Gonzaga Journal of International Law

Gonzaga University 721 N. Cincinnati St Spokane, WA 99202 Phone 800 986 9585

Click on the icon above to download article.

Cite as:  Paul F. Diehl, Reconceptualizing Treaty Consent, 6 Gonz. J. Int’l L. (2002-03), available at http://www.gonzagajil.org/.

RECONCEPTUALIZING TREATY CONSENT

 By

Paul F. Diehl[1]

             In his classic statement, James Brierly argued that international law arises from the consent of nations.[2]  With respect to whether a state consents to a treaty, the traditional conceptualization is a binary one: the state is either a party to the treaty or it is not.  The Vienna Convention on the Law of Treaties[3] is reflective of this dichotomous classification, and the United Nations system for the deposit of ratification instruments is public affirmation of those who consent to the treaty and those who are not.  In the aggregate, the number of states that are party to a treaty is often used to gauge the breadth of acceptance of a given agreement or the norms embedded in it; much has been made of the fact that over 180 states are parties to the Nuclear Non-Proliferation Treaty (NPT)[4] while at the same time noting the number of nuclear capable states (e.g., India, Pakistan) who have not ratified the agreement. 

            There are obvious advantages to assessing treaty consent based on whether a given state or set of states are parties to the agreement.  As any young viewer of Sesame Street will recognize (“these go together and these are different”), there are strong elements of parsimony and clarity in such simple conceptualizations.  The transparent and generally formal consent involved in treaty ratification is also an advantage for policymakers and scholars alike in determining a state’s position on a given agreement. 

Despite these obvious benefits, there are some drawbacks and limitations to this simple conceptualization.  Generally, such a classification ignores the differences in legal obligations that may accrue to those within the broad consenting and non-consenting categories.  Furthermore, and perhaps more significantly, the dichotomous classification does not capture the degree of political support that a state might have for the treaty in question.  Whether a state is a party or not to a given treaty is, at best, only preliminary indicator of possible treaty acceptance and gives little indication of (1) how strongly it accepts that agreement, (2) how willing it is to comply with that agreement, and (3) how tolerant it will be of violations committed by others.  A state may be a party to a given agreement, but it may be a strong supporter of the agreement or a weak one.  It may also be more acceptant of certain provisions in the agreement than of others.  As noted below, a state that ratifies a treaty may place reservations on its acceptance such that portions of the treaty or its obligations vis-à-vis certain other parties are precluded.  Conversely, that a state is a not a party to a treaty does not necessarily indicate that the state’s government rejects the treaty in whole.  Instead, this may be a reflection that domestic political objections to some set of provisions within the agreement prevent it from being a party.  Current binary conceptualizations lose these nuances and thereby ignore the tremendous variation within the categories of parties and non-parties.

Whether a state is a party to an agreement is also not necessarily a good indicator of whether it will follow the precepts contained in it or not.  The willingness of the United States to follow the provisions of the unratified SALT II agreement[5] (at least until 1986) is illustrative of one side of the coin.  That Iraq’s human rights policies did not improve (and probably worsened) after it accepted the International Covenant on Civil and Political Rights in 1971 demonstrates the reverse. [6]  Oona Hathaway actually finds a negative association between treaty ratification and observance of some human rights practices.[7]  In general, one cannot always wait for a state to comply or not comply with an agreement to assess its support of a treaty.  Looking only at actual compliance behavior makes prediction impossible and such a post hoc approach may be too late for policymakers seeking an assessment of state policy. 

In other cases, some states have little “opportunity” to violate an agreement or not.  For example, many parties to the aforementioned NPT lack the technological capability to build nuclear weapons.  Other states are unlikely to exceed pollution or fishing limits when they engage in few activities of the sort covered by agreements to which they are parties.  In those cases, support for an agreement is largely symbolic and their support is instead important vis-à-vis their willingness to uphold the principles in the agreement and participate in sanctioning others who violate such principles.

At the international system level, international “regimes” or global governance structures are composed of rules, norms, and institutions, many of which are formalized in treaties. [8]  Successful operation of regimes depend not merely on how many states are members, but also on the degree to which they (the regimes) are accepted by the international community.  Merely counting the number of states party to the agreements governing the regime will not necessarily reveal the degree of commitment present for the regime, and perhaps provides a misleading picture.  Indeed, W. Michael Reisman argues that even unperfected legal acts, such as unratified treaties, can influence the expectations and behavior of states.[9]  A more sophisticated view of how broadly and to what degree international rules are accepted is a prerequisite to addressing many of the questions about international law and international relations posed by Slaughter, Tulumello, and Wood.[10]

If the traditional mode of classifying treaty consent is inadequate, what conceptualization might take its place?  To answer this, I turn to the concept of “fuzzy sets,” developed in the field of mathematics.[11]  Fuzzy sets begin with the notion that objects more or less correspond to a category; that is, membership in a category is a matter of degree rather than an absolute.  For example, rather than classifying a voter as a Democrat or Republican, a voter could be regarded as 75% Republican and 25% Democratic reflecting her particular mixture of split-ticket voting.  Applied to our concern, we should look at the degree to which a state consents to a treaty, rather than assume that all states that ratify agreements are equally acceptant of that agreement.  Similarly, there is variation within the broad category of non-parties, with some states exhibiting various degrees of consent. 

