, and was himself litigated against for the pace of his inquiry, the resigned members contending that an inquiry into motive was beyond the proviso. In the event the Speaker disqualified all seventeen members under the Schedule on 25 and 28 July 2019, the disqualification proceeding being held capable of running alongside and after the tendered resignations, with the disqualification, where found, relating back to the conduct. On the challenge, Shrimanth Balasaheb Patil v. Hon’ble Speaker, Karnataka Legislative Assembly, decided on 13 November 2019 and reported at (2020) 2 SCC 595, a bench of Ramana, Khanna and Murari JJ upheld the disqualifications in full, and simultaneously confined the institution at both ends: the Speaker’s power over a resignation extends only to whether it is voluntary and genuine and cannot travel into the member’s reasons, and the Speaker’s power on disqualification does not extend to barring the disqualified member from contesting elections for the remainder of the term, the offending portion of the order being struck down for want of support in Articles 191, 164(1B) or 361B. The December 2019 by-elections then returned most of the disqualified members who contested on the ticket of their new party, and several later held ministerial office. Every institution in the chain did what the law empowered it to do, and the defection’s transfers of allegiance emerged from the far end substantially intact.

Madhya Pradesh in March 2020 shows the resignation route operating at maximum political effect without any disqualification at all. Twenty-two Congress members of the Assembly, supporters of Jyotiraditya Scindia, who had joined the Bharatiya Janata Party, resigned their seats, more than enough to destroy the majority of the Kamal Nath government. The Speaker accepted some of the resignations promptly and held others for inquiry, the Governor directed a floor test, the Supreme Court on 19 March 2020 ordered the test to be held the next day, and the Chief Minister resigned on 20 March 2020 ahead of it, the remaining resignations being accepted as the government fell. No petition under the Tenth Schedule decided anything in that episode, because there was nothing left for it to decide: the seats were vacant by resignation, the government was gone, and the subsequent by-elections, in which the resigned members contested on their new party’s ticket, were the only adjudication the episode received. The contrast with Karnataka, where the chair resisted and disqualified and the same political result followed one litigation later, is the cleanest demonstration in the record that the Schedule’s sanction and the defection’s success are independent variables. A chair who fights produces a judgment; a chair who accepts resignations produces a vacancy list; the government changes either way.

The deeper reason resignation escapes the Schedule is conceptual, and practice has never seriously disputed it. The Schedule attaches consequences to the conduct of a member: voluntary giving up of party membership, defiance of the whip. Resignation of the seat is conduct of a different constitutional character, an exercise of the member’s own right to vacate, supervised only for voluntariness and genuineness, and it is available to the honest dissenter and the purchased defector on identical terms. A regime that disqualified the resigned member from the by-election would need a constitutional text that does not exist, as Patil’s striking down of the Karnataka Speaker’s attempt to construct one establishes. What the practice has settled into, therefore, is a division of consequences by route: cross and stay, and face a petition of indefinite duration under a chair of variable speed; cross and resign, and face only the electorate of the by-election, at a moment of the receiving party’s choosing, with the government already changed by the vacancies. Reform proposals that address the adjudicator and the clock leave this door exactly as open as they found it, a point the reform discussion must carry honestly.

The genuineness inquiry itself contains a miniature of the whole problem, and practice has noticed. The Thirty-third Amendment’s proviso empowers the chair to refuse a resignation that is not voluntary or genuine, after such inquiry as the chair thinks fit; it fixes, once again, no period for the inquiry and prescribes no consequence for its prolongation. In the Karnataka episode the inquiry became the episode’s second front: the resigned members, several of whose letters had reached the chair while the members themselves were away from the State capital in the company of the receiving party’s managers, complained that the inquiry was a device for keeping their seats in suspense while the government’s fate was negotiated, and the chair answered that letters procured from members incommunicado were exactly what the proviso existed to test. Both accounts were available on the law, because the proviso is a power without a calendar, like paragraph 6. Patil in the end confined the inquiry’s scope to voluntariness and genuineness and excluded motive, which disciplines its content; nothing has disciplined its duration, and a chair minded to hold resignations in suspense holds, in the inquiry, a second instrument of delay that operates before the Schedule is even reached. The government’s majority in that interval is counted with the resigned members still in their seats, which is to say the inquiry’s pace, like the petition’s pace, is a political fact dressed as a procedural one.

