International treaties are generally negotiated and signed by national bureaucracies. Defense ministries negotiate military agreements, education ministries negotiate educational exchanges, and so forth. When they face challenges or disagreement in carrying out an agreement, such ministries often avoid tit-for-tat punishment strategies in favor of more flexibility in implementation and even renegotiation of agreements. They also avoid bringing legal action, even if international dispute resolution is available in the World Trade Organization or elsewhere. More typically they tend to resolve disagreements politically, through ongoing bargaining or, exceptionally, by bringing disagreements to a higher political level for resolution.Footnote 1
National bureaucracies cooperate in much the same way when they negotiate memoranda of understanding with one another. Despite being governed by a single domestic legal system, enforcement of domestic agreements is also largely political. For example, the US National Park Service (NPS, in the Department of Interior) and the US Forest Service (USFS, in the Department of Agriculture) do not sue each other over violations of any agreement.Footnote 2 They only rarely refer disputes to a higher political level—the President—because the demands on any president's time are too great. Monitoring, enforcement, implementation, and renegotiation are largely political, not legal, problems for these agencies.
Agencies interacting domestically do differ from those interacting internationally in being responsible to different kinds of principals. Two US agencies will ultimately report to the President, while a Canadian agency will report to the Prime Minister. In a single country, a common superior can enforce cooperation on subordinates. For example, the 1993 Northwest Forest Plan resolved the spotted owl controversy among multiple federal and state agencies by bringing together the President, Vice President, and five cabinet-level officials—a level of intervention that will rarely be practical.Footnote 3 Even such exceptional cases have international analogues, however. European summits can mandate domestic agencies to work together, as can the Council of Ministers, European Commission, or the European Court of Justice under various conditions. Other summits, such as the G-7, can also produce pressure on agencies “from above.”Footnote 4 These pressures from above differ in many respects, but they also have more similarities than our separation into “international relations” and “comparative politics” typically acknowledges.
I argue that cooperation between two agencies presents much the same problem whether these agencies are found in different countries or in the same country. This claim extends insights from Graham Allison's bureaucratic politics approach—though Allison conceived as interagency politics as something that determined the foreign policy of a “state” interacting with other “states.”Footnote 5 The approach here aims at a fuller account of international/interagency relations instead of a theory of foreign policy in a single state.
The similarity between interagency cooperation and international cooperation is generally overlooked because the issues over which agencies negotiate often differ. For example, the defense agencies of the US and Canada negotiate over missile defense, while the states of Massachusetts and California do not. Such differences between “international relations” and “interagency relations” may simply reflect the fact that defense policy and education policy differ as issues. Footnote 6 Agencies that negotiate with the threat of military force behind them presumably represent an analytically-different problem than agencies that negotiate in the absence of such threats. However, academic theories of cooperation have not acknowledged that the shadow of force represents the exception and not the rule. The issue and not the unit represents the more important analytical distinction.
I will argue that variation in cooperation does not depend on whether two actors cooperate across an international boundary, nor does it depend on the nature of the units involved (nation-states, domestic agencies, and so on). To show this, I hold issue area and other variables constant while allowing interstate and intrastate units to vary. Comparing similar cooperation within and across international boundaries provides the simplest possible test of the claim that agencies, and not states, are the proper unit of analysis.
Testing this central claim requires a theory that can explain variation in cooperation using variables other than the nature of the unit. The theory that I develop yields straightforward predictions about the effects of homogeneity or heterogeneity in preferences; and symmetry or asymmetry in the problem being addressed. Within this general approach, I use a mid-range theory of a particular issue area, cooperation on wildlife management, to provide predictions that are sufficiently specific to allow testing. Variation between successful and unsuccessful cooperation in this issue area is governed solely by characteristics of the species and agency goals for the habitat they manage. For example, two agencies that both desire more animals will cooperate if these animals are migratory but will not cooperate otherwise. The use of variables such as preference homogeneity and issue symmetry should generalize to many issues beyond the one addressed here, and the concrete theory developed here is meant to represent a larger set of theories addressing cooperation at both the domestic and international levels.