There is little doubt that the ratification of a treaty by a state is a significant indicator of consent by that state; indeed, this is the traditional standard and I do not wish to eliminate this aspect.  There are, however, other points on the continuum between ratification and non-ratification.  With the idea that states can vary in their consent to an agreement (the degree of fuzziness, if you will), one looks to the factors that define its degree of consent.  In the following three sections, I offer a series of considerations designed to suggest possible differences across states.

Methods and Breadth of Consent: A First Cut

            One place to begin the reconceptualization of treaty acceptance is with other forms of acceptance that engender legal rights and obligations.  Accession is the process by which a state subsequently binds itself to a treaty, after a negotiation and ratification process in which it did not participate.  For example, a number of states (e.g., Denmark, Canada) joined the Antarctic Treaty[12] by accession, even though they were not original members.  This type of treaty adherence is permitted so as to promote broader principles, and in theory, universalism within international law; it also limits future negotiation costs with respect to including new parties in the agreement. 

Legally, those states that accede to a treaty have the same rights and obligations as those who were part of the original negotiating process and then ratified the agreement.  It is not immediately obvious that acceding states are less supportive of a treaty than ratifying states.  Indeed, as in the case of China joining the World Trade Organization (WTO), some acceding states pay significant costs to become part of a treaty regime.[13]  Yet, it is conceivable that accession may be suggestive of less consent than original treaty ratification.  Much depends on the reasons for a state not being an original participant.  Of course, some states might not have been in existence at the time of the original treaty, and therefore their accession is merely a function of not having the legal capacity to commit to the agreement rather than some indication of reluctance to participate.  Thus, East Timor, newly independent in 2002, will undoubtedly accede to many extant treaties.  Some states may also have been excluded from the treaty making process, for any number of political or economic reasons.  For example, some recent additions to the NATO membership (e.g., Czech Republic, Hungary) were states that were considered enemy states from the time of NATO’s founding and through most its history.  Exclusion has become less common over time as international organizations and multilateral forums are now the preferred mechanisms for negotiating significant international agreements.[14]  Yet states may not have participated in the original treaty formation process because they objected to the goals of the exercise or opposed some of the likely treaty provisions.  In those circumstances, the state clearly did not originally support the treaty.  A subsequent “change of heart” in becoming party to a treaty may indicate that such provisions are now minimally acceptable, but it may still object to parts of the agreement and may only choose to become a party in order to enjoy certain benefits. 

As suggested above, some states that accede to a treaty may be less acceptant of an agreement because they are not as politically vested in the agreement as original parties.  Acceding states did not have to pay the negotiation costs or make the political commitments at the time to ensure the success of the negotiation process.  Furthermore, they did not have direct input into the formulation process, and indeed may oppose some parts of the agreement.  Certain states, especially less developed ones, may join a treaty regime out of feelings of necessity (e.g. trade regime) rather than because they strongly support the values underlying it.  Thus, under some circumstances, treaty accession may include less consent than ratification.  The extent of this difference (if any) and its impact on compliance are ultimately empirical questions for future study.

Lower on the consent scale are states that sign an agreement, but do not ratify it.  For example, as of October 2002, Cyprus has not ratified the so-called Mine Ban Treaty despite being among the first signatories.[15]  These states are often classified with those that never participated in the process, never signed, and never ratified the agreement.  Legally, however, states that sign an agreement do have some legal obligations.  Most notably, they must refrain from any actions that undermine the object and purpose of the treaty; the length of this obligation into the future is unclear, although it expires if the state explicitly rejects ratification or the intent to ratify.  For example, US President George W. Bush withdrew the signature of his predecessor to the treaty creating the International Criminal Court[16] and announced that he has no intention of seeking US Senate approval for that agreement.[17]  That a state was willing to sign an agreement suggests that its executive or head of state supported the agreement.  This is quite different from a state that never takes the first step of affixing its signature to the document.  Furthermore, in the absence of a formal rejection of the treaty, a signatory could in short order become a party to an agreement with ratification; this may takes years, however, as the US experience with the Genocide Treaty indicates.[18] 