When the Chair Moves Fast

Against the long record of delay, the episodes of speed are valuable precisely because they isolate the variable. Himachal Pradesh in February 2024 is the cleanest case. In the Rajya Sabha election held on 27 February 2024, six members of the ruling Congress, Rajinder Rana, Sudhir Sharma, Inder Dutt Lakhanpal, Devinder Kumar Bhuttoo, Ravi Thakur and Chaitanya Sharma, voted for the candidate of the Bharatiya Janata Party, and on 28 February 2024 they absented themselves from the proceedings on the Finance Bill and the cut motions despite the party’s whip. The parliamentary affairs minister filed Disqualification Petition Nos. 112 to 117 of 2024, and Speaker Kuldeep Singh Pathania decided them on 29 February 2024, holding, in a reasoned order running to thirty pages, that by their conduct the six had voluntarily given up membership of their party within paragraph 2(1)(a), applying the line of authority from Ravi S. Naik through Subhash Desai that such conduct may be inferred from what a member does, and recording that the six ceased to be members of the Assembly with immediate effect. The order was the first disqualification in the State’s history. The six challenged it in the Supreme Court within days; within weeks they had joined the party they had voted for, and in June 2024 they faced the by-elections their disqualification had made necessary. From conduct to vacancy took two days. No provision of the Schedule had changed to make that possible; the same paragraph 6, the same absence of a clock, produced a two day disposal in Shimla and a three year pendency in Imphal. The variable was the chair, and behind the chair, the alignment of the chair’s government with the petition’s success.

The Himachal disposal is not the only exhibit. The Chairman of the Rajya Sabha disposed of the petitions against Sharad Yadav and Ali Anwar, members whose party had petitioned after they aligned against it, in December 2017, within about three months of their filing, and defended the disposal publicly in 2019 by recalling the piloting minister’s standard of 1985, fearlessness and expedition, in terms that read as a standing rebuke to slower chairs. What speed requires, institutionally, is modest: a petition in order, comments on a short leash, hearings on consecutive days, and a chair prepared to write a reasoned order without waiting for the political weather. The rules already permit every element of it, since the seven days for comments and the hearing are the only temporal features the scheme contains and both are compatible with a monthly timetable. Speed, in other words, is administratively available in every House in the country on any given Monday. Its rarity is the measure of how rarely the incentive structure supplies it, which is why the study treats the fast cases as the control group of the whole experiment: they prove the delay is chosen, by the structure if not by the individual, and not imposed by the law.

A second lesson of the fast cases concerns the quality of decision that speed makes possible, and it cuts against an assumption the slow cases encourage, that delay at least purchases deliberation. The Himachal order was reasoned, rested on identified conduct and applied the governing authority. The Maharashtra order, by contrast, arrived after eighteen months and a constitutional reference’s worth of litigation with reasoning that turned on a contestable choice between two versions of a party constitution, a choice the petitioners have continued to dispute. Delay is not deliberation; it is delay. A tribunal that sits on a file for eighteen months has not spent eighteen months thinking, and the record of outcomes gives no support to the notion that the long-decided cases are better decided than the short ones.

The Votes That Cannot Be Uncast

One consequence of the system deserves a section of its own, because it is the point at which delay becomes irreversible in the strictest sense. A member under a pending petition votes, and the vote counts. If the member is later disqualified with effect relating back to the defection, the votes cast in the interval are not reopened. The constitutional basis is express: Article 100(2) for Parliament and Article 189(2) for the state legislatures provide that a House has power to act notwithstanding vacancies and that proceedings are valid notwithstanding that it is discovered afterwards that a person who was not entitled to do so sat or voted or otherwise took part in them, and Articles 122 and 212 bar courts from inquiring into the validity of proceedings on the ground of irregularity. The defector’s budget votes stand; the confidence motion carried or lost on the defector’s vote stands; the legislation passed by the majority the defector composed is law in every respect. The disqualification, when it comes, is a judgment about the member. It is not, and cannot be made into, a judgment about the interval.

The same constitutional habit of leaving the interval undisturbed runs through the neighbouring territory. The comparison the petitioners themselves draw is with the election petition, the other proceeding that can unseat a member mid-term, and the comparison is instructive in both directions. An election petition under the Representation of the People Act, 1951 is tried by a High Court, with Section 86(7) directing that endeavour be made to conclude the trial within six months of presentation, a direction honoured loosely enough that election petitions and disqualification proceedings alike are known to outlast the Houses they concern. Yet the election petition carries structural features the defection petition lacks: a court as the tribunal, a statutory expression of the expected pace, and a consequence, the voiding of the election, that erases the member’s tenure in law even if it cannot unpass the laws the member voted for. The defection proceeding has none of the three. Its tribunal is interested, its pace is unexpressed, and its consequence is personal. The Schedule’s framers cannot have been unaware of the model; the 1951 Act was the standing template for member-removing proceedings, and the departures from it were the substance of the 1985 choice.