Testing the central claim that the international-domestic divide does not affect cooperation requires examining an issue area that is “scalable,” falling within the jurisdiction of government units both large and small. Many environmental issues share these characteristics, being regulated both locally and internationally. Wildlife, which easily crosses human boundaries, provides an excellent subject for this study because it raises similar land management problems across multiple levels of government. There is a clear policy need for transboundary cooperation in wildlife management, especially for larger mammals and many migratory species. Many carnivore and ungulate populations are found in islands isolated from one another, with a concomitant increase in the risk of local extinction.Footnote 7 Both gray wolves and grizzly bears, for example, are found in Yellowstone and Glacier National Parks but are not established in the 400 miles of territory between these parks. Such transboundary problems create an incentive for transboundary cooperation (TBC), defined as cases in which two actors jointly agree to change their actions in some policy area, each condition on the other.Footnote 8
Variation between symmetric and asymmetric problems can be used to study other environmental issues too. Shared lakes present symmetric problems while upstream and downstream actors on a river face an asymmetric problem. Managing pollution in air sheds and river basins or water usage in shared rivers would also provide good, scalable issue areas for a study seeking to replicate or extend the analysis here.
Though this is but one theory, and it addresses behavior in a relatively minor issue area, this study was designed to have wider implications. The issue area was chosen for its scalability, to provide an illustration of how students of international relations can compare cooperation at the domestic and international levels so as to determine whether “international” is a causally important variable. The particular theory of wildlife cooperation depends on general theoretical notions such as whether actors' preferences are the same or different, and whether the substantive problem is symmetric or asymmetric. By looking at TBC on wildlife issues, I simply make the general claim concrete in a mid-level theory that can be tested.
Thinking of international cooperation as a form of interagency cooperation changes how we analyze international cooperation. For scholars who think in terms of nation-states interacting either in an anarchic international systemFootnote 9 or in an international society,Footnote 10 this points to a very different unit of analysis—the agency, not the “state.” This approach also suggests a recasting of the nature of anarchy. In particular, the approach questions the extent to which the political and legal environment for the US Departments of Interior or Agriculture differs from the environment faced between the US National Park Service and Parks Canada. For those who emphasize the domestic politics of international cooperation,Footnote 11 the analysis here moves us away from executives constrained by legislatures to look at sub-units within each executive.
At an extreme, the study of international cooperation need not assume that the adjective “international” has any causal importance at all. This study of cooperation would suggest that the burden of proof be imposed on those who believe the adjective is important.
Why Might International Politics Differ from National Politics?
Before examining how international boundaries do not change the basic transboundary problem, we should consider several plausible reasons why the category of “international” might play a key role in TBC problems. The foremost reason why international TBC may differ from domestic TBC is the problem of anarchy, and the related issue of sovereignty.Footnote 12 Anarchy refers to the lack of any international government or legal system with the power to make legally binding, authoritative commands to the units (states) in the international system. These states are thus externally sovereign, in that no external actor can give them binding instructions. Though it need not follow logically, these states are also internally sovereign in that each has ultimate legal authority to issue binding commands within its territory. In this framework, then, there is a very large qualitative difference between interactions within a country and interactions across national boundaries.
As a result of this difference, Realists and many others argue that cooperation across international boundaries will be more difficult than cooperation within national boundaries. First, they argue, the international system lacks a legal order that can enforce TBC agreements, while domestic systems are characterized by such orders. This difference can easily be overstated, however. Compare for example the legal structure available to the European Commission if it wishes to force a member state to carry out an EU directive with that available to the NPS and USFS, should these agencies need to enforce an agreement between them. The Commission can, and does, bring legal action to compel implementation, and states almost always comply because they wish to retain the dense network of cooperation within the EU.Footnote 13 In contrast, by long-standing policy and (almost) law, the NPS and USFS do not bring suit against one another. It seems that the United States, and not the international system, is anarchic in such a case.
