majority opinion and I concur in the result reached with respect to part II. I would arrive at that result, however, by a different route.
Contrary to the view of the majority, I am persuaded that the Connecticut constitution includes a governmental obligation to provide a minimal safety net to our poorest residents. As all of our historical sources indicate, the framers of our constitution believed that the government that they were establishing would not be permitted to stand idle while people, without food, shelter, clothing or medical care, were left to die in the streets. In addition, an impressive array of contemporary sociological, economic and legal texts support the recognition of such a governmental duty.
The existence of a minimal government obligation has always been so framed, however, as to afford elected officials substantial discretion both to choose the manner in which the obligation will be implemented and to impose reasonable conditions on the provision *617of support. In light of these limitations on the constitutional obligation to provide support, the plaintiffs were not entitled to injunctive relief without making at least a preliminary showing that General Statutes (Rev. to 1993) § 17-273b, as applied to them, exceeded the large degree of discretion retained by the state. In my view, the plaintiffs in this case did not meet this demanding burden of proof. I, therefore, concur in the judgment that the trial court’s denial of the plaintiffs’ motion for a preliminary injunction must be affirmed.
I
The Connecticut constitution concededly contains no express provision obligating government to provide minimal subsistence. In my view, however, our constitution includes unenumerated constitutional rights that encompass a constitutional obligation on government to act. My conclusion depends upon the proper resolution of two distinct issues: (1) insofar as the state constitution includes unenumerated constitutional rights, must these unenumerated rights be confined to rights that arise under the umbrella of the due process clause? and (2) must unenumerated constitutional rights be categorically limited to “negative” rights—i.e., rights to be free from governmental intrusion—or may they also encompass affirmative rights—i.e., rights to demand governmental protection?
A
Our case law and our state constitutional history, text and theory already demonstrate that the constitution protects unenumerated rights. We have no express state constitutional provisions to afford protection against “cruel and unusual punishment” or against “double jeopardy.” Nevertheless, we have held that both rights are constitutionally protected. State v. Ross, 230 Conn. 183, 246, 646 A.2d 1318 (1994); Kohlfuss v. *618Warden, 149 Conn. 692, 695, 183 A.2d 626, cert. denied, 371 U.S. 928, 83 S. Ct. 298, 9 L. Ed. 2d 235 (1962).
Recently, we have characterized both of these unenumerated constitutional rights as elements of the due process guaranteed by article first, §§ 8 and 9, of the state constitution, a guarantee that the plaintiffs here do not invoke. That characterization, however, is a historical oversimplification. We never have held (and could not have held) that the framers of our written constitution enumerated a right to “due process” with the intent also of protecting unenumerated rights such as the rights against double jeopardy and cruel and unusual punishment. It is only in the second half of the twentieth century that either of the latter concepts has been denominated a right of due process. See E. Peters, “Common Law Antecedents of Constitutional Law in Connecticut,” 53 Albany L. Rev. 259, 261 (1989); see also Benton v. Maryland, 395 U.S. 784, 794-96, 89 S. Ct. 2056, 23 L. Ed. 2d 707 (1969) (incorporating fifth amendment prohibition of double jeopardy into fourteenth amendment right to due process); Robinson v. California, 370 U.S. 660, 666, 82 S. Ct. 1417, 8 L. Ed. 2d 758 (1962) (same for eighth amendment prohibition of cruel and unusual punishment); Kohlfuss v. Warden, supra, 149 Conn. 695. The cases recognizing state constitutional rights against double jeopardy and cruel and unusual punishment, therefore, cannot be read for the proposition that unenumerated rights will be recognized only when they are directly attached to the due process clause or some other particular constitutional phrase.
Instead, the case law in this state has subscribed to a broader proposition—that the framers did not intend our written state constitution to be an all-encompassing exhaustive enumeration of individual rights. Although *619we did not expressly recognize this proposition in the so-called “due process” cases, we repeatedly have done so in the past.
In 1861, for example, in deciding whether a retroactive act infringed the defendant’s rights, we explicitly held that the enumeration of particular rights in our written constitution did not preclude the existence of other quasi-constitutional rights: “Nor can it be claimed that the act in question conflicts with any provision of the constitution of this state. There is nothing in any of the provisions of that constitution which can restrain the legislature from passing retrospective laws; and it is their practice every year to do so, and not unfre-quently acts which affect antecedent vested rights. But the power of the legislature in this respect is not unlimited. They can not entirely disregard the fundamental principles of the social compact. Those principles underlie all legislation, irrespective of constitutional restraints, and if the act in question is a clear violation of them, it is our duty to hold it abortive and void.” Welch v. Wadsworth, 30 Conn. 149, 155(1861).
We again adhered to this position in 1895: “[Ujnlike the constitutions of many States, [our constitution] contains no specific limitations on the exercise of legislative power, except some slight restrictions in one or two recent amendments. The limitations, however, are no less real, and perhaps more effective, than if phrased in specific terms. ... No legislative Act is law, that clearly and certainly is obnoxious to the principle of equality in rights thus solemnly made the condition of all exercise of legislative power. It is patent that not everything that can be called a right is included in this guaranty. The protected rights are those that inhere in ‘the great and essential principles of liberty and free government’ recognized in the course of events that resulted in our independence, and established by the adoption of our Constitution. The language used is pur*620posely broad, as the language in reference to the absolute power of legislation is broad; and the relation of limitation to power can, in the nature of things, be settled only through specific applications as emergencies arise.” State v. Conlon, 65 Conn. 478, 489, 33 A. 519 (1895).