The practical teaching of the uncastable vote is absorbed by every participant in the system. Governments defend their majorities on the floor, not in the tribunal, because the floor is where the majority exists while it matters. Petitioners treat the petition as a record for history, a deterrent for the next episode and a lever for judicial directions, rather than as an instrument for recovering the government they lost. Defectors price the petition at its real value: a personal risk, deferred, survivable, and in the merger and resignation variants avoidable altogether. And voters, whose verdict is the input the whole apparatus exists to protect, are presented with a fait accompli that no institution will later undo on their behalf. That the votes cannot be uncast is not a defect of drafting; it is a principle of legislative continuity without which no legislature’s acts could ever be final. But it completes the study’s account of why timing is everything under the Tenth Schedule. A remedy that cannot reach backwards must arrive in time, or it does not arrive at all.

Nor is there a side sanction that reaches the interval. Articles 104 and 193 impose a monetary penalty on a person who sits or votes in a House when disqualified, recoverable as a debt, and the provisions are sometimes pointed to as a deterrent against the defector’s interval. Their text addresses the person who sits or votes knowing the disqualification, and their operation presupposes the disqualification as an established fact; a member under a pending petition is, until the order, a member in full standing whose sitting and voting are authorised, and the subsequent relation back of a disqualification to the date of conduct has never been carried so far as to convert authorised interval participation into the penalty’s subject matter. The interval is therefore not merely unremedied in the sense that its political products stand; it is unpriced. No fee, fine or recovery attaches to the months during which a member later held disqualified composed a majority, drew a ministerial salary or cast the deciding vote on a budget. The full cost of the missing clock is borne by the side that played by the timetable of an ordinary litigant, which is to say by the petitioner and, behind the petitioner, by the electorate whose verdict the defection rearranged.

Even the endgame, when it finally arrives, runs on a further clock that dilutes it. A disqualification vacates the seat, and Section 151A of the Representation of the People Act, 1951 then requires the by-election to fill the vacancy to be held within six months, subject to exceptions where the remainder of the term is short or the Election Commission certifies difficulty in holding it. The six month allowance is sensible electoral administration and, in the arithmetic of a finely balanced House, a further transfer of advantage: for up to half a year after the tribunal has spoken, the seat stands empty, the majority that the disqualified member composed is reduced by absence rather than restored to its opponents, and the government the defection built governs on the smaller number. When the by-election is held, the disqualified member may contest it, Patil having settled that the chair cannot bar the attempt, and may return to the House within the same term, converted by the interval and the election from a defector under a cloud into a member with a fresh and personal mandate. The full cycle of the remedy, petition to decision to vacancy to by-election to possible return, is thus measured in years even when every institution in the chain performs respectably. Set beside that cycle the defection’s own cycle, which is measured in days, and the study’s subject stands complete: the law and its breach move at speeds so different that they might be described as inhabiting different constitutions, one written and slow, the other practised and immediate.

Changing the Decider, the Clock, or Neither

Reform of defection adjudication has a settled menu, unusually stable across four decades, and each item on it can be assessed against the practice record rather than in the abstract. The oldest and most repeated proposal is to change the decider: to take the question away from the presiding officer and give it to the President or the Governor, to be decided on the opinion of the Election Commission, which is the route Articles 103 and 192 prescribe for every other disqualification of a sitting member. The Dinesh Goswami Committee on Electoral Reforms recommended exactly this in 1990, five years after the Schedule’s enactment; the Law Commission’s 170th Report recommended it in 1999, alongside the deletion of the split and merger exemptions; the National Commission to Review the Working of the Constitution recommended it in 2002; and the Election Commission’s own statement of proposed electoral reforms endorsed it in 2004. The proposal’s strength is that it severs the adjudication from the majority the adjudication affects, since the Commission’s opinion binds through the formal decision of the President or Governor and the Commission itself holds no stake in a state government’s survival. Its weakness, which the practice record throws into relief, is that it substitutes one institution’s calendar for another’s: the Commission route under Articles 103 and 192 has itself produced long pendencies in non-defection disqualification references, and a change of decider without a discipline of time would reproduce the central defect of the practice in a more respectable office. The question the proposal answers is who decides. The record’s question is when.