Second, many analysts argue that the international system relies on self-help, forcing each state to pursue military and economic power as a means to ensure its own security and continued existence. This makes international cooperation more difficult because states worry not only about the absolute gains from any agreement but also about each state's share of those gains. Certainly such concerns may play an important role in issue areas such as arms control. However, trade cooperation poses risks for security only under highly-constrained conditions,Footnote 14 and in an issue such as wildlife preservation the security externalities will almost always be vanishingly small. Some environmental issues, such as water usage, probably lie closer to the trade issue in that they plausibly have security externalities under some conditions. Other environmental issues lie closer to the wildlife issue area in generally lacking such externalities—though some transboundary cooperation across the US-Mexico border is clearly inhibited by the border fence, as is wildlife management in Israel.Footnote 15 Since security concerns matter only occasionally, this suggests that issue area, and not anarchy by itself, poses the key analytical distinction. Indeed, security analysts increasingly see their issue area as distinctive,Footnote 16 defining problems found both within the boundaries of nation-states as well as between them.Footnote 17 Any linkages between, say, trade and security need not be any different in principle than linkages between environmental regulations and economic competitiveness or any other two issue domains.
Anarchy does contribute to another challenge for international TBC, namely, the weaker institutional framework available to would-be cooperators at the international level.Footnote 18 Neo-liberal theorists of international relations have argued that institutions and institutional networks can make cooperation easier by reducing the transaction costs of future negotiations.Footnote 19 The argument is plausible enough, but clearly some regions such as the European Union are highly institutionalized. In contrast, weak federations such as Canada or Switzerland are much less centralized, and can in fact be compared to the EU on many dimensions.Footnote 20 The key variable across these cases is “institutionalization” and not “international/domestic,” and we should analyze the problem as such. Seen in this way, my approach is analogous to George Tsebelis' theory of veto players,Footnote 21 which does not distinguish between veto players in a single country, a federation, or an institution such as the European Union.
In addition to the problems of anarchy and sovereignty, we can imagine other reasons why international TBC might differ from national TBC. Some have argued that international environmental problems are characterized by much larger scales than domestic environmental problems.Footnote 22 This is usually true but it need not be. To take one salient comparison, the Greater Yellowstone Ecosystem is about 7.7 million hectares in size, while the three Benelux countries comprise only about 7.4 million hectares. Even if the scale of a problem is correlated with whether a problem is international, surely it would be better to study the variable “scale” directly instead of making causal claims about the variable “international.”
Finally, some have proposed that cultural heterogeneity makes international cooperation more difficult than intranational cooperation because of differences in understanding and communication.Footnote 23 Again, this is a plausible claim. My research design largely holds cultural heterogeneity constant by looking only at the US and Canada—two countries with significant cultural differences between them but also significant internal cultural differences from, say, British Columbia to Nunavut to Québec or Connecticut to Hawaii to Wyoming. In many cases the cultural differences within each country are mirrored in the other. Both have ethnically European urban environmentalists, conservative rural communities dependent on resource extraction, Native Americans/First Nations, and other such groups, each more important in some places than in others. Theoretically, cultural heterogeneity may be correlated with the presence of international boundaries but this is not a necessary relationship. Thus, it too should be studied directly instead of assuming a relationship with the variable “international.” Cultural heterogeneity can also be captured directly by expressing some aspects of these differences in terms of “preferences,” as in the theory here.
For each of these arguments, then, there is a plausible claim that scholars have misspecified the problem. Rather than international/domestic being the central causal variable, variation in issue area, scope, institutionalization, or preference heterogeneity may be more important. Each of these alternative arguments has some grounding in another literature working to break down the international/domestic divide, adding plausibility to the central claim here. However, none have tried to control for international boundaries in an empirical study such as the one here.
Theory of Transboundary Cooperation
To study wildlife TBC both within and across national boundaries, I focus on land-management agencies. Most of these, such as the NPS and USFS, are simply land managers. Other agencies, such as the US Fish and Wildlife Service (USFWS) only partly fit this definition—the USFWS is a land manager of national wildlife refuges (NWRs), but also a regulatory agency responsible for the endangered species act (ESA) and other legislation through which it can regulate other agencies' lands.Footnote 24
To simplify the analysis, I treat each agency as a unitary actor, though each land manager normally consists of a staff and not a single individual. This simplification is a pragmatic decision, not an ontological one. In addition, I treat preferences as exogenous though recognizing that each decision-maker responds to both internal and external pressures that shape her preferences over time. Again, this is a matter of pragmatism for the present article. Developing a theory connecting endogenous changes in preferences to variation in cooperation over time would provide a natural extension of the theory here.
