Development of the text of the Ramsar Convention: 1965–1971

Keywords: conservation, legislation, multilateral environmental agreement, protected areas, wetland.

The Convention on Wetlands of International Importance Especially as Waterfowl Habitat (the Ramsar Convention) was novel upon its agreement in 1971 (Ramsar Convention 1971a, 1971b) and remains so through being the only global multilateral environment agreement focused on a group of defined ecosystem types. It is also unusual in that it was not negotiated through the United Nations (UN) system, through which most subsequent environment agreements have been developed. The Convention has been of significance not just for its own considerable achievements, but also for its influence on other international biodiversity treaties and processes; for example, in its creation of a framework for protecting sites of international importance (Bowman et al. 2010), the broader application of its criteria to select these (Stroud and Davidson 2021) and its pioneering application of the ‘wise use’ concept to steer the management of both these protected sites and other wetlands (Finlayson et al. 2011; Pritchard 2021).

Geoffrey Matthews, the appointed Secretary-General of the final negotiating conference that finalised the Convention in Ramsar, Iran, in February 1971 detailed an account of the initial 25 years of the Convention’s development, including describing the genesis of the text from first international discussions in 1962 (Matthews 1993). Although Matthews (1993) describes six different negotiating texts circulated to interested governments, to our knowledge none of these early drafts has previously been published, and all are included as supplementary material to this paper (see list in Table 1). Using historical papers from the files of Great Britain’s Nature Conservancy, the UK government’s statutory conservation agency at the time, and retained by Geoffrey Matthews, we compare, in detail, an early draft with the final text agreed in Ramsar to explore the early development of some of the Convention’s key concepts.

Table 1.  Timeline of decisions and drafts of the Convention on wetlands (drawn from Matthews 1993)

In the second half of the 1960s there were few, if any, examples of what a broad-ranging and, more importantly, an effective multilateral environmental treaty may look like. In North America in the early 20th century, there had been early multilateral treaties related to sustainable utilisation of shared resources of economic importance, such as northern fur seals (Callorhinus ursinus) and sea otters (Enhydra lutris), as well as for migratory waterbirds, but they were geographically limited (Dorsey 1998). Regional conventions, as well as global conventions such as the 1946 International Convention on the Regulation of Whaling (see http://www.iwcoffice.org/private/downloads/1r2jdhu5xtuswws0ocw04wgcw/convention.pdf, accessed 6 April 2022), had been singularly unsuccessful in managing shared international resources (Dorsey 2014). From a European perspective, both the 1902 Convention for the Protection of Birds Useful to Agriculture (see http://www2.ecolex.org/server2neu.php/libcat/docs/TRE/Full/En/TRE-000067.txt, accessed 7 April 2022) and its successor, the 1950 International Convention for the Protection of Birds (see http://www2.ecolex.org/server2neu.php/libcat/docs/TRE/Full/En/TRE-000066.txt, accessed 7 April 2022) (both called the Paris Convention), were substantially ineffective, with weak provisions, limited accession and, critically, no international coordination, all limitations that were to influence the later drafting of the Ramsar Convention (Hayden 1942; Lyster 1985; Ferrero-García 2013).

In the Americas, the Convention on Nature Protection and Wildlife Preservation in the Western Hemisphere, or Western Hemisphere Convention, had entered into force in 1942. Although ‘visionary’ in its scope (Bowman et al. 2010) for provisions for protected areas (albeit national sites, rather than internationally important sites), migratory species other than waterbirds and international cooperation (and even trade), it lacked administrative and enforcement mechanisms. This had severely limited its effectiveness.

The scope of the Organisation of African Unity’s 1968 African Convention on the Conservation of Nature and Natural Resources, developed at the same time as the wetlands (Ramsar) convention, was theoretically ambitious, covering issues of relevance to wetland conservation such as water supply and the need to establish protected areas. However, it lacked administrative structure to facilitate implementation or oversee its enforcement (van der Linde 2002; Bowman et al. 2010).

In the 1960s then, the prospects for a comprehensive and effective global wetland convention were limited, with W. A. Panis noting in 1968 that:

…so far hardly any experience has been gained of general agreements concluded by a plurality of States and concerned with natural environmental planning… [see File S7 of the Supplementary material]

and that

Experience with, for example, the [1950] Paris Convention is not very encouraging.