This court’s longstanding recognition of unenumer-ated constitutional rights reflects this state’s unique constitutional history. Of the thirteen original states, Connecticut was the second to last to adopt a written state constitution, failing to do so until 1818. The historical record suggests that, when Connecticut joined the Union as a state, Connecticut citizens did not see an immediate need to draft a written state constitution, because Connecticut already had a consensually adopted framework for how the government would be “constituted,” and that framework already had been in effect for more than one hundred fifty years. That framework—a “constitution” in the older, broader sense of the word—was based on principles contained in various written documents of the Connecticut colony such as the Fundamental Orders of 1639, the Ludlow Code of 1650 and the Charter of Charles II of 1662, as well as on unwritten principles of natural law. See generally 1 Z. Swift, A System of the Laws of the State of Connecticut (1795) pp. 55-59;1 H. Cohn, “Connecticut Constitutional History 1636-1776,” in H. Cohn & W. Horton, Connecticut’s Four Constitutions *621(1988) (Four Constitutions) pp. 1-15; W. Horton, “Connecticut Constitutional History 1776-1988,” in Four Constitutions, supra, pp. 18-19; C. Collier, “The Connecticut Declaration of Rights Before the Constitution of 1818: A Victim of Revolutionary Redefinition,” 15 Conn. L. Rev. 87, 94 (1982); E. Peters, supra, 53 Albany L. Rev. 259. Our early judges relied on these “constitutional” principles to decide legal disputes and to provide a remedy for misconduct by the executive branch of government. See E. Peters, supra, 261-64; see also C. Collier, supra, 15 Conn. L. Rev. 94 (“to Connecticut jurists, common law meant more than judicial precedent and case law; it included natural law as well”).
Approximately thirty years after Connecticut attained statehood, its citizens finally decided that it was necessary to adopt a written state constitution, and a convention for that purpose was assembled. The historical record suggests that there were three main reasons that the constitutional convention was thought necessary: (1) to disestablish the Congregational church; (2) to create a tripartite government involving the separation of powers and a system of checks and balances;2 and (3) to expand the suffrage to all white males who paid taxes or had served in the militia.3 See generally *622W. Horton, supra, Four Constitutions, pp. 19-23; W. Horton, “Annotated Debates of the 1818 Convention,” 65 Conn. B.J. SI-3, SI-4 (1991) (Constitutional Debates). The constitution that was adopted accomplished each of these goals, and it further placed into one written document many of the ideals that had been reflected in the seventeenth century texts or in the common law.
Nothing in either the 1818 constitution itself nor the history surrounding its adoption, however, suggests that it was intended to supersede or displace other aspects of the preexisting “constitution” of the state that, by 1818, were thought to be inherent in the structure of Connecticut society and government.4 Indeed, the text of the constitution in several places strongly suggests that the framers desired to retain and merely to improve upon the preexisting legal framework. For example, the preamble to the constitution provides that “[t]he People of Connecticut acknowledging with grati*623
Our state’s constitutional text and unique constitutional history are not, however, the only reasons that we have been, and should continue to be, open to recognizing unenumerated rights. Such openness also is demanded by the theory of constitutionalism itself. “[A] constitution intended to endure for ages to come . . . [must] be adapted to the various crises of human affairs.” McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 415, 4 L. Ed. 579 (1819), quoted in Cologne v. West-farms Associates, 192 Conn. 48, 80, 469 A.2d 1201 (1984) (Peters, J., dissenting). “Although state constitutions are more readily amended than is the federal constitution, state judges bear an independent responsibility for making state constitutions adaptable to current conditions.” E. Peters, “State Constitutional Law: Federalism in the Common Law Tradition,” 84 Mich. L. Rev. 583, 586 (1986). Without such adaptations, cultural and technological changes will disable courts from discharging their constitutional responsibility for protecting the democratic framework that the constitution establishes. See Cologne v. Westfarms Associates, supra, 70 (Peters, J., dissenting) (discussing need to recognize free speech protections in modern mall, which is modern analogue to eighteenth century public green and appropriate place for public discourse).
*624B
The state constitution, furthermore, affords no reasonable basis for limiting the recognition of unenumer-ated constitutional rights to “negative rights.”5 When the framers of the 1818 constitution expressly enumerated rights in the written constitution, they provided both negative and affirmative rights. See, e.g., Conn. Const. (1818) art. I (Bill of Rights), and art. VIII, § 2 (now art. VIII, § 4) (“[t]he fund, called the SCHOOL FUND, shall remain a perpetual fund, the interest of which shall be inviolably appropriated to the support and encouragement of the public, or common schools throughout the state, and for the equal benefit of all the people thereof”); Horton v. Meskill, 172 Conn. 615, 646-48, 376 A.2d 359 (1977) (relying on preconstitu-tional history of support of public schools and article VIII, § 4, to declare that constitution creates fundamental right to education). The preamble to the constitution also speaks of both types of rights. See Conn. Const., preamble (expressing framers’ desire “more effectually to define, secure, and perpetuate the liberties, rights and privileges which they have derived from their ancestors” [emphasis added]). The social compact that was serving as Connecticut’s “constitution” before 1818 was understood to provide certain affirmative rights. "See infra and part II B of this opinion. Finally, the history of our constitution’s adoption does not demonstrate any intent to preserve unenumer-*625ated negative rights but to abrogate unenumerated affirmative rights.6
Even more importantly, the constitutional theory that supports the recognition of unenumerated rights applies equally whether those rights are affirmative or negative rights. The drafters of the constitution intended to create a constitutional framework that would protect democratic values whatever the future circumstances. See Cologne v. Westfarms Associates, supra, 192 Conn. 79-80 {Peters, J., dissenting). That original intention should not be frustrated by an unwarranted distinction between affirmative and negative unenumerated rights.