Agencies follow a mandate, or set of rules, telling managers what objectives to pursue. Other goals also affect managers' decision, such as budget or staff maximization,Footnote 25 but I will not discuss those here because these latter goals are common to all agencies. If the goals do not vary, they cannot help explain variation in interagency cooperation.
In the narrower case of wildlife management, managers will have preferences over the number of animals of each managed species. They may also have preferences over natural processes such as predator-prey relationships, disease or pest processes, or wildfire management, but I will set these issues aside here.Footnote 26 Wildlife numbers may be important for a variety of reasons. State game management agencies, for example, manage game species to ensure a large harvest for hunters.Footnote 27 Managers of endangered species seek a population above some target threshold. Other land mangers want to reduce animal populations. For example, the state of Wyoming wants to reduce or eliminate politically controversial wolves and, to a lesser extent, grizzly bears. Most prairie states want fewer black-tailed prairie dogs, which their rancher constituents dislike. National park and forest managers occasionally reduce oversized ungulate populations.
Wildlife management problems only become transboundary wildlife management problems if the animals in question cross boundaries. Isolated populations of, say, boreal toads in the Central Rockies may not pose such issues, but larger animals certainly do. For TBC problems, we can reduce managers' possibly-complex preferences over populations to simple preferences over the wildlife that cross boundaries.Footnote 28 Any given manager will either want an increase in immigration into her unit, an increase in emigration out of her unit, or no change at all (this latter possibility includes managers who wish not to intervene at all). For example, Wyoming state land managers would generally like elk to move into their lands but wolves to move out of them, with both goals understandable in terms of ungulate harvest maximization and other political goals. For simplicity, I call managers who want an increase in animals moving into their land the “more” type, and those who want an decrease in animals moving into their land (equivalently, an increase in animals moving out) the “less” type. This captures well the situation for land units whose animals emigrate or immigrate permanently, since net emigration is the relevant variable.
For migratory species, where the same animals return, the induced preferences at the unit boundary are defined slightly differently. In this case, more-type preferences imply that the manager wants a larger population. This implies greater numbers of “emigrants” out of their unit and more “immigrants” into their unit. Managers with less-type preferences over migratory species simply want fewer animals.
Implicit in this discussion is the existence of two types of wildlife, “migratory” and “dispersing.” In a migratory relationship, the animals in one land unit move to another land unit for a season, and then return. In such a case, the two managers are jointly managing the population and, as I will discuss below, these common interests generally create favorable conditions for TBC.
In a dispersing (or source-sink) relationship, one particularly productive habitat (the “source”) regularly produces surplus animals. When the population exceeds the land's carrying capacity, surplus animals from this source disperse elsewhere in search of food or unoccupied breeding sites.Footnote 29 The “sink” habitats would not be productive enough to maintain their current population if they were fenced off from the world, but these sinks can maintain a stable population as long as they can receive a regular inflow of animals from adjacent sources.
Manager preferences and the relationship between two habitats for a given species interact in a straightforward way. Managers with identical goals may have either compatible or incompatible interests. Consider first the case of a dispersing species whose source and sink habitats are managed by different agencies. Immigration into one agency's unit may come at the expense of adjacent land managers who want to reduce emigration out of their unit (equivalently, increase immigration into this second unit). If both managers want more animals, they have a direct conflict of interest when each manager's desire for more comes at the expense of the other. TBC will be difficult or impossible. In the case of migratory species, however, joint management would increase the animals available to both land units. Increases in the summer breeding population will also increase the number of animals that return to the winter population, and increases in the winter population will also increase the summer population. In this case, more-type managers share a common interest in that the two managers' desire for more animals can be achieved jointly. TBC will be easier.
The outcomes in the other logical possibilities follow easily from the same analysis. If the managers of a migratory species' summer and winter habitats both want fewer animals, TBC will again be easy because the managers have the same preferences. Reducing the winter population will reduce the summer population, and vice versa. TBC will be hard for two less-type managers in a source-sink setting because each manager will want to encourage emigration into the other manager's unit. Preference homogeneity in a symmetric setting leads to cooperation while this same homogeneity does not lead to cooperation in an asymmetric setting.