Why does a state sign, but not ratify, an agreement?  There may be dozens of reasons, but one of the most prominent is a change in government within the signatory state such that its preferences vis-à-vis the agreement also changes.  Another possibility is that the ratification process involves some legislative body that opposes the executive body that signed the agreement (different ideologies and/or political parties may be to blame).  Either circumstance suggests that at least some strong domestic actors support the agreement.  In the latter case, the executive will likely continue to adhere to the agreement (or not actively seek to violate it), for both principled reasons as well as in the hope that domestic political winds will change so as to permit ratification.  In these instances, signature without ratification cannot be considered equivalent to ratification, but neither should it be categorized the same as rejectionist positions.

At the aggregate level, additional concerns arise with respect to treaty approval.  The number of states who are party to the treaty (or who indicate lesser forms of consent) is clearly important.  Treaty regimes represent a network of legal relationships.  If a significant number of states are not part of that regime, its overall effectiveness will be diminished.  Perhaps equally important is the identity of those states who have not given their consent.  For example, a maritime agreement that is only supported by landlocked states is almost meaningless.  Yet even a widely accepted treaty, such as the NPT, is diminished when the handful of states most likely to acquire nuclear weapons are neither signatories nor parties to the agreement.

In this section, we note two possible modifications (with variations possible) between the extreme categories of ratifying states and non-party states:  those states that merely accede or sign an agreement.  Legal differences may exist between these different states, but certainly the political differences are most evident.  With these initial differences established, we now move to draw further distinctions between those in the highest category – treaty ratifiers.

The Timing of Ratification:  A Second Cut

            Among those states that ratify a given international agreement, there is tremendous variation on the timing of that acceptance.  Some states, which might be labeled “early adopters,” were among the first to ratify the treaty.  At the other extreme are states that only became parties to an agreement years or decades after the treaty was completed.  As example of the latter, note that Algeria became a party to the International Covenant of Civil and Political Rights[19] in 1989, almost three decades since its original formulation and over a decade since it entered into force.[20] 

With this in mind, a second modification to the bimodal classification of treaty acceptance is to consider the length of time it takes for a state to ratify an agreement.  The greater that period of time, the less acceptant or supportive of the agreement that state might be.  Of course, there are a number of caveats to this simple formulation.  First, the length of the ratification process is correlated with the subject matter of the treaty.[21]  Second, because the legal implications are more complex, multilateral treaties may take longer to ratify than bilateral ones, and there may be a secular trend toward longer ratification over time.[22]  Third, some states must undergo more rigorous and time consuming ratification processes, as dictated by domestic law, than others.  In those instances, it might be inaccurate to regard them as less supportive than lesser constrained governments.  Even with these limitations, an appropriate baseline standard might be whether the state ratified the treaty before or after the treaty came into force.  Most multilateral treaties require a certain threshold number of states before the treaty becomes binding on any party.  After that threshold is achieved, the treaty comes into force.  For example, the United Nations Framework Convention on Climate Change[23] required fifty ratifying parties to enter into force, and did so less than two years after it was opened for signature.[24]  Those who are early adopters, ratifying before it comes into force, might be considered as the most supportive or demonstrating the most consent.  The speed of ratification suggests strong support and the willingness to commit to the agreement without guarantee of benefits.  Note that when a state is an early adopter, it makes a public commitment to support a treaty with no guarantee that the treaty will ever come into force.  Furthermore, it makes a commitment without full knowledge of which other states will become parties.  For example, the first state that ratifies an agreement that restricts nuclear acquisition or testing (e.g., NPT or Comprehensive Test Ban Treaty[25]) does so with the risk that other key states, perhaps its enemies, will never become parties to the agreement.  That the state accepts such risks suggests strong support for the agreement and its principles.[26] 

In contrast, late adopters, those who ratify after the treaty is in force, may be less committed to the agreement.  They may have problems with some certain provisions and significant domestic political opposition may be present.  Such concerns or opposition may have dissipated over time or perceived benefits of treaty ratification may have increased such that the state becomes a party to the agreement.  Pressure from other signatories may also be a critical factor.  Late adopters also have the luxury of getting signals on which other states will become parties to the agreement.  They also may have the opportunity to gather information about how the agreement is implemented and works in practice.  Thus, their consent might be regarded as much more conditional than those who sign on to the agreement under conditions of incomplete information.  Crudely one could measure the level of consent by the number of years after the treaty went into force that a state joined the agreement.  In the aggregate, one might regard a treaty regime that was rapidly ratified by signatories as stronger or more consensual than one in which the agreement took many years to enter into force.