The history of the 1965 constitutional convention, during which article eighth, § 1, was added to our constitution, does not detract from the conclusion that the constitution continues to protect unenumerated affirmative rights. As the majority correctly notes, several delegates to that convention argued that the constitution should be amended so that it would provide an express right to public elementary and secondary edu*626cation. See 3 Proceedings of 1965 Connecticut Constitutional Convention pp. 1039-68. The statements of those delegates, however, do not demonstrate that the delegates believed that rights not expressly enumerated were without constitutional protection. Nearly all of the delegates who discussed the right to education spoke of the right as if it already existed; many expressed surprise that the right was not already enumerated; and the sole question was whether a universally agreed upon principle should be expressly enumerated in the new constitution. See id. The statements therefore show only that, having had the issue of education brought to their attention, and in light of the state of constitutional jurisprudence in 1965, the delegates thought it prudent expressly to enumerate a previously unenumerated right to guarantee its recognition. There is no evidence that the delegates believed that the right to education should be the only previously unenumerated right incorporated into the 1965 constitution; nor is there evidence that the delegates even examined the historical record relating to the existence of other such rights. The 1965 reenactment of the 1818 constitution, with only minor changes, therefore, must be read to preserve the earlier framers’ understanding that the constitution was not an all-inclusive document.7
In determining what claims rise to the level of unenumerated constitutional rights and assessing the scope of such rights, we must exercise judicial constraint. Because we cannot rely on a specific text in recognizing unenumerated rights, we must take care *627not to extend such rights beyond the contours that the historical record supports. Our exercise of constraint does not, however, warrant a refusal to recognize the existence of unenumerated affirmative constitutional rights.
II
The conclusion that our constitution embodies affirmative as well as negative unenumerated constitutional rights does not itself resolve the principal issue in this appeal. The majority concludes that, even if there are unenumerated affirmative obligations inherent in our state constitution, there is no basis for a constitutional obligation to support poor people under any circumstances. I disagree.
The plaintiffs have offered considerable historical evidence that, before the promulgation of the Connecticut constitution of 1818, the governmental authorities in this state recognized their obligation to provide minimal subsistence to those who were in dire need of help. The majority does not dispute the accuracy of the historical picture painted thereby but declares that this history has no constitutional significance, because it finds the historical record relating to an obligation for support of the poor to be "ambiguous and contradictory.” In particular, the majority argues that the historical sources are ambiguous and contradictory as to whether the government supported the poor out of a sense of constitutional duty, charity or punitiveness. The majority further argues that the historical sources are ambiguous and contradictory as to the level of support that the government was obligated to provide, because some of the sources do not specify a particular level of support and others suggest that the government had to provide a level of support above the minimum for which the plaintiffs contend. These arguments are unpersuasive.
*628A
The test that we apply to interpret our state constitution requires us to consult, inter alia, history. See State v. Geisler, 222 Conn. 672, 685, 610 A.2d 1225 (1992). In rejecting the historical sources because the legislators’ motivations and the scope of the alleged right were “ambiguous and contradictory,” the majority makes that test meaningless, for two reasons.
First, history of legislative motivations is always ambiguous and contradictory. Even with the most complete sources, we never will be able to say for certain that persons framing our constitution, drafting our statutes or interpreting our common law more than two hundred years ago had a unitary and specific intent. The purpose of the historical inquiry under Geisler, therefore, is not to determine to an absolute certainty what motivated people to take particular actions, but rather to make a prediction about what most likely motivated people to take those actions. Ambiguity is the beginning, not the end, of that inquiry.
Second, the mere fact that the historical record is ambiguous as to the scope of a constitutional obligation tells us nothing about the existence of such an obligation, because the scope of every constitutional principle is ambiguous by design. All constitutions— whether enumerated or unenumerated—attempt to set out only “ambiguous” governing principles, so that those principles are general enough to be adapted to fit the very different facts and circumstances that may exist in the future. The scope of constitutional principles, in other words, can never be predetermined, but always must be left for future determination on a case-by-case basis. See, e.g., Nielsen v. Kezer, 232 Conn. 65, 74-75, 652 A.2d 1013 (1995) (scope of principle of separation of powers); Woodcock v. Journal Publishing Co., *629230 Conn. 525, 536-37, 646 A.2d 92 (1994) (scope of defenses to libel action based on first amendment); State v. Cosgrove, 181 Conn. 562, 583-84, 436 A.2d 33 (1980) (scope of rights afforded by confrontation clause); State v. Cordon, supra, 65 Conn. 489 (calling for case-by-case analysis of boundaries between legislative actions and unenumerated rights).
It is true that a well established jurisprudential doctrine counsels us to construe ambiguous constitutional principles narrowly. We rely on that doctrine when, in deciding constitutional challenges to particular government conduct, we demand proof that the conduct violates the constitution “beyond a reasonable doubt.” See, e.g., State v. Ross, supra, 230 Conn. 236; State v. Joyner, 225 Conn. 450, 460, 625 A.2d 791 (1993); State ex rel. Andrew v. Lewis, 51 Conn. 113, 127-28 (1883).