When the managers have different preferences, cooperation is likely for dispersing species but not for migratory ones. In the migratory setting, the managers have directly opposed interests and TBC will not occur. In a source-sink setting, dissimilar preferences can make TBC possible. A less-type manager of a source habitat who wants to encourage emigration has the same interests as an adjacent sink manager with more-type who wants to encourage immigration, so TBC should occur. However, if the source manager has more-type preferences and wants to discourage emigration, there is little scope for TBC with an adjacent sink manager with less-type preferences who wants to encourage emigration because the sink does not have surplus animals to disperse.
Table 1 summarizes the predictions across all logical possibilities. As a rule of thumb, homogenous preferences facilitate cooperation in symmetric settings, while heterogenous preferences facilitate cooperation in at least some asymmetric settings. With some complications, this should generalize to other issue areas, as discussed below.
Incentives for transboundary cooperation
* TBC if the source wants an increased emigration into the sink but not if the source wants a decrease in emigration.
The theory makes no reference to the nature of the units. Whether land managers are provincial parks, national forests, Indian reservations, or the Department of Defense should make no difference whatsoever. Though these agencies have a variety of goals, the TBC problem depends more narrowly on their preferences over animals at the border of their lands, which collapse into more-type and less-type goals. Decision-makers in many environmental issues could be characterized similarly as seeking more clean water, less air pollution, with or without additional goals such as employment or economic development.
Because I assume that each habitat has a unique land manager, I neglect the interests of “overlay” wildlife managers. I have already mentioned the USFWS overlay role as executive of the ESA, and Canada's federal government plays an analogous “overlay” role over the Crown lands managed by the provinces.Footnote 30 Cooperation between a land manager and an overlay manager, such as cooperation between the USFWS and Indian tribes,Footnote 31 represents a different analytical problem than one between two land managers.
Second, the theory concerns only the presence or absence of cooperation, and not the policy content of any agreement. Some successful agreements may be shallow, or may not manage the TBC problem very well. Moreover, agreements divide the gains from cooperation in many possible ways.Footnote 32 As is well known in international relations, asymmetric information about preferences or about the issue at hand, such as wildlife numbers in this case, can affect the allocation of benefits from any agreement. Clearly the content of agreements is important, but it is much more difficult to compare this content systematically in the wildlife context—as opposed to, say, trade policy where tariffs can be measured in a fairly straightforward way.
A third point also requires clarification to define the scope of transboundary cooperation as defined here. Given the theory here, I look only at policy change and not information sharing. The literature on international cooperation regularly maintains that information exchange is a relatively low form of cooperation, and the environmental literature on TBC tends to agree.Footnote 33 More important, information-sharing has a somewhat different logic than the policy cooperation here.Footnote 34 For example, a less-type source and more-type sink would be able to exchange information even though policy coordination would be difficult or irrelevant. Building collective databases or conducting joint research projects also present a different analytical problem that may be described as joint public good provision subject to economies of scale.Footnote 35 Analyzing such problems requires an extension of the theory in future research.
Fourth, this simple theory also contributes to our understanding of environmental TBC. Professional biologists and managers regularly lament the lack of transboundary coordination in many areas, typically calling for greater goodwill, less “divisiveness,” efforts to develop a shared vision, or greater political will in order to improve TBC.Footnote 36 Only a few scholars have identified agency mandates as an important effect on the success or failure of TBC.Footnote 37 Even those that have identified mandates have not teased out the general pattern examined here.
Finally, this specific theory of wildlife cooperation is embedded in a large approach to cooperation based on the homogeneity or heterogeneity of actor preferences and the symmetry or asymmetry of the problems that they face. Realism and its relations see a world of like units with homogenous preferences facing identical (or symmetrical) problems of security. Clearly, this characterizes many security problems but it does not exhaust all international relations. In trade policy, in contrast, nation-states and the agencies that work on their behalf seek some combination of goals such as economic wealth, domestic political support, and foreign policy goals. Countries with a common interest in lower tariffs generally lower them reciprocally, a situation seen most easily in the case of small, industrial market economies. Yet countries that share general interests may interact in asymmetric settings, as when one state is a hegemon. Similarly, two countries whose firms have oligopolistic positions in a given market face the same asymmetric problem as the dispersing species problem in the wildlife theory, where one actor's gain comes at the expense of the other. In these examples, actors with similar preferences cooperate when facing a symmetric problem, but find cooperation difficult when the policy problem is asymmetric.