Thus, another indication of treaty consent involves when a state ratifies a treaty, with special attention to the timing of that ratification vis-à-vis when the treaty came into force.  States who are original adopters may be regarded as more supportive of an agreement than those whose ratification is long delayed.  A third modification moves beyond the timing of ratification to the conditions attached to that ratification.

Reservations: A Third Cut

            Even when a state ratifies a treaty, it may do so only under a specified set of conditions.  States may impose reservations on their ratification.  A reservation is “a unilateral statement, however phrased or named, made by a State, when signing, ratifying, accepting, approving or acceding to a treaty, whereby it purports to exclude or to modify the legal effect of certain provisions of the treaty in their application to that State.”[27]  Whatever their content, reservations indicate that some part of the treaty as originally written is not acceptable to the state in question.  It seems reasonable on first glance that reservations indicate a lesser degree of treaty consent than an unconditional ratification.  Of course, the degree of acceptance indicated by reservations depends on the kind, number, and scope of reservations made by the ratifying state.

            There are a number of different ways of classifying reservations.  One simple way is to divide reservations according their primary focus: (1) substance of the agreement, (2) non-recognition of other parties, or (3) concerns with dispute settlement provisions.[28]  Each has different implications for the conceptualization of treaty consent here.  Substantive reservations indicate that reserving states object to parts of the treaty.  In effect, reserving states may be responsible for (legally) and supportive of (politically) only a portion of the overall agreement, even though they remain formal parties.  The degree of acceptance depends on the scope of the reservations.  Although now a somewhat dated study, John King Gamble found that most substantive reservations were relatively minor[29]; many have involved objections or clarifications of treaty language, such as Luxembourg’s interpretation of the term “lawful sanctions” in the Convention on Torture.[30]  More seriously are major substantive revisions, which are attempts by states to sign onto parts of the treaty they agree with, while relieving themselves of obligations to which they object.  For example, the United States indicated in its reservation to the UN Convention to Combat Desertification[31] that it would not have to make changes to existing land management practices and programs in order to meet its obligations, which seems to undermine the provisions relating to land planning.  A corollary to the substantive reservation type is one in which states indicate that they do not accept treaty provisions that conflict with domestic law.  This can be a largely symbolic or unimportant caveat.  Nevertheless, such a reservation can also emasculate the application of the treaty to that state depending on the substantive focus of the treaty.  For example, Saudi Arabia did ratify the Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW)[32], but placed a reservation that it would not observe those treaty provisions in contradiction to norms of Islamic law.[33]

            Reservations with respect to non-recognition are generally those in which states declare that being a party to a multilateral treaty does not imply recognition of certain states or engender any obligations vis-a–vis those states.  For example, many Arab states make this declaration vis-à-vis Israel, which is a party to many multilateral treaties.  These reservations again indicate a diminished acceptance of the breadth of an agreement, and states that offer them consent to rights and obligations of a lesser scope than states without such reservations.

            States may also put reservations on their ratification that limit enforcement of the agreement.  Classically, many states have put reservations on their acceptance of the International Court of Justice by indicating that domestic jurisdiction issues, as determined by the reserving state are excluded.  Many states made similar reservations not recognizing an investigatory committee or arbitration provisions in the Convention on Torture.  The net effect may be to undermine the treaty as a whole or at least as it applies to a significant number of parties.

            In general then, states that place reservations on their ratification exhibit more limited consent than those who ratify unconditionally.  The degree of difference in these two classes of ratifiers depends on the number, type, and implications of those reservations for the treaty.  Thus, one should assess reservations in determining the extent to which a state (or all states) actually consent to a given agreement.  In the aggregate, a treaty with many reservations from many states can leave a “Swiss-cheese” mosaic of rights and obligations that may weaken the overall regime.

Conclusion
        In this essay, I have suggested that the traditional binary conception of treaty consent (party vs. non-party) is too narrow.  Scholars and policymakers should also consider the methods and breadth of acceptance, the timing of that acceptance, and the reservations attached by states in assessing the degree of consent for each state and for the treaty regime as a whole.  Looking at treaty consent in this fashion is only a first step.  Consideration should then be given to how variations in consent affect a variety of state behaviors with respect to the treaty:  Are states with lower degrees of consent more likely to violate treaty provisions than other states?  Are states with higher levels of consent more willing to sanction members of the treaty regime who violate the agreement?  In the aggregate, are treaty regimes with higher levels of state consent from its members more likely to be efficient and effective?  These are largely empirical questions that must be answered after systematic study.  They have significant legal and political implications.  Yet they can only be asked if we reconceptualize the way we think about treaty consent.