That jurisprudential doctrine, however, does not and cannot apply to our initial decision of whether a constitutional principle exists in the abstract—i.e., without reference to a particular government action. In resolving the initial question of whether any constitutional principle exists, we cannot require proof of its existence “beyond a reasonable doubt”; historiography tells us that such proof is impossible. We also cannot demand an unambiguous definition of the scope of the constitutional principle; constitutional theory tells us that such a definition can be constructed only on a case-by-case basis.
It is only after we have determined that a constitutional principle exists that we must scrutinize it carefully, to determine whether it bars the particular government action being challenged. See, e.g., part IV of this opinion; see also Horton v. Meskill, supra, 172 Conn. 648-50 (first determining that constitution provides right to education; then determining whether legislative fund*630ing mechanism violates constitution beyond a reasonable doubt). If we demand unambiguous history before deciding that rights exist, our search for such rights necessarily will become a pro forma exercise that leads to a foregone conclusion.8
B
Despite its intrinsic ambiguities, the historical record persuades me that in Connecticut, before 1818, it was understood that the government had an inherent duty to ensure that no one who complied with reasonable conditions would be left to starve or freeze in the streets. That duty is sufficiently embedded in our history to warrant its constitutional recognition.
The existence of the constitutional obligation to provide minimal subsistence is demonstrated most clearly by the writings of our two most preeminent eighteenth century jurists, Chief Justice Swift and Judge Root.9 Both Swift and Root expressly stated that the government had an “obligation” or a “duty” to help the poor. For example, Swift wrote that “[t]he selectmen are bound to provide necessaries for all the inhabitants of the town, who are incapable of supporting themselves. Towns are obliged to support their respective inhabitants, whether living in the town to which they belong, or any other town, either with or without a certificate, who may need relief . . . .” 1 Z. Swift, supra, p. 119. Root further explained that this govern*631mental obligation was prestatutory and that the statutes merely implemented the obligation. As Root wrote: “[T]he highest character given of any ruler on earth is, that he judgeth the people in his righteousness, and the poor with judgment; that he delivereth the needy when they cry, and the poor that hath no helper; that he dealeth bread to the hungry, and delivereth him that is ready to perish. It is the duty of every government to protect and to provide for the poor; the laws of the state therefore humanely enact and ordain . . . that every town shall take care of, provide for and maintain, its own poor.” (Emphasis added.) J. Root, 1 Root’s Reports, pp. xxviii (1789-93).
The governmental obligation also is reflected—either implicitly or explicitly—in every codification of the statutes of Connecticut, beginning with the Ludlow Code of 1650, and continuing to the present day—an unbroken line stretching for nearly 350 years. See footnotes 58 through 64 of the majority opinion (reprinting statutes). Contrary to the majority’s suggestion, moreover, these statutes do contain “evidence that there was an inherent obligation on the part of the towns, other than that mandated by the colonial government, to care for the indigent.” An obligation to support the poor was not expressly imposed by statute until 1672. See footnote 59 of the majority opinion (reprinting statute of 1672). Nevertheless, the Ludlow Code reflects that, even before 1650, there were disputes between the towns about where indigent persons had been “lawfully settled,” and it was necessary to establish a procedure by which the general court could decide these disputes. See footnote 58 of the majority opinion (reprinting Ludlow Code). Such disputes would only have surfaced, of course, if the towns already were operating under the assumption that they were obligated to support poor people who had been settled in their towns, but not poor people from other towns. The *632Ludlow Code is thus consistent with the interpretation of Judge Root that the duty to support the poor preceded the enactment of statutes explicitly requiring such support.10
The existence of some kind of governmental obligation is entirely consistent with the observation of the majority that “in the entire period before the constitution of 1818, local elected officials had wide discretion to decide the nature of, amount of and eligibility criteria for receiving” governmental assistance. During the relevant time period, elected officials even had the power to impose various harsh conditions on the provision of support. Neither type of discretion, however, undermines the premise that the government retained some obligation to support the poor.
Beginning in 1702, the laws and statutes explicitly gave elected officials discretion to decide who was eligible for support, how much support was necessary, and the form of such support. See footnotes 60 through 63 of the majority opinion (reprinting statutes). Nevertheless, the laws and statutes also provided in unambiguous terms that there was a baseline obligation to maintain the poor at a level of minimum subsistence. For example, a 1702 law provided that “each Town within this Colony shall maintain their own poor; and the Select-men or Overseers of the Poor (where any such are Chosen) shall at times keep the Town Stock; *633who shall have full power to disburse or expend what they shall judge meet from time to time for the relief and supply of any of the poor belonging to their Town, so far as Five Pounds will extend; and if more be needful, the said Select-men or Overseers, or the major part of them, shall with the advice of the Assistants (if there be any in the Town) disburse what shall be by them judged needful for the relief of the poor as aforesaid . . . for the supplying their Poor, or any of them with Victuals, Cloathing, Fire-wood, or any other thing necessary for their support or subsistance.” Acts and Laws of His Majesties Colony in Connecticut in New England (B. Green & J. Allen pubs. 1702) p. 94; see footnote 60 of the majority opinion. Except for changes in the applicable dollar amounts, the statute of 1808, which was in effect when the constitution was enacted, was to the same effect. General Statutes (1808 Rev.) tit. CXXX; see pp. 606-607 of the majority opinion.