In short, the theory develops a straightforward explanation of variation in TBC that depends solely on preferences and species type without reference to whether these problems of cooperation are national or international. Evaluating these claims requires that we consider both the theoretical explanation of TBC and the assertion that the nature of the units does not explain TBC. This is the task of the next two sections.
Research Design
I will test two sets of claims based on the earlier analysis, both the theoretical explanation of TBC and my claim that the nature of the unit does not affect whether TBC occurs. Ideally, we would have a database of all cases of possible TBC. This is currently unrealistic,Footnote 38 and it is even more difficult to identify cases of TBC that should occur but have not.
Failing this, I use comparative case studies chosen by a geographically-limited decision rule. Limiting the geographical scope of the study constrains my discretion, reducing potential researcher bias. At the international level, I focus on the US-Canadian issues from Montana westwards. Comparable domestic problems around the Greater Yellowstone Area (GYA) provide non-international controls. The species that live in this area enjoy national and international reputations, making them more politically salient than other possible study areas. I focus on mammals and birds that enjoy greater political salience.Footnote 39 Political salience makes these cases more researchable, and is not correlated with the variables in the theory so it does not introduce bias. To fill an otherwise empty cell, I include one case outside the geographic constraint, the monarch butterfly in the US and Mexico.
For testing my claim about the non-importance of the nature of the unit, the United States and Canada provide excellent variation across all levels of government. They are strongly federal systems that also have significant First Nations (Indian tribes) playing roles in the wildlife management issue area.Footnote 40 I limit attention to government agencies only. Private companies and non-governmental organizations play an important role in some transboundary collaboration, most famously in the Quincy Library Group,Footnote 41 but further research will be necessary to determine whether they can be brought under the same theoretical umbrella as the agencies analyzed here.
The cases here are largely illustrative, and intended to show the causal mechanisms at work in explaining variation in TBC. As appropriate in such cases,Footnote 42 I select cases to provide variation in the independent variables of species type, manager goals, and nature of the units. By taking all cases from a restricted geographical domain, I minimize the possibility of biased sampling. The cases are not randomly chosen, but the choices were highly constrained by this geographical rule, which is itself unrelated to the theoretical variables of interest.
Determining whether agency goals are “same” or “different” can pose some challenges. The theory does not require that we compare the agencies' actual mandates, only their preferences over transboundary migration. Agencies that want a larger population of the target species want more immigration (or, equivalently, less emigration), and will therefore have more-type preferences. This will be true even for agencies that manage hunting on their lands, since increased immigration makes possible a larger harvest. Most of the land management agencies in the region analyzed here have these “more” type preferences in that they want more animals than they currently have.Footnote 43 Only a few agencies want smaller wildlife populations. For example, some agencies that manage severely overpopulated ungulates want to reduce their numbers. Other agencies seek to reduce or eliminate politically undesirable species such as wolves, prairie dogs, and sometimes grizzly bears, among others.
Table 2 shows the relevant independent variables (manager goals, species type, and nature of the unit), and the prediction of the theory for each combination of variables, listing those TBC cases that I will discuss in the next section. Because less-type preferences are unusual for most managers, the international cases in which managers have different goals are weaker fits—the prairie dog case is more latent than manifest as an international issue, and the butterfly case falls outside the geographical constraint. This does not affect the test of TBC predictions across all manager and species types, since I have a full set of good cases at the domestic level. The claims about the nature of the units depend on comparisons that include these weak cases, though the conclusions would remain the same if one or both of these rows were dropped. As a result, this set of cases suffices to demonstrate that variation depends on the variables found in the theory and not on the nature of the units.
Variation in transboundary cooperation: Cases
* Requires that source managers want to reduce dispersion of bison, wolves, and grizzlies for political reasons.