As well as the drafts of the convention themselves (Table 1), accompanying informal (and latterly formal) cover notes provide very important narratives as to the intention of drafters. There were six primary drafts of the convention prior to the 1971 conference at Ramsar, and we have systematically numbered these based on Matthews (1993) and his unpublished archive of drafts. These are all given as supplementary documents, as outlined in Table 1. At the outset, there was no agreed timetable or sequence of meetings to develop the convention, which occurred rather opportunistically.

We present the first draft in Appendix 1, but this was rather ‘unformed’ and drafted without input from lawyers. In Appendix 2, we analyse the content of Draft 3, matching this against the final text inasmuch as this is possible. We have chosen Draft 3 for this comparison because, by 1968, the draft had already received legal commentary from the International Union for Conservation of Nature (IUCN) and multiple international inputs, and its content had become significantly elaborated. However, the structure of the draft text changed considerably between Draft 3 and the final text, not just the high-level sequence of Articles, but also their content.

The first suggestion for an international convention on wetlands arose from Baron Le Roy, a representative of a hunting organisation (Association Nationale des Chasseurs de Gibier d’Eau), at the 1962 Mar Conference, co-organised by the International Wildfowl Research Bureau (IWRB; now Wetlands International), the IUCN and the International Committee for Bird Preservation (ICBP; now BirdLife International; Hoffman 1964; Matthews 1993, p. 12).

This conference brought together waterbird and wetland scientists who concluded that ‘in temperate regions drainage of wetlands is proceeding at an increasing rate and without reference to their diverse values…’ (Hoffman 1964, p. 29). The conference recommended that:

IUCN compile a list, in accordance with an internationally agreed classification, of European and North African wetlands of international importance … to be placed at the disposal of conservationists, and those responsible for development schemes; and further recommends that this list may be considered as a foundation for an international convention on wetlands. [Hoffman 1964, p. 29].

The list was published by Olney and IWRB/Mar Bureau (1965), presenting an initial list of sites largely important for waterbirds, but with further additions of sites projected as knowledge grew.

Although Mar was essentially a scientific conference, the need to engage governments more formally was recognised. The following years saw a sequence of government-sponsored European Meetings on Wildfowl Conservation, at St Andrews, Scotland (1963), then Noordwijk ann Zee, Netherlands (1966), and subsequently in Leningrad, USSR (1968).

IWRB initiated the development of the convention text, with the circulation to ‘appropriate authorities’ in 32 countries of 8 issues for inclusion (Draft 0; see File S1 of the Supplementary material) in October 1965, with a first draft also circulated in 1965, consisting of a Preamble and 6 articles (Draft 1; File S2 of the Supplementary material). The following year, the Noordwijk ann Zee meeting reviewed the IWRB proposals and requested the Dutch government to draft a convention based on the Draft 0/1 suggestions as modified by discussions at that conference.

Matthews (1993, p. 15) recorded that the Dutch government circulated, on 12 October 1967, a draft of a ‘Wetlands Convention’ with 21 Articles (Draft 2; File S3 of the Supplementary material). Matthews (1993) noted that within a month, a document entitled Comments on the First Draft of a Wetlands Convention had been prepared for the IUCN. In this, the draft was criticised for not taking into account the relatively low appreciation of wetlands then current. Instead, far-reaching obligations were imposed, and wetland protection made paramount. An international authority was planned to have final judgement on national affairs. The IUCN felt that few states would become parties to such a convention and considered that a much more modest approach may be more fruitful.

Draft 2, with commentaries from the IUCN and the international environmental lawyer Cyril de Klemm, was discussed at a meeting of IWRB’s Executive Board at the IUCN’s Morge headquarters in November 1967, and the conclusions (File S4 of the Supplementary material) transmitted to the Dutch government with a request that a redraft be prepared for the 1968 Leningrad conference.