Throughout the eighteenth century, the statutes also afforded the towns wide authority to impose conditions on support. For example, a town that “was bound by law to support a person, had a right to restrain his locomotion, and place him where his support would be most cheap and convenient.” Backus v. Dudley, 3 Conn. 568, 573 (1821). In particular, a town could place its employable poor into workhouses; it could require them to work in exchange for food; and it could impose corporal punishment on them if they did not work. A town even was permitted to separate families, binding out the children of the poor as apprentices or servants. General Statutes (1808 Rev.) tit. CXXX, § 5, and tit. CLXXVI, § 21; see pp. 607 and 608, respectively, of the majority opinion.
The towns’ historical discretion to impose conditions on the provision of support demonstrates that the towns’ obligation was not unlimited, but it does not demonstrate that the obligation did not exist. The stat*634utes explicitly recognized the underlying obligation of the towns to “maintain their own poor; and ... [to supply] their Poor, or any of them with Victuals, Cloath-ing, Fire-wood, or any other thing necessary for their support or subsistance.” Acts and Laws of His Majesties Colony in Connecticut in New England, supra, p. 94; see footnote 60 of the majority opinion. The majority’s construction of these statutes as having been designed merely to punish and isolate the poor cannot be reconciled with the fact that the statutes do not merely establish workhouses and provide for the binding out of children but also expressly state the underlying obligation to provide support of some kind.
The historical evidence shows, therefore, that, at the time of the adoption of the 1818 constitution, a governmental obligation to support the poor had served as the implicit or explicit premise of more than 168 years of legislation. This obligation was understood to be fundamental to the organization of our state and inherent in natural law. The obligation was in no way absolute, as elected officials had wide discretion to decide how that obligation would be implemented or to impose conditions on the provision of support. Nevertheless, if a person met the applicable conditions and criteria and otherwise would have no other means to survive, support could not be denied.11
*635III
Recognition of this limited, conditional governmental obligation is supported not only by the historical record, but also by contemporary considerations of law and public policy. The majority declines to address most of these contemporary considerations on the grounds that (1) any judicial recognition of a right to minimal support will invalidate all general assistance eligibility requirements, and (2) the choice to provide such support is a question of “morality, ‘policy and value’ ” that is reserved to the legislature. I disagree with the majority that recognition of the constitutional right asserted here will fetter legislative attempts to establish reasonable eligibility requirements. Moreover, in my view, our duty to construe the state constitution requires us to undertake the plenary analysis outlined in State v. Geisler, supra, 222 Conn. 684-85, which includes consideration of issues of morality, policy and value. Contemporary economic, sociological, legal and moral considerations overwhelmingly support the recognition of a qualified right to some limited form of governmental assistance to assure minimal support for our poor citizens.
*636A right to governmental support is even more important today than it was 350 years ago, when our forebears recognized the right. In their largely agricultural society, with a vast, unsettled western frontier, paid work presumably was available to everyone who wanted it. In today’s postindustrial society, on the other hand, almost all jobs that pay a living wage require skills or educations that many poor people neither possess nor have the economic resources to acquire. There also is no longer any opportunity to “go west,” homestead new land, and grow the food necessary for survival. See A. Amar, “Forty Acres and a Mule: A Republican Theory of Minimal Entitlements,” 13 Harv. J. L. & Pub. Policy 37, 41 (1990) (comparing today’s “safety net” to nineteenth century “safety valve” of western frontier).
Finding a governmental obligation of minimal subsistence in the constitution also is justified by our contemporary notions about democracy and universal suffrage. Although the constitution of 1818 did not so provide, our state constitution now extends the suffrage to all adults, regardless of race, gender, or economic circumstances.12 Individuals who have no food, clothing, shelter or medical care will, in all likelihood, be severely hampered in their ability to vote or otherwise to participate in the political process. The right to minimal subsistence can therefore be justified as necessary to ensure that the unsheltered poor—a truly discrete and insular minority group—have access to the political process. See generally F. Michelman, “Wel*637fare Rights in a Constitutional Democracy,” 1979 Wash. U. L.Q. 659, 676-78 (arguing that right to subsistence is one of rights properly enforceable by courts as “representation-reinforcing” under Professor John Ely’s analysis); see also J. Ely, Democracy and Distrust: A Theory of Judicial Review (1980); L. Tribe, American Constitutional Law (2d Ed. 1988) p. 778.
These contemporary economic circumstances and contemporary conceptions of democracy already have led the international community to incorporate a right to subsistence into the international law of human rights. For example, article 25 (1) of the Universal Declaration of Human Rights declares that “ [everyone has the right to a standard of living adequate for the health and well-being of himself and of his family, including food, clothing, housing and medical care and necessary social services, and the right to security in the event of unemployment, sickness, disability, widowhood, old age or other lack of livelihood in circumstances beyond his control.” Universal Declaration of Human Rights (1948), U.N. Doc. A/811, reprinted in Basic Documents on Human Rights (3d Ed. 1992) pp. 21, 26 (Basic Documents).
Article 11 (1) of the International Covenant on Economic, Social and Cultural Rights (International Covenant), which supersedes the Universal Declaration of Human Rights for the 104 states that are parties to it,13 similarly provides that “[t]he States Parties to the present Covenant recognize the right of everyone to an adequate standard of living for himself and his family, including adequate food, clothing and housing, and to the continuous improvement of living conditions. The States Parties will take appropriate steps to ensure the realization of this right . . . .” International Cov*638enant on Economic, Social and Cultural Rights, Jan. 3,1976, 993 U.N.T.S. 3, reprinted in Basic Documents, supra, pp. 114,118; see also Centre for Human Rights, United Nations, Right to Adequate Food as a Human Right (1989); The Right to Food: Guide Through Applicable International Law (K. Tomasevski ed. 1987); Symposium, “The Right to Food: The International Perspective,” 30 Howard L.J. 223 (1987); The Right to Food (P. Alston & K. Tomasevski eds. 1984).