The second item is the tribunal: the permanent, independent body that Keisham Meghachandra Singh invited Parliament to consider in January 2020, headed by a retired Supreme Court judge or a retired Chief Justice of a High Court or constituted as another outside independent mechanism. The tribunal proposal answers the partiality objection directly and, if constituted with a small and specialised docket, could in principle decide fast as well as independently; it is the only proposal that addresses both variables of the One Test at once. Its costs are constitutional and practical. Divesting the chair under paragraph 6 requires amending the Schedule, and an amendment replacing the internal adjudicator with an external tribunal reopens the settlement of Kihoto Hollohan, under which the chair’s role was sustained partly on the dignity of the House’s self-government; and a tribunal appointed through the ordinary processes of the executive would face, at its first controversial order, the same contest over its composition that attends every appointment to bodies that decide political fortunes. The case for the presiding officer, stated at its full strength because it deserves to be, is that the composition of a House is the House’s own affair, that a chair embedded in the House knows a whip from an intimation and a revolt from a misunderstanding in a way no outside body can, and that the remedy for chairs who abuse the trust is the one the Constitution already supplies for Speakers generally, removal by the House, supplemented by the judicial directions that have demonstrably worked in Manipur, Maharashtra and Telangana. The case for change, at its full strength, is the record itself: forty years of asymmetry between petitions that threaten governments and petitions that threaten only the opposition, set out in the cases above, and a Supreme Court that has moved from trusting the office in 1992 to directing it by dates in 2020, 2023 and 2025 and to recording in terms that the trust placed in it has in many cases not been honoured.

A time limit fixes the wait and not the decider

A written time limit of the Keisham measure, three months from filing, would make expedition a rule instead of a favour obtained by litigation, and it would protect petitioners who cannot reach a constitutional court. It would not change who decides, so a chair minded to favour its own side would still hold the pen when the period expired.

A statutory limit, written into the Schedule, of ninety days or three months from filing, with the petition deemed decided or transferred on default, is the reform aimed squarely at the missing clock. It would convert Keisham’s litigated standard into a rule that operates without anyone’s leave, and it is the only proposal under which a petitioner in an unreported assembly proceeding would be protected as fully as a petitioner who can reach the Supreme Court. It leaves the interested chair in place, and so leaves the asymmetry of outcomes untouched in cases the chair decides quickly in its own interest; and a limit without a default consequence invites the jurisprudence of condonation to grow around it. What it solves is the variable the record shows to be decisive in the worst cases, the open ended wait. What it does not solve is who holds the pen when the wait ends.

The Constitution’s comfort with undated trusts, it should be noted, is not peculiar to the Schedule. The text elsewhere arms high offices with powers whose timing it leaves substantially to the office holder’s judgment, as with the ordinance making power vested in the President and the Governors, bounded by the life limits of Articles 123 and 213 rather than by any discipline of occasion; and in those fields too, practice has filled the silence with convention, litigation and judicially announced limits, re-promulgation doctrine being the closest cousin of the Keisham standard. The Schedule’s silence is thus an instance of a general constitutional technique, distinguished only by the stakes: an undated adjudicatory trust over the very membership of the deciding body, in which the passage of time does not merely inconvenience a litigant but transfers power between governments. Reforms that would touch neither the decider nor the clock, and there are respectable examples, address the margins of the practice: tighter rules on the form and service of whips, published cause lists for defection hearings, reasoned interim orders on a petitioner’s application for expedition. Each would improve the record’s transparency. None would alter its central asymmetry, because the central asymmetry is not an accident of administration to be tidied but a consequence of two design choices, the insider decider and the missing clock, which between them allocate the outcome of every contested episode to the politics of the chair and the calendar. A reform debate that does not name those two choices is a debate about a different law.

The study takes no side in that debate beyond the side the evidence occupies, which is this: the criteria by which any proposal should be judged are the two the One Test identifies, the independence of the decider and the certainty of the timetable, and a proposal that improves neither is not a reform of the practice whatever else it may be. The Goswami route improves the first and leaves the second to the habits of a new office; the tribunal improves both if it is given a docket discipline and fails the second if it is not; the time limit improves the second completely and the first not at all; and the status quo, judged by the same criteria, offers a decider whose partiality the Supreme Court has described in terms and a timetable whose existence depends on litigation. Readers weighing the proposals also carry a plainer duty of accuracy: this article explains the law and the record for general information and is not legal advice, and anyone advising a member, a party or a petitioner in a live proceeding must work from the text of the Schedule, the rules of the House concerned and the orders in that proceeding.