The second Dutch government draft (under the authorship of Dr W. A. Panis but ‘designed in close co-operation with the chief of Treaty Department of the Dutch Ministry of Foreign Affairs’; W. A. Panis, cf. File S7) was circulated informally to (at least) Phyllis Barclay-Smith (a major UK motivator of international bird conservation and Secretary-General of the ICBP) seeking comments. This Draft 3.0, which we explore, is presented in Appendix 2 (File S5 of the Supplementary material). The subsequent draft circulated formally to the Leningrad conference (Draft 3.1; File S6 of the Supplementary material) was changed in relatively few respects (as indicated in Appendix 2) and with an introduction provided by the Dutch Government (File S7 of the Supplementary material).

Following the USSR-led invasion of Czechoslovakia by troops of the Warsaw Pact in August 1968, the Leningrad conference in September was boycotted by most Western European governments, with the IWRB withdrawing from its formal organisation. Despite this disruption to the negotiating processes, the draft convention survived with participants:

…consider it expedient to hasten the adoption of a convention concerning wetlands conservation, and to provide for a strict protection of those wetlands that have an international importance [Anonymous 1970].

Subsequently, Matthews stated:

…on 11 February 1969, IWRB received from the USSR Ministry of Agriculture another text for ‘An International Convention on Wildfowl and Wetlands’ [Draft 4; File S8 of the Supplementary material]. Of its 13 articles, a number were rather similar to those of the Dutch text, but others differed quite substantially [Matthews 1993, p. 20].

This version and the Dutch text (Draft 3.1) were considered point by point at the IWRB’s next Executive Board meeting in Vienna in May 1969. The conclusions (File S9 of the Supplementary material) were transmitted to the Dutch government, which agreed to draw up a compromise text by July 1969, although failure to do this by August 1969 resulted in the IWRB producing Draft 5.0 (File S10 of the Supplementary material) and shortly afterwards producing Draft 5.1 (File S11 of the Supplementary Material) At the same time, Iran offered to host an international conference to finalise a wetland convention in early 1971 (Matthews 1993; File S12 of the Supplementary material).

In September 1969, the Ninth Biennial Conference of the International Union of Game Biologists in Moscow gave an opportunity to restore international relations (Matthews 1993). The USSR generally accepted the IWRB’s compromise text (Draft 5.0) subject to approval by a technical meeting to be called in Finland the following year.

In December 1969, the Dutch government circulated an official draft proposal for a Convention on Wetlands as Wildfowl Habitat (Draft 5.1; File S11 of the Supplementary material).

This was very nearly identical with the compromise draft [Draft 5.0] drafted by IWRB and sent to [the Dutch] Ministry in August. [Matthews 1993, p. 23].

Following very wide circulation, Draft 5.1 was discussed in Helsinki, Finland, in March 1970. Line-by-line examination of the text resulted in some significant changes, as summarised by Matthews (1993). The IWRB was asked to produce a final draft to be sent to all countries likely to attend the Iran conference (Draft 6; File S13 of the Supplementary material), together with a substantial cover note (File S14 of the Supplementary material).

In March 1970, the Imperial Iranian government extended formal invitations ‘to the Ministries of Foreign Affairs of the countries of Europe, North Africa and Asia’ to participate in an international conference, initially planned for the city of Babolsar but then shifted to Ramsar (Matthews 1993, p. 26).

The IUCN, which previously ‘had been unenthusiastic about the convention’ (Matthews 1993, p. 26), had made late and ambitious suggestions for textual changes in September 1970, but these were rejected by the IWRB’s Executive Board to avoid reopening discussions on Draft 6 prior to final negotiations in Iran. Matthews (1993, pp. 26–30) gives a detailed account of the discussions held in Ramsar in 1971.

One of the key issues to emerge from the negotiating history of the convention was the critical engagement with the USSR. This was especially notable during the ‘coldest’ part of the cold war. USSR’s strong support for an ambitious convention was expressed through their hosting of the 1968 Third European Wildfowl Conference in Leningrad, and despite the Western boycott following the invasion of Czechoslovakia, continued engagement in producing their own draft (Draft 4) convention text, subsequently to be reconciled with earlier circulating drafts (Table 1). It is clear from Matthews’ (1993) more detailed account that exceptional personal relationships underpinned this crucial east–west collaboration.