Although the United States is not a party to the International Covenant, and although no right to subsistence may yet apply to this country as part of customary international law; see, e.g., R. Bard, “The Right to Food,” 70 Iowa L. Rev. 1279, 1289 (1985); the wide international agreement on at least the hortatory goals identified in the human rights documents strongly supports the plaintiffs’ claim. See B. Stark, “Economic Rights in the United States and International Human Rights Law: Toward an ‘Entirely New Strategy,’ ” 44 Hastings L.J. 79 (1992) (arguing that International Covenant should be read to inform United States state courts’ interpretation of state constitutions); comment, “Human Rights and Basic Needs: Using International Human Rights Norms to Inform Constitutional Interpretation,” 34 U.C.L.A. L. Rev. 1195 (1987) (same); see also Boehm v. Superior Court, 178 Cal. App. 3d 494, 502, 223 Cal. Rptr. 716 (1986) (relying on Universal Declaration of Human Rights to interpret state statutory duty to support poor); cf. Sterling v. Cupp, 290 Or. 611, 617 n.9, 622 and n.21, 625 P.2d 123 (1981) (relying on Universal Declaration of Human Rights to interpret state constitutional provision relating to treatment of prisoners).
Finally, the writings of many of this country’s leading constitutional scholars support the recognition of the right to minimal subsistence. See, e.g., A. Amar, supra, 13 Harv. J. L. & Pub. Policy 37; C. Black, “Fur*639ther Reflections on the Constitutional Justice of Livelihood,” 86 Colum. L. Rev. 1103 (1986); P. Edelman, “The Next Century of Our Constitution: Rethinking Our Duty to the Poor,” 39 Hastings L.J. 1 (1987); F. Michelman, supra, 1979 Wash. U. L.Q. 659; F. Michel-man, “In Pursuit of Constitutional Welfare Rights: One View of Rawls’ Theory of Justice, ” 121 U. Pa. L. Rev. 962 (1973); L. Tribe, supra, p. 778.
To be sure, there are scholars who dispute that such a right should be recognized. Most of their objections, however, center on problems with finding such a right in the federal constitution. They point to limitations of the federal constitutional text, to the constitutional theory that the federal government is a government of limited powers and to the absence of a history of federal support for the poor until the 1930s. See, e.g., Bowers v. DeVito, 686 F.2d 616, 618 (7th Cir. 1982) (Posner, J.) (federal constitution “is a charter of negative liberties ... it does not require the federal government or the state to provide services”); R. Bork, “Commentary—The Impossibility of Finding Welfare Rights in the Constitution,” 1979 Wash. U. L.Q. 695; H. Monaghan, “Commentary—The Constitution Goes to Harvard,” 13 Harv. C.R.-C.L. L. Rev. 117 (1978); R. Winter, “Poverty, Economic Equality, and the Equal Protection Clause,” 1972 Sup. Ct. Rev. 41.
The only scholarly objections that appear relevant to our inquiry—and indeed, the only objections that the majority expresses—are that recognizing a right to governmental support will be: (1) judicially unmanageable, as it will force judges to decide the constitutionality of every change to the welfare program and to second-guess policy judgments of the legislature; and (2) counterproductive, as it will decrease poor people’s incentives to become self-sufficient and thereby extend the cycle of poverty. See, e.g., R. Bork, supra, 1979 Wash U. L.Q. 699-701; R. Ellickson, “The Untenable Case *640for an Unconditional Right to Shelter,” 15 Harv. J. L. & Pub. Policy 17 (1992); S. Neilson, “Right to Shelter Under the Connecticut Constitution,” 67 Conn. B.J. 441 (1993).
Such objections do not foreclose the recognition of a limited constitutional right that obligates government to provide only minimal subsistence but reserves to government broad discretion about the manner in which such an obligation is to be implemented. The history of the right to subsistence teaches us that, although there is a minimal obligation to provide support, elected officials have discretion to determine who is eligible for support, to fix the level of support, to determine the means by which support is to be provided, and to impose conditions on the provision of support, such as the condition that recipients perform work or acquire other skills.14 Judicial intervention will not be warranted to enforce the constitutional obligation except in the most extreme cases—where individuals demonstrate that: (1) without government support, they actually will be unable to secure the necessaries of life such that they will face a grave threat to their health or welfare; and (2) for reasons beyond their control, they could not comply with the ’conditions the statute imposes. Judicial intervention to enforce a constitutional obligation only in such narrowly defined circumstances of severe deprivation meets all legitimate jurisprudential objections. See, e.g., R. Ellickson, supra, 15 Harv. J. L. & Pub. Policy 17; note, “Making Shelter Work: Placing Condi*641tions on an Employable Person’s Right to Shelter,” 100 Yale L.J. 491 (1990).
In sum, our constitutional framers, the contemporary academy, and the international community all support the conclusion that the government may not stand idle while its poorest residents die in the streets because of lack of food, shelter, clothing or medical care. The government has wide discretion in implementing its constitutional obligation and in imposing reasonable conditions on the provision of minimal support. The government, nonetheless, has a constitutional obligation to provide minimal subsistence.