One design question runs beneath the menu and determines whether any item on it works: the consequence of default. A time limit without a consequence is a request, and the four decades of the Schedule are, in one reading, the record of a request being declined. The variants a serious amendment must choose among are each instructive. The petition might stand transferred, on the expiry of the period, to another forum, which converts the limit into a jurisdictional trigger and gives the chair a reason to decide that no exhortation has supplied; or the member complained against might stand suspended from voting on the expiry, which prices delay for the side that profits from it but punishes a member against whom nothing has been proved; or the petition might be deemed allowed, which no responsible design would adopt, since it would let an unexamined accusation unseat a member by efflux of time. The transfer variant is the only one that disciplines the decider without visiting the default on the parties, and its absence from the enacted text, and from most of the debated proposals, is of a piece with the Schedule’s general habit: the system is reluctant, at every joint, to attach consequences to its own inaction. The Goswami route and the tribunal route inherit the same design question on arrival, since a Commission opinion or a tribunal can also be awaited indefinitely unless the instrument constituting it answers for the wait. Reform of this subject is therefore, at bottom, a single question asked of whichever decider is chosen: what happens, automatically and without further litigation, if you have not decided in time? The present text’s answer, nothing, is the practice this study has recorded.

The Remedy Outlives the House

The Tenth Schedule in practice is a study in the distance between a sanction and its object. The sanction is disqualification, and it is severe: loss of the seat, and under Articles 164(1B) and 361B a bar on ministerial office and remunerative political posts until re-election or the expiry of the term. The object was the stability of governments and the integrity of the voter’s mandate, and against that object the practice records a consistent pattern across four decades and every region: lone defectors and small groups are caught, sometimes quickly; organised crossings executed at scale, by merger arithmetic, by mass resignation or behind the shield of a friendly chair’s calendar, succeed in their political object whether or not they are eventually condemned in law. The provision disqualifies the defector. It does not, in the cases that matter most, defeat the defection, because the defection’s work is done in the interval the provision leaves ungoverned, on the floor of a House whose proceedings the Constitution will not afterwards disturb.

The One Test survives the record intact. Ask of any episode who decided and when, and the outcome is explained: Imphal’s three years and Shimla’s two days proceeded under the same text; the Maharashtra petitions were decided by a chair the defection had installed, at a date the defection’s consolidation had made safe; the Karnataka disqualifications were complete in law and irrelevant in arithmetic within five months; the Madhya Pradesh government fell without the Schedule being engaged at all. The grounds, the whips, the texts and the doctrines, all the matter of the companion study, are the language in which these outcomes are afterwards expressed. The outcomes themselves are decided upstream, in the allocation of the adjudication to an interested office and in the absence of any period within which that office must speak. Until one of those two allocations changes, the practice of the Tenth Schedule will continue to be written by the calendar, and the remedy will continue to arrive, where it arrives, in time to judge the past and too late to govern it.

What the record would look like under a working design can be stated briefly, because the fast cases have already shown it. Petitions would be decided while the governments they concern are still the governments of the day, so that disqualification would operate on live majorities and the threat of it would enter the calculations of members contemplating a crossing before the crossing, as deterrents must. Chairs would write their orders on fresh evidence about parties whose structures had not yet been remade by recognition and time. Courts would return to the end of the process, reviewing decisions that exist, instead of spending their authority on directions to produce decisions at all. None of this requires imagining an institution India lacks; it requires only that the two allocations the record indicts, the interested decider and the missing clock, be addressed by the amendment the reports of 1990, 1999 and 2002 already drafted in substance. That the amendment has not come, across four decades in which the defect it would cure has been documented by committee, commission and Constitution Bench alike, is the final datum of the practice, and perhaps its most eloquent: the system’s beneficiaries sit in the only chamber that can repair it, and the remedy, in the assemblies as in the Republic, continues to outlive the House.

Frequently Asked Questions

Q: Why do defection petitions take so long to decide?