Of the supplementary documents we present, perhaps the most instructive are the information notes accompanying Draft 3 (W. A. Panis, cf. File S7) and Draft 6 (IWRB 1970; File S14). These give unique narrative accounts, by the drafters, explaining their rationale and approach. The 1968 note by W. A. Panis clearly outlines his recognition of the risks (see especially pp. 2–4): that there had been no previous effective multilateral environmental treaties, but the state of wetland loss and degradation was such that international actions were urgently required. Despite this there would likely be little political appetite for states in the 1960s to accede to an overambitious convention, especially with prescriptive measures and a compliance or sanctions regime. Thus, drafts needed to find a way that brought parties on board but also delivered a useful framework that would deliver national actions. In this, Panis recognised the need ‘to interfere as little as possible with national autonomy’ (see File S7, p. 2), such that ‘the draft goes no further than the acceptance of the principle stated at the beginning of the section concerned, nor does it give directives on the measures the parties are to take’ (see File S7, p. 4).

Regarding compliance,

No provision of any kind is made for sanctions if a State should fail to observe these directives. The emphasis in the convention is advisedly not so much on the establishment of a system of enforceable standards as on the institutionalization of permanent international consultation on as broad a base as possible. [W. A. Panis, cf. File S7, p. 3].

His logic was that:

No sanction whatever would be realistic in the majority of cases, for any decision to ignore a recommendation would normally be a result of weighing national and international considerations, and in many cases the national interest will be deemed to be so predominate that it will have to prevail over international aspects. [W. A. Panis, cf. File S7, p. 4].

(Softer ‘naming and shaming’ provisions have since proved a more pragmatic but weaker alternative; Stroud et al. 2021.)

This was a skilful and politically aware approach to the task. In contrast, in the drafting period as described by Matthews (1993, p. 26), the IUCN seem to have badly misread the politics, initially being unenthusiastic about a wetlands convention and failing to engage with it, but then, in 1970, making last-minute unrealistically strong proposals, especially on sanctions, at odds with the more nuanced approach of Panis for the Dutch government and of the IWRB. From Matthews’ (1993) tactful account, it appears these IUCN proposals were quietly dropped.

The general drafting approach was to oblige the creation of national wetland policies ‘while no exact indication is given as to what form that policy or those measures should take’ (W. A. Panis, cf. File S7, p. 2). This was playing a long game and, in subsequent years, the Conference of Parties (COP) to the Convention has adopted formal guidance on an increasingly wide range of relevant conservation issues, including on policy, as summarised in the library of 20 handbooks published by the Ramsar Secretariat (https://www.ramsar.org/resources/the-handbooks, accessed 21 September 2021). The COP process has progressively created the detail missing from the formal 1971 convention text, a general approach similar to that adopted by more recent multilateral environment treaties such as the UN’s Framework Convention on Climate Change and Convention on Biological Diversity (both of 1992).

The development of the wetlands convention outside the UN system has led to consequences that are still being debated in recent COPs. Yet, in the late 1960s there was much environmental activity within the UN, with preparations for the 1972 Stockholm Conference on the Human Environment ongoing from 1968.

The lack of UN status results from the wetlands convention being developed initially as a European and North African regional initiative (as witnessed from the invitations to the Ramsar conference) and being driven (initially, at least) by non-government organisations ‘bottom-up’, rather than as a ‘top-down’ global initiative via the UN system. Indeed, the manner of its development seems to have confused UK government officials. The UK file contains a 18 June 1968 letter from the Home Office, London to the Nature Conservancy (P. E. Baker, in litt., 1968):

I attach a copy of a letter and enclosure [Draft 3.0] I have received from Miss Barclay-Smith [of IWRB] about a proposed convention on the conservation of Wetlands and Wildfowl. The draft convention does not appear to have been produced by one of the recognised international bodies, and this no doubt accounts for the unusual manner in which it has been brought to our attention.

However, the final text was usefully silent on its geographic scope, thus allowing its eventual adoption globally.

The fact that the convention agreed in Ramsar became global is entirely due to its subsequent promotion by the IWRB. Constitutionally, the IWRB was a hybrid international organisation with both non-government and government members, and it heavily advocated accession with its state members. Based on this effort, the Ramsar Convention already had 33 parties by the time the Stockholm-mandated Migratory Species Convention came into force in November 1983 with 15 parties (Bowman et al. 2010).

Based on the above-mentioned documentation, we now focus on a few important thematic issues in the convention’s development.