IV
The existence of the government’s constitutional obligation to provide minimal subsistence is a necessary but not a sufficient predicate for the plaintiffs’ cause of action. I would affirm the trial court’s denial of a temporary injunction in this case because, in my view, on the facts presented, the plaintiffs have failed to prove that General Statutes (Rev. to 1993) § 17-273b was an improper exercise of the legislature’s broad discretion to implement its obligation to provide minimal subsistence.
On its face, § 17-273b is entirely consistent with historical limitations on governmental support. Like the laws and statutes of 1702 and 1808, which limited monetary support to a particular dollar amount but allowed the towns to exceed that limit if they thought it necessary; see part II B of this opinion; § 17-273b limits the duration of some benefits to nine months but allows the towns to exceed that limit if they think it necessary. Moreover, at the same time that § 17-273b was amended to impose the nine month limit on those benefits, funding was increased for other programs designed to benefit poor people. See footnote 68 of the majority opinion (describing programs).
*642Similarly, the intent behind the nine month limit is consonant with, rather than in abrogation of, the governmental obligation to care for the poor. As one legislator explained, and as witnesses at trial confirmed, the statute was designed to help poor people in the long run, by creating stronger incentives for employable people to find jobs and thus breaking the cycle of dependence on welfare. Legislators reasonably could have believed that it was better to provide full benefits for nine months, so as to encourage recipients to seek employment during the three month hiatus, rather than to provide benefits for twelve months at a lower monetary amount. See, e.g., 35 H.R. Proc., Pt. 24, 1992 Sess., pp. 8053-54, remarks of Representative Joseph D. Courtney (“[B]ut most fundamentally is the much more concerted effort which will be made to move able-bodied employable recipients of this program into work. . . . First of all, it will put limits on what the program will pay. There will be an end of the road for able-bodied recipients which the[y] must, at that point, contend with the fact that they either are going to have to do better in their efforts to find employment or they lose cash benefits.”); id., Pt. 23, p. 7626, remarks of Representative Joseph D. Courtney (“perhaps one way of finding some savings, since there were other addbacks occurring during the budget process, was to reduce the grant level for employables to give that additional incentive to go out and fin[d] work”); see also footnote 67 of the majority opinion (discussing trial testimony).
In this case, furthermore, there is no evidence of record that, because of the enactment of § 17-273b, any plaintiff has in fact been unable to secure the specific “necessaries” of life to which he or she might be constitutionally entitled as a matter of minimal subsistence. There is no evidence that, as a result of the statutory scheme at issue, any plaintiff faced or now faces a grave threat to his or her health or welfare due to the lack *643of any of these life necessities. The plaintiffs have, therefore, not demonstrated that they are entitled to relief.
I respectfully concur in affirming the judgment of the trial court.
Chief Justice Swift wrote: “Some visionary theorists, have pretended that we have no constitution, because it has not been reduced to writing, and ratified by the people. It is therefore necessary, to trace the constitution of our government to its origin, for the purpose of showing its existence, that it has been accepted and approved of by the people, and is well known and precisely bounded.” 1 Z. Swift, supra, p. 55. Swift went on to explain how Connecticut’s “constitution” descended from the voluntary agreement of the settlers in the early seventeenth century to form a government, and how that social compact had endured through both monarchical rule and independence. Id., pp. 55-59.
Before 1818, Connecticut had what is now known as a “parliamentary” system of government. The General Assembly was the final source of most power in the state. The governor was merely the presiding officer of the upper house of the General Assembly, then known as the Council. The other members of the Council, known as the assistants, exercised powers of all three branches: executive when assisting the governor, legislative in passing bills, and judicial in that the Council sat as the Supreme Court of Errors. W. Horton, supra, Four Constitutions, pp. 19-22.
Before 1818, for all practical purposes, the only people eligible for the suffrage were adult white males who owned land. Because of shifts from an agricultural to a more industrial economy, the proportion of landowners in the population decreased throughout the latter part of the eighteenth *622and early part of the nineteenth centuries, and, by 1816, fewer than one half of the adult white male residents of Connecticut were eligible to vote. W. Horton, supra, Four Constitutions, pp. 22-23.
For example, in voting for the adoption of the Bill of Rights, a majority of the delegates to the constitutional convention implicitly rejected the view of several delegates that “such a detailed specification of individual rights was superfluous and tended to abridge them, because all governmental powers not granted by the constitution were reserved to the people.” Cologne v. Westfarms Associates, 192 Conn. 48, 60, 469 A.2d 1201 (1984); see Constitutional Debates, supra, 65 Conn. B.J. SI-16-SI-24, remarks of Governor John Treadwell, Alexander Wolcott and Judge Stephen Mitchell. As this court previously has noted, the majority of delegates likely was moved by “the political sentiment of that time that the basic liberties of the people should be enshrined in a written constitution to ensure their protection from governmental infringement.” Cologne v. Westfarms Associates, supra, 60. There is no evidence, however, that any delegate believed that the enumerated rights were meant to be the only rights the constitution would protect. Indeed, one delegate who expressly had argued that the framers should “not undertake to tell the people their rights—they could not be enumerated” went on to vote for the constitution anyway. Constitutional Debates, supra, 65 Conn. B.J. SI-22, remarks of Judge Stephen Mitchell.
As the majority notes, in Cologne v. Westfarms Associates, supra, 192 Conn. 61, this court stated that the rights enumerated in article first of the state constitution only protect individuals against government action. Contrary to the majority’s implication, however, our statement in Cologne was not meant to distinguish between rights against government action and rights to government action; in that case, we merely enunciated a difference between rights against government action and rights against private action. Moreover, even if Cologne could be read to suggest that the rights enumerated in article first are purely “negative” rights, that case certainly does not decide that enumerated rights also would be so limited.