Case Studies
To examine the theoretical claims, this section reviews those cases chosen previously. Each case describes the manager goals, species type, and whether the outcome is consistent with the theory. For most cases, I also briefly consider some alternative explanations as a form of control. Taken as a whole, the cases clearly demonstrate that the theoretical variables, and not the nature of the units, determine outcomes.
One matter of operationalization warrants mention. In some cases, there are trade-offs between mangers' preferences over animal numbers and their preferences over processes. For example, managers might prefer to leave a species alone, relying on “natural regulation” or other biological processes, and yet recognize that some human intervention will be required to make natural processes work correctly. In the simple theory and brief case studies here, I reduce such complications to a simple matter of whether the manager wants more or fewer animals compared to the status quo.
International, Dispersing Species, Homogenous Goals: No Cooperation
The US-Canadian border is characterized by a large number of source-sink relationships. Several highly visible mammals were extirpated, or nearly so, south of the border—including wolves, grizzly (brown) bears, Canada lynx, and others possibly including the wolverine. In most of these cases, the animal remains fairly abundant in much of Canada. Some Canadian animals have dispersed into the United States over the last few decades, establishing foothold populations in the Northern Rockies and North Cascades. The current situation seems to recapitulate the successful dispersion of gray wolves out of the Quetico-Boundary Waters source into the upper Midwest, where the Wisconsin and some other populations remain sinks.Footnote 44
In each of these cases, US recovery efforts would benefit from TBC. If Canada were to manage border areas to increase animal production, for example, more animals would disperse into the United States. Useful forms of cooperation could be as simple as prohibiting hunting in Canadian lands adjacent to targeted US regions, with Canadian hunters directed farther north and away from the border. As it stands, hunting is sometimes prohibited in border districts for game management or public safety reasons, as in E.C. Manning Provincial Park, but these decisions reflect provincial goals and not any cooperation across boundaries. When populations increase to a level that will allow sustainable hunting, the provinces generally allow hunting regardless of the implications for the US population. For example, British Columbia allows hunting and trapping of wolves, wolverines, cougars, and black bears, among other species, most of which are of interest to adjacent US land managers.
The United States and Canada do not cooperate in any way to manage these populations jointly. All the agencies involved have the same goals, an influx of animals into their lands. The reasons differ—American land managers seek species recovery, while Canadian managers generally want a larger game animal population base for hunters in the most populous southern parts of the country. Even for non-game animals, none of the Canadian mangers have goals involving increased emigration out of their land units. As a result, cooperative land management does not occur.
Interestingly, the lack of cooperation does not depend on whether the population source is located in the United States or Canada. For example, grizzly bears around Glacier-Waterton International Peace Park may better be described as having a US-based source in Glacier National Park with population sinks in Waterton Lakes National Park and some adjacent areas in Alberta and southeast British Columbia. When population levels permit, Alberta has allowed grizzly hunting, which keeps its population below a sustainable level, dependent on immigration from the US source.Footnote 45
Consistent with my claim that information-sharing differs from policy cooperation, various agencies do exchange information about some listed species. For example, Canada's Rocky Mountain Grizzly Bear Planning Committee (RMGBPC), formed in 1997, makes advisory recommendations to wildlife ministers in British Columbia and Alberta, and to the federal minister responsible for the national parks.Footnote 46 The RMGBPC includes some Americans who provide information to Canadians, but they have no legal standing in either US or Canadian policy making. This simple exchange of information falls outside the definition of TBC used here.
The US and Canada also cooperate in many environmental areas that do not involve land management.Footnote 47 Many treaties and agreements affect wildlife, such as the agreements concerning the Ross High Dam on the Skagit River, but do not regulate wildlife directly. Others concern migratory species, and are discussed later in this section. A few others govern endangered species (“species at risk” in Canada) but do not require changes in habitat management. For example, the 1973 Convention on Trade in Endangered Species (CITES) prevents trade in listed animals, furs, and other parts. Cooperation may occur in law enforcement at the border, but CITES does not require either country to change how it manages public lands. The two countries also signed a “Framework for Cooperation between the US Department of the Interior and Environment Canada in the Protection and Recovery of Wild Species at Risk” in 1997. This provides a legal framework for future cooperation, though it has not yet led to any such cooperation within the domain here.