The Mar Conference in 1962 had been essentially European, although there had been single attendees from Australia, Canada and Morocco, with six US delegates. The Introduction for Draft 3 of the Convention (Appendix 2; File S7) expresses ambivalence as to the scope of the proposed Convention:

The area covered by the Convention, which by its very nature cannot be world wide, will have to be decided by specialists from the European, North and Central African and South-West and Central Asian regions. [W. A. Panis, cf. File S7].

Matthews (1993, p. 18) commented that ‘Such a myopic viewpoint is now difficult to comprehend.’

Subsequent drafts grew in geographic aspiration, progressively leaving aside initial ‘cannot be world wide’ doubts, but retaining the potential convention’s scope as ‘international’. Nowhere in the papers we have is the word ‘global’ used. The formal Preliminary Notice for the Ramsar Conference (File S12 in the Supplementary material) states:

The time is now ripe for a move towards positive international action to encourage and co-ordinate conservation measures by individual countries. This is particularly with regard to the complex of wildfowl populations breeding in Northern Eurasia and wintering in south and west Eurasia and in Africa south to the Equator.

Iran only invited countries in Europe, North Africa and Asia to its negotiating conference. A total of 28 countries attended (including observer states), all but 6 of which (India, Iran, Jordan, Pakistan, South Africa and Turkey) were European. As a result of lack of invitation, there was no representation from East Asia, the Americas or Oceania (in contrast with the earlier technical meetings in the UK (1963), Netherlands (1966) and Leningrad (1968), which had attendees from Canada, Ceylon, Ethiopia, India, Israel, Senegal and USA). Despite this Eurocentric start, the Convention has progressively grown to a current global membership of 172 parties, although membership growth has not been regionally uniform (Fig. 1), being significantly slower in Oceania and Asia.

Fig. 1.  Regional accession rate to the Ramsar Convention expressed as a cumulative proportion of all states in each of Ramsar’s six regions. Green, North America (three parties); dark blue, Europe (potentially 48 parties); yellow, Latin America and the Caribbean (potentially 32 parties); light blue, Africa (potentially 54 parties); orange, Asia (potentially 42 parties); grey, Oceania (potentially 16 parties).

By 1968, Draft 3 (Appendix 2) was attempting to define the scope of the incipient convention as a list of specific wetland types (lagoons, swamps, ponds etc.). However, the accompanying information note (W. A. Panis, cf. File S7) clearly notes that ‘the summary of areas in the second paragraph is enumerative and does not exclude the possibility that areas of other kinds might be designated as wetlands.’ Although, with hindsight, this clearly did not represent the full range of interest from potential contracting parties, with the later inclusion of a wider range of wetland types (missing, for example, at that time are wetland types such as mangroves); given the then ambivalence as to the proposed convention’s geographic extent (see above), the approach is not unexpected.

The final draft (Draft 6; File S13) adopts an approach based on wetland attributes (e.g. whether the wetlands are ‘natural or artificial, … with water that is static or flowing, fresh brackish or salt,…’). Such an approach allows any area to be defined as wetland or non-wetland without the need for its inclusion as a particular wetland type in a list. The approach finally adopted has proved robust for the general purposes of the Convention, although its recognition in other national and international contexts has been limited. Thus, both the UN’s Aichi Targets and Sustainable Development Goals split wetlands across targets, and there are multiple examples of different national wetland classifications (Gerbeaux et al. 2018).

Central to the initial concept of the Convention was a COP. This has been fundamental to the effectiveness of the treaty. However, a number of important elements as to its working were missing from the final text, including the authority of the COP to adopt financial provisions for the organisation of its meetings, funding a secretariat and other purposes.

Article 11 of Draft 3.0 had set out a first, inadequate, attempt to create a finance mechanism. This established (Article 11.1) that nationally incurred expenses ‘shall be borne by the Contracting Party concerned’. However, Article 11.2 attempted to establish a funding mechanism for those collective tasks required under Article 9.3b, essentially the provision to the COP of relevant contextual ‘scientific reports, statistics, observation and other documents’. Four categories of contributions were proposed ranging from 10 000 to 100 000 Dutch guilders, but without criteria as to assignation. Such budgetary prescription was quickly dropped: later in 1968, Article 11.2 of Draft 3 just had four unspecified categories of contribution. By 1971 the final text is silent on the issue of budgets.