At least one constitutional convention delegate appeared to anticipate early in the convention that the provision of certain enumerated negative rights in the Bill of Rights might lead to confusion as to whether there remained unenumerated affirmative obligations. Governor John Treadwell, a federalist delegate to the convention, suggested that instead of having a Bill of Rights that specified only the rights of the people, “it might be provided what the legislature may do, what they may not do, and what they shall do; the powers of the executive and judiciary might be enlarged, and the rights of the people, to do this, should be incorporated under the several distinct heads. Such a list of distinct propositions [in the Declaration of Rights], it was improper to pass by themselves; it would be more advantageous, as well as systematical, to apportion the rights to be incorporated under their distinct heads.” (Emphasis added.) Constitutional Debates, supra, 65 Conn. B.J. SI-17. Governor Treadwell’s suggestion obviously was not adopted, and he voted against the constitution as drafted. Id., SI-7 n.10. One possible interpretation of the other delegates’ rejection of Governor Treadwell’s suggestion is that they did not believe that the structure of the constitution would create the problem that he feared.
Our decision in Horton v. Meskill, supra, 172 Conn. 615, moreover, makes clear that the 1965 constitutional amendment was not necessary to create a fundamental constitutional right to education. In that case, we recognized the existence of the right to education based only on the pre-1818 history of educational funding and the provision in the constitution of 1818 that there was to be a perpetual school fund, and not on the constitutional provisions that had been added in 1965. See id., 646-48.
The majority’s reasoning would be the equivalent of stating that, if our written constitution did not explicitly enumerate a right to “due process,” we would be forced to declare that there was no such right because some parts of the historical record did not specify exactly why such a right was being recognized, and because other parts of the historical record lent themselves to more generous interpretations of the alleged right than others.
As the majority notes, we repeatedly have treated these writings as authoritative when examining the common law antecedents of our constitutional law. See, e.g., State v. Chapman, 227 Conn. 616, 628, 632 A.2d 674 (1993); State v. Joyner, supra, 225 Conn. 467; State v. Oquendo, 223 Conn. 635, 650, 613 A.2d 1300 (1992).
The statutes reenacting the governmental duty to support the poor in 1821 and thereafter; see footnotes 63 and 64 of the majority opinion; further demonstrate that, by the time the written constitution was adopted, the support of the poor was firmly entrenched as a civil, and not a religious, duty. In 1821, the statutes were amended with the intent of implementing the constitutional disestablishment of the Congregational church and the theory of separation of church and state. See General Statutes (1821 Rev.) preface (1824 Ed.) pp. x-xi; see also Snyder v. Newtown, 147 Conn. 374, 385, 161 A.2d 770 (1960) (relying on 1821 statute to determine meaning of constitutional prohibition against compulsory support of a church).
The majority is correct that a series of our post-1818 cases contain a dictum to the effect that “[a] town is under no common-law obligation for the support of paupers. Its sole liability in this respect is imposed by statute, and it is not liable for a pauper’s support any farther than statute makes it so. . . (Citation omitted; internal quotation marks omitted.) Middle-sex Memorial Hospital v. North Haven, 206 Conn. 1, 5, 535 A.2d 1303 (1988); William W. Backus Hospital, Inc. v. Norwich, 146 Conn. 686, 689,155 A.2d 916 (1959); State v. Bristol, 139 Conn. 469, 471, 95 A.2d 78 (1953); Beacon Falls v. Seymour, 44 Conn. 210, 214 (1876).
These statements, however, have neither much weight nor much relevance. In none of the cases in which the court issued the dictum was the court even asked to decide whether the common law provided rights greater *635than those contained in the statute: in each case, the plaintiffs only argument was that the statute required the town to reimburse the plaintiff for support that had been provided. Indeed, none of these cases even reviewed the common law before enunciating the dictum. In addition, all the cases significantly postdate the drafting of our written constitution.
Even were we to assume that the court’s enunciation of the dictum grew out of a considered and reasoned interpretation of our law, none of the cases could be read to decide the issue now before us—whether, in the absence of any contemporary statute requiring support, the constitution would require the government to maintain a minimal safety net for the poor. The dicta suggest only that the common law imposes no obligation to support the poor “any further than statute” already imposed at the time the cases were decided. The dicta are thus consistent with the result I reach in this case—that the statute adequately fulfills the government’s constitutional obligation. See part IV of this opinion.
The constitution of Connecticut, article sixth, § 1, as amended by article nine of the amendments, provides: “Every citizen of the United States who has attained the age of eighteen years, who is a bona fide resident of the town in which he seeks to be admitted as an elector and who takes such oath, if any, as may be prescribed by law, shall be qualified to be an elector.”
This figure is as of 1991. B. Stark, “Economic Rights in the United States and International Human Rights Law: Toward an ‘Entirely New Strategy,’ ” 44 Hastings L.J. 79, 80-81 n.8 (1993).
Of course, there may be constitutional limitations on the types of conditions that the legislature may impose on the provisions of benefits. See, e.g., footnote 48 and p. 608 of the majority opinion (discussing constitutional rights to travel and equal protection and the constitutional prohibition of involuntary servitude); see also note, “ ‘Unconscionable' Conditions: A Contractual Analysis of Conditions on Public Assistance Benefits,” 94 Colum. L. Rev. 193, 196-220 (1994). These limitations already constrain the actions of the legislature, however, and do not depend on whether we recognize a right to minimal subsistence.