The Tenth Schedule fixes no period for decision, and the House rules made under paragraph 8 prescribe the steps of the proceeding, petition, forwarding, comments, hearing, without prescribing the pace between them. A chair who wishes to move slowly breaches no deadline, because none exists. The incentives of the office point the same way: the chair is elected by the majority the petition usually threatens, and delay preserves that majority. Hearings also compete with the session calendar, and a reference to the Committee of Privileges adds a further stage. Expedition, where it happens, is therefore a choice of the chair or the product of a court direction, not a requirement of the scheme.

Q: Has the Supreme Court set a time limit for the Speaker?

The Schedule itself contains no time limit, and the Court has not written one into it. What the Court did in Keisham Meghachandra Singh in January 2020 was to hold that the Speaker, acting as a tribunal, must decide within a reasonable period, and to state that three months from the filing of the petition should be the outer limit in the ordinary case, absent exceptional circumstances. That standard has been applied through directions in later cases, including the Maharashtra proceedings in 2023 and the Telangana proceedings in 2025. It remains a judicial standard that a petitioner must invoke, not a rule that operates by itself, and a High Court’s attempt to convert it into a fixed timetable was disapproved even while the duty of reasonable expedition was reaffirmed.

Q: What did the Court decide about the Speaker and the whip in the Maharashtra case?

In Subhash Desai, decided on 11 May 2023, the Constitution Bench held that the whip and the leader of a legislature party are appointed by the political party, not by the legislature party, and that directions to vote or abstain issue from the political party. The Speaker, when rival claims arise, must recognise only the whip and leader duly authorised by the political party, determined after an enquiry conducted with reference to the party’s constitution. On that footing the Court held that the Maharashtra Speaker’s recognition in July 2022 of the breakaway faction’s whip was contrary to law. The Court still declined to decide the disqualification petitions itself, holding that they were for the Speaker to decide.

Q: Who issues the whip, the party or the legislature party?

The political party issues the whip. Subhash Desai settled the question in May 2023: the authority to appoint the whip and the leader, and to direct how members vote or whether they abstain, belongs to the political party, and the legislature party acts through the office holders the political party has authorised. A direction that comes from the legislature party without the political party’s authority behind it will not sustain a disqualification under paragraph 2(1)(b). Where factions dispute the authority, the presiding officer must test the claims against the party’s constitution rather than against the numbers each faction can parade, which is itself one of the inquiries that consumes time in a live proceeding.

Q: Does resigning avoid disqualification?

Resignation, once accepted, vacates the seat, and a disqualification proceeding has no membership left to act upon, so a member who resigns before any order is never disqualified and may contest the resulting by-election at once. The presiding officer may hold the resignation only to inquire whether it is voluntary and genuine, and that inquiry, as Shrimanth Balasaheb Patil settled, cannot extend to the member’s motives. The route is not quite a safe harbour: proceedings can run alongside a tendered resignation, and in Karnataka in 2019 the Speaker disqualified the members notwithstanding their letters, the disqualification relating back to the conduct. Where resignations are accepted first, as in Madhya Pradesh in 2020, the Schedule is never engaged at all.

Q: Has the anti defection law stopped defections?

The reported record supports a divided answer. Individual crossings and small group defections are deterred and, when attempted, are punished often enough, especially where the chair’s incentives favour expedition, as the Himachal Pradesh disqualifications of February 2024 show. Organised crossings at scale continue: the Goa merger of July 2019, the Karnataka and Madhya Pradesh resignation episodes of 2019 and 2020, the Arunachal Pradesh migration of 2016 and the Maharashtra split of 2022 each changed or sustained governments, and none was undone by the law. The Supreme Court’s own assessments have hardened over the same ground, from the trust expressed in Kihoto Hollohan to the observation in the Telangana judgment of 2025 that the trust reposed in the Speaker’s expedition had in many cases not been honoured.

Q: Should an independent tribunal decide defection cases?

The proposal has the strongest institutional pedigree of any reform in this field: the Supreme Court itself, in Keisham Meghachandra Singh, invited Parliament to consider substituting the Speaker with a permanent tribunal headed by a retired Supreme Court judge or a retired Chief Justice of a High Court, or another outside independent mechanism, so that disputes are decided swiftly and impartially. Supporters point to the chair’s structural interest in the majority under threat. Opponents answer that the composition of a House is the House’s own affair, that an external tribunal would need a constitutional amendment and would face contests over its own appointment, and that the judicial directions of 2020, 2023 and 2025 show the existing forum can be made to work. The proposal remains unadopted.