The need for a financial provision was formally recognised at COP 1 (Resolution 1.8) and ultimately delivered through amendments made to the convention text through a two-stage process: (1) the separate agreement of a Protocol to the Convention (the ‘Paris Protocol’) adopted by an Extraordinary Conference in 1982 that established legal provision to amend the Convention text; and (2) through amendments adopted by the second Extraordinary Conference (in 1987 prior to COP3) which inter alia finally settled the issue of budgetary competence by insertion of new Articles 6.5 and 6.6 allowing the COP to establish procedures for financial contributions (see below).

During the drafting process there was increasing recognition of the critical need for a secretariat to service the Convention and its parties, given the lack of such was one of the main reasons of the non-functionality of the Paris Convention and other early treaties. Draft 3.0 proposed coordination be provided by a ‘Secretariat of the Conference’ (Articles 8.1 and 8.2), with a biennially convening COP proposed as competent, under Article 9.3c, ‘to establish directives covering the work and the size of the Secretariat of the Conference for the coming two-year period’. By the final text (Article 8) it had been resolved that the IUCN ‘perform the continuing bureau duties’ with a list of tasks needing to be undertaken.

Notwithstanding the more detailed breakdown of support tasks in the final text, the practical and organisational mechanisms for delivering permanent secretariat support was to take some years to stabilise, being immediately recognised as a need at the first COP (Recommendations 1.8 and 1.10), as documented by Matthews (1993). From 1971, interim secretariat functions to the Convention were undertaken by the IWRB on a voluntary basis and, from 1981, under formal IUCN subcontract. From 1987 these services were undertaken by a semi-autonomous ‘Bureau’, since 2005 called a ‘Secretariat’.

More recent multilateral environment treaties have included, from the outset, a scientific subsidiary body to provide parties with technical and scientific advice. Although explicitly recognised as a need at COP 1 (Recommendation 1.8), such provision for Ramsar only came significantly later at COP 5 (Ramsar Convention 1993) with the establishment of the Scientific and Technical Review Panel (STRP). Prior to this, a number of organisations, in particular the IWRB and IUCN, provided scientific and technical advice to the Convention through participation in technical sessions at the COPs, or through individual activities and alignment of their own programs to the scientific interests of the Convention.

At the heart of early thinking about a wetland convention was the need to identify and conserve remaining wetlands of perceived international importance. This came directly from outcomes of Project Mar (Hoffman 1964), which developed and then published (Olney and IWRB/Mar Bureau 1965) a list of internationally important wetlands defined according to specified criteria. Although these were especially, although not exclusively, identified on the basis of their importance for waterfowl, Project Mar was explicit that such interest was a surrogate for other ecological values for which international data did not then exist. The use of ornithological criteria was seen as an expediency because it was readily available. The ‘list’ of waterbird sites was explicitly seen as an initial list (Olney and IWRB/Mar Bureau 1965) intended for subsequent addition with other interests, such as, for peatlands, with data from the International Biological Programme and IUCN’s Project Telma (Bellamy and Pritchard 1973).

Thus the primary role of the proposed wetland conservation convention was to ensure, through government policy actions, the sustained conservation of these few critical sites. At the time of drafting the convention there were no other models of what international site-based protection under a multilateral treaty may look like, notwithstanding a long history of nationally protected areas such as national parks, notably in Australia and USA (Holdgate 1999), so for the drafters this was entirely new territory.

The initial thinking is presented as Article 6 of Draft 3.0 (Appendix 2), although by 1970 (Draft 6), and because of the centrality of the issue, this had become Article 2. In Draft 3.0, parties are obligated (but only ‘so far as it considers it possible’) to inscribe wetlands ‘in the List pertaining to this Convention’ (Article 6.1). Draft 3.0 then makes clear, as in the final text, that the first listing shall occur when a country joins the convention (Article 6.2) and that further listings may follow (Article 6.3).