Domestic, Dispersing Species, Homogenous Goals: No Cooperation
Bordering both Alberta and Glacier National Park, the Blackfeet Indian Reservation also manages significant numbers of wildlife. Blackfeet Indian Reservation is home to the largest of the three tribes of the Blackfeet Nation, alongside two small tribes in Alberta. Like Glacier National Park, Blackfeet Indian Reservation lies at the intersection of prairie and montane regions, and it is home to many species familiar to each ecosystem. These species include dispersers such as grizzly bears, wolves, and swift foxes.
The tribal government's wildlife management goals vary. In the case of game animals, it seeks first to provide hunting opportunities for tribal members, and then to earn revenue from trophy hunting by outsiders. For these species, Blackfeet Indian Reservation seeks increased immigration, most of which comes from Glacier National Park, making this a source-sink relationship. Its interests in non-game animals depend on both cultural and economic motives. The endangered swift fox, like black-footed ferrets and black-tailed prairie dogs, is culturally important to the Blackfeet Nation. This cultural interest has been supplemented by economic interests as Blackfeet Indian Reservation has received grants from government and non-government sources for swift fox recovery programs, revenue that helps support the tribal fish and wildlife department.
The reservation's neighbors also seek larger populations of the animals considered here, especially listed (or once-listed) species such as grizzlies and wolves. More typically, both tribal members and fee-paying outsiders cluster near the boundary in the hunting season to shoot animals from Glacier National Park as they exit the park. A hunting brochure for the Nation makes this connection explicit for the Boulder Flat Top area, where the “abundant wildlife of Glacier Park wander out onto the mountain front and prairies of the Reservation, making this area [a] prime hunting destination.”Footnote 48 The most lucrative species is grizzly bear, for which the tribe opened a non-member trophy season in Fall 2006, charging $50,000 per license.Footnote 49
There would be a clear benefit from TBC, since these species are quite wide-ranging and require large regions to support a sustainable population. Despite this, the reservation and the national park find it very difficult to cooperate on joint management of anything.Footnote 50 Only after years of disagreement have they even reached a modus operandi in presenting the history and culture of the region. Thus, an alternative explanation would be the poor relations between many Indian reservations and other units of government, reflecting a long and tragic history. We will see below that TBC does occur between federal agencies and tribal governments, however, so the real task is to explain variation and not the admittedly low average level of cooperation. Some of this variation depends on internal tribal politics, especially the common alternation in power of traditionalists and modernizers.Footnote 51
These issues of internal tribal politics have been more important for the issue of swift fox reintroduction. Because there are not yet any adjacent swift fox populations, TBC is not an issue for that species, but here too the tribe would seek increased immigration of foxes. Interestingly for the argument here, the three Blackfeet units have been unable to cooperate in swift fox reintroduction. With support from the Cochrane Ecological Institute and Defenders of Wildlife, the US tribe ran a reintroduction program from 1998 to 2002. The Canadian reserves are much more densely populated than the US reservation, and earn more income from agriculture and forestry. The Blood (Kainai) tribe was not interested in participating until 2005, when they worked separately with NGOs instead of joining the Blackfeet Indian Reservation to the south.Footnote 52 The North Peigan (Canada) tribe was unable to reach a decision either way despite support from tribal elder Joe Crowfoot, who died in 1998. Thus, intertribal TBC on swift foxes proved impossible as well, despite the likelihood that a sustainable population on the South Peigan (US) lands could have served as a source supporting reintroductions across the border in Canada.
These failures of TBC among the constituent members of the Blackfeet Nation, consistent with the theory here, suggests that the history of mistrust between Blackfeet and white governments is not a necessary condition for non-cooperation. In addition, I will show below that occasional examples of successful TBC occur when the theory predicts that they should.
Domestic, Dispersing Species, Heterogenous Goals: Outcome Varies
As mentioned several times in this article, less-type preferences are fairly uncommon because most land managers in the study area either seek larger animal populations or wish to let natural processes stabilize the population at higher levels than have been common since European contact. The only