Q: What happens to a government while a defection petition is pending?

Legally, nothing changes. A member under a pending petition remains a member in full, entitled to sit, speak, vote and hold ministerial office, and a government is entitled to count that member in every division and in a floor test. The Manipur petitions of 2017 remained pending for about three years while the member they concerned sat as a minister in the government his crossing had helped to form. A government’s exposure during pendency is therefore political rather than legal: if the petitions are eventually allowed, the seats fall vacant and by-elections follow, and the arithmetic is recalculated then. Until a decision is recorded, the House as it stands is the House the Constitution recognises.

Q: Who can file a defection petition, and in what form?

A defection proceeding begins with a petition to the Chairman or the Speaker, ordinarily moved by a member of the House, in the form the rules of that House prescribe. Under the Members of Lok Sabha (Disqualification on Ground of Defection) Rules, 1985, whose pattern the state rules follow, the petition must set out the grounds and the material relied on: the member’s original party and ticket, the whip and its communication where paragraph 2(1)(b) is invoked, or the conduct from which a voluntary giving up of membership is to be inferred under paragraph 2(1)(a). If the petition complies with the rules, copies are forwarded to the member and, where the petitioner is not the leader of the member’s legislature party, to that leader, and written comments are due within seven days, extendable for sufficient cause.

Q: Can a court intervene in a defection case before the Speaker decides?

A court will not ordinarily intervene on the merits while the proceeding is before the chair: the quia timet reasoning of Kihoto Hollohan bars attempts to injunct or pre-empt the tribunal’s decision. Intervention is available against the delay itself. Since Rajendra Singh Rana, a failure to decide is reviewable as a violation of constitutional duty, and courts direct the chair to decide, in the later cases by a date: four weeks in the Manipur case, deadlines in the Maharashtra proceedings in October 2023, and three months in the Telangana judgment of July 2025. What courts will almost never do is decide the disqualification themselves, the Constitution Bench in Subhash Desai and the Telangana bench alike remitting the petitions to the Speaker.

Q: What did the Supreme Court direct in the Manipur defection case?

In Keisham Meghachandra Singh, decided on 21 January 2020, the Court dealt with petitions filed in 2017 against a Manipur member who had crossed from the Congress to the Bharatiya Janata Party within days of the election results and had become a minister, and which the Speaker had left undecided for about three years. The Court held that the Speaker must decide within a reasonable period, fixed three months from filing as the outer limit in the ordinary case, and directed a decision in that case within four weeks. When the direction was not met, the Court on 18 March 2020 restrained the member from entering the Assembly and recorded that he had ceased to be a minister, and the Speaker disqualified him on 28 March 2020.

Q: Are votes cast by a member later held disqualified counted as valid?

Yes, the votes and the proceedings they formed part of stand. Article 100(2) for Parliament and Article 189(2) for the state legislatures provide that proceedings are valid notwithstanding that it is afterwards discovered that a person who was not entitled to do so sat or voted, and Articles 122 and 212 bar courts from questioning the validity of proceedings for irregularity. A disqualification relates back to the conduct for the purposes of the member’s own status, but it does not reopen divisions, unpass budgets or annul a confidence vote carried in the interval. This is why the timing of the decision matters more than its content: a remedy that cannot reach backwards must arrive while the configuration it challenges still exists.

Q: Do defection petitions survive the dissolution of the House?

The petition lapses with the House. The proceeding lies before the presiding officer of the House whose member is complained against, and dissolution ends both the membership in question and the forum, so an undecided petition is not carried over to the successor Assembly. Petitions from the Manipur term that the Speaker had still not decided lapsed when that Assembly was dissolved ahead of the 2022 elections, notwithstanding the standard Keisham Meghachandra Singh had announced two years earlier. A party that wishes to pursue the matter in the new House must begin again with fresh petitions, and members returned at the election hold their seats under the new mandate, not under the shadow of the old file.

Q: Can a member disqualified for defection contest the by-election for the same seat?

Yes. In Shrimanth Balasaheb Patil, decided on 13 November 2019, the Supreme Court upheld the disqualification of the seventeen Karnataka members but struck down the part of the Speaker’s order that barred them from contesting elections for the remainder of the term, holding that the Speaker has no power to specify a period of disqualification or to bar a disqualified member from contesting, and that nothing in Articles 191, 164(1B) or 361B supports such a bar. Disqualification vacates the seat and bars ministerial office and remunerative political posts until re-election or the expiry of the term, but the member may seek the seat again at the by-election, which Section 151A of the Representation of the People Act, 1951 requires within six months of the vacancy, subject to exceptions.