Although Article 7 in Draft 3.0 states that ‘the wetlands inscribed in the List shall be the subject of the joint care of the Contracting Parties’, in seeming contradiction it also states that such care should be ‘without prejudice to the exclusive rights attaching to the sovereignty of each of the Contracting Parties individually over the wetlands inscribed by it in the List’. The nature of such joint international care is unclear, although Article 6.4 envisaged a process where any party would need to obtain a formal ‘opinion’ from the COP before a party could delist a site on the list or reduce its boundaries.

The COP is also envisaged to have a formal role in providing advice to a party following that country’s notification to the COP ‘…that changes are likely to occur in the ecological nature of any of its wetlands figuring on the list…’ (Article 8.1). The COP would then provide a formal recommendation to the party concerned (Article 8.3) ‘if possible’ appending ‘concrete proposals for preserving the whole or part of the wetlands in question, limiting the effects of any encroachment on the wetlands and for co-ordinating these interests with others affected by the changes’ (Article 8.4).

Such proposals, notwithstanding the purported claim that they were ‘without prejudice to the exclusive rights attached to … sovereignty’ were quickly found unacceptable. Matthews (1993) noted that ‘the curtailment of national sovereignty would be unacceptable, as would the prevention of any planned changes considered undesirable by nature conservancy experts’.

However, the suggested model is of interest in that aspects have since developed through subsequent decisions. Thus, final Article 3.2 requires that parties report cases of changes in the ‘ecological character’ to the Secretariat. Final Article 8.2d requires the Secretariat to bring such cases to the attention of the COP, at which this issue is a standing agenda item. Final Article 8.2e envisages a further role for the Secretariat in communicating to the party concerned ‘the recommendations of the Conferences in respect of such alterations to the List or of changes in the character of wetlands included therein’.

Although early COPs adopted recommendations explicitly addressing specific sites (e.g. separate COP 2 recommendations related to the trilateral Wadden Sea, Sahel wetlands, and two related to the Doudj National Park in Senegal), more recent COPs have rarely adopted detailed site-specific recommendations, especially not in regard to management prescriptions. Rather, separate advisory missions, initially as a management guidance procedure and now Ramsar advisory missions (Gardner et al. 2018a, 2018b; Jones and Pritchard 2018) provide a means of giving formal advice to parties with regard to best management of sites with problems. Removing such advice from the political arena of a COP has been helpful to positive outcomes, as shown by the success of Ramsar advisory missions (Gardner et al. 2018a), such as that for Chilika Lagoon, India (Finlayson 2020), for the Ouse Washes, UK (Mansfield 2014), which triggered a range of wetland creation initiatives, and at Parc national des Virunga, Democratic Republic of the Congo (Gardner et al. 2018b).

Notwithstanding the centrality of the obligation to designate Ramsar sites, the draft and final text lack criteria for their selection. Stroud and Davidson (2021) separately document the history of these criteria.

There is no suggestion in Draft 3.0, nor indeed in the final text, as to the need for regular reporting on the status of listed sites. The need for a standard description of listed sites through an information sheet was to come later in 1990 (Ramsar Convention 1990) and the regular update of such status descriptions after that (Ramsar Convention 1996).

As noted above, the interaction between sovereignty and collective obligation was a significant discussion area when drafting Ramsar’s text. With respect of changes in the ecological nature of a listed Ramsar site, Article 8.2 of Draft 3.0 mandated the COP to ‘make known its opinion on the notified change in the form of a recommendation to the Party or Parties concerned’. This requirement largely survived the drafting process with an obligation on the Secretariat to ‘make known to the Contracting Party concerned, the recommendations of the Conferences in respect of such alterations to the List or of changes in the character of wetlands included therein’. This results in a formal resolution at each COP on listed sites where, in theory, such recommendations can be made. In practice, this now rarely occurs with recent resolutions on listed sites (e.g. Ramsar Convention 2018) substantially silent as to recommendations to other parties.

The accompanying information note (W. A. Panis, cf. File S7, p. 2) recognised the issue that many important wetlands straddle international boundaries, noting that:

…some room has been left [in Article 1, paragraph 1] to enable a number of Contracting Parties jointly to designate areas to be protected. This can be particularly important to areas situated on the territories of two or more States.

By the final text (final Article 5), this notion of transboundary designations had changed to a more general obligation for parties to ‘consult with each other … especially in the case of a wetland extending over the territories of more than one Contracting Party’. General guidelines for international cooperation, especially with respect to shared wetlands were subsequently adopted (Ramsar Convention 1999), although the concept of joint designations as an expression of such cooperation has been slow to develop. To date there are 57 Ramsar sites in 22 parties described as transboundary.

Initial (Draft 3.0) obligations to maintain the ‘ecological character’ of wetlands were poorly expressed, referring to ‘changes in the … ecological nature’ of listed sites, although the final Article 3.2 was scarcely more detailed. The term ‘wise use’, so central to the convention’s obligations, appears to have been added at the Ramsar conference. Article 4.1 of the final Draft 6 (File S13) requires that:

Each Contracting Party shall promote the conservation of wetlands and waterfowl by organising nature reserves on wetlands whether they are included in the List or not.

The final formulation of Article 3.1 to ‘formulate and implement their planning so as to promote the conservation of the wetlands included in the List, and as far as possible the wise use of wetlands in their territory’ establishes a much wider concept and moves away from a particular mechanism (the establishment of nature reserves) to establish a wide palate of policy options. As explored by Pritchard (2021), what the wise use concept means and what the policy options are for its delivery have needed significant subsequent development.

In the second half of the 1960s there were no examples of what a broad-ranging and effective environmental treaty may look like. With respect to habitat conservation and land use, the only precedent was the ambitious 1968 African Convention on Nature and Natural Resources, which required parties to take conservation measures in respect of soil and water and to ‘maintain and extend’ conservation areas. However, there is no evidence for significant African input into the development of the Convention (the Mar Conference was attended by a single delegate from Morocco, with the Ramsar meeting attended by just a single (South) African delegation), suggesting that during its evolutionary phase the wetlands convention developed independently from the African Convention.

The Western Hemisphere Convention has been in force since 1942 but lacks administrative provisions (Bowman et al. 2010), whereas the International Convention on the Regulation of Whaling had, since 1948, established an International Whaling Commission that had met annually, although an adequately funded secretariat was not in place until decades later, materially impeding its effectiveness (Dorsey 2014).

At the time then, what was proposed for wetlands following the 1963 Mar Conference was visionary.

In parallel to the development of a wetland convention, the UN was increasingly focusing on environmental issues. This led to the 1972 Stockholm Conference on the Human Environment, the outcomes of which were presented to the General Assembly of the UN later in 1972, which accepted the declaration and action plan in Resolution 2995 (XXVII) mandating, inter alia, the establishment of the UN’s Environment Programme (Chasek 2020).

There were no global conventions on biodiversity conservation before Ramsar, and it has pioneered international processes and ways of working, acting as a procedural ‘test bed’ for conventions that were to come later, such as those on trade in endangered species and on migratory species. In particular, it was the first international biodiversity treaty to recognise the need (in principle at least) for a permanent secretariat to drive activity, notwithstanding that the final structures that exist now took some time to put in place. It also gave the first recognition of the need for regular meetings of the parties through a formal conference of the government parties. Both these elements were within the very first drafts of the Convention text. Thus, beyond its critical importance for wetlands themselves, the treaty has been enormously influential for international environmental conservation more widely.

The availability, for the first time, of the early convention drafts makes clear not only how the text convention developed, but also the thinking of its drafters and proponents. As well as being of importance for the historical record, these documents give pointers as to how the convention may further evolve. Accordingly, and 50 years on, it could prove useful to reappraise the influence and, more specifically, the direction of the Convention given the greatly increased membership and experience from other agreements.

Data sharing is not applicable because no new data were generated or analysed during this study.

Nick Davidson and Max Finlayson are editors for Marine and Freshwater Research but did not at any stage have access to this manuscript while in peer review, as is standard practice when handling manuscripts submitted by an editor to this Journal. Marine and Freshwater Research encourages its editors to publish in the Journal and they are kept totally separate from the decision-making processes for their manuscripts. Other authors have no conflicts of interests. The authors collectively have had long-involvement in the Convention’s international development and national implementation, variously as national delegates to multiple COPs, as chairs and members of STRP, as chairs of national Ramsar committees, as representatives of the Convention’s international organisation partners and as Deputy Secretary General of the Secretariat.

This research did not receive any funding.