Q: What must the Speaker check before accepting a member’s resignation?

Only whether the resignation is voluntary and genuine. Article 190(3)(b), for state legislatures, and the corresponding Article 101(3)(b) for Parliament, carry a proviso under which the presiding officer shall not accept a resignation if satisfied, after such inquiry as the officer thinks fit, that it is not voluntary or genuine. Shrimanth Balasaheb Patil confined the inquiry to those two qualities: the chair may test whether the letter is in fact the member’s act and was intended, but may not inquire into the member’s reasons, bargain with the member over them, or refuse the resignation because the motive is a defection. The inquiry has no fixed period, and in Karnataka in 2019 its pace itself became a subject of litigation.

Q: What did the Maharashtra Speaker finally decide in January 2024?

Speaker Rahul Narwekar decided the Shiv Sena petitions on 10 January 2024, some eighteen months after they were filed. He held that the faction led by Eknath Shinde was the real Shiv Sena when rival factions emerged on 21 June 2022, a conclusion he reached from the party constitution of 1999 and the leadership structure on the Election Commission’s record, rejecting the structure of 2018 relied on by the other faction. It followed that Bharat Gogawale was the validly appointed chief whip and that Sunil Prabhu had ceased to be the party’s whip from 21 June 2022. On those findings the Speaker dismissed the disqualification petitions filed by both factions, and no member of either side was disqualified.

Q: Can the Election Commission and the Speaker proceed on the same split at the same time?

Yes. Subhash Desai held that the two proceedings answer different questions and may run concurrently: the Speaker decides whether named members have incurred disqualification for past conduct under the Tenth Schedule, while the Election Commission decides, under paragraph 15 of the Election Symbols (Reservation and Allotment) Order, which faction is the party for the purposes of its name and symbol. Neither authority must await the other, and the Commission’s decision operates prospectively, so it does not decide or disturb the disqualification petitions. Maharashtra ran the two in parallel: the Commission recognised the breakaway faction as the Shiv Sena in February 2023, and the Speaker concluded the petitions in January 2024.

Q: Does a pending defection petition stop a floor test from being held?

No. Pendency suspends nothing: the members under petition remain members, the House transacts its business with them, and a floor test is conducted on the House as it stands. In Maharashtra the Governor’s floor test of 30 June 2022 was called while the petitions of that month were pending, the Supreme Court refused to stay it, and the Chief Minister resigned the day before it was to be held. In Madhya Pradesh in March 2020 the Supreme Court itself directed a floor test for the following day amid a resignation and defection crisis. Subhash Desai later held that a Governor needs objective material to call a floor test and cannot use one to resolve an intra-party dispute, but the pendency of petitions is not itself a bar to the test.

Q: Has any presiding officer decided a defection petition within days?

Yes. In Himachal Pradesh, six Congress members voted for the Bharatiya Janata Party candidate in the Rajya Sabha election on 27 February 2024 and absented themselves from the Finance Bill proceedings on 28 February 2024 despite the party whip. Disqualification Petition Nos. 112 to 117 of 2024 were filed by the parliamentary affairs minister, and Speaker Kuldeep Singh Pathania decided them on 29 February 2024, in a reasoned order of thirty pages, holding the six disqualified under paragraph 2(1)(a) for voluntarily giving up membership by their conduct, with immediate effect. It was the first disqualification in the State’s history. The Rajya Sabha Chairman’s disposal of the petitions against Sharad Yadav and Ali Anwar in about three months in 2017 offers a second, slower instance of expedition being administratively possible.

Q: Which committees and commissions have proposed changing who decides defection cases?

Four institutional voices across fourteen years proposed the same change: that disqualification for defection be decided by the President or the Governor on the opinion of the Election Commission, as Articles 103 and 192 provide for other disqualifications. The Dinesh Goswami Committee on Electoral Reforms recommended it in 1990; the Law Commission’s 170th Report recommended it in 1999, together with deleting the split and merger exemptions; the National Commission to Review the Working of the Constitution recommended it in 2002; and the Election Commission’s own proposed electoral reforms endorsed it in 2004. None of the recommendations has been enacted, and the Supreme Court added a different suggestion in Keisham Meghachandra Singh in 2020, proposing a permanent independent tribunal in the Speaker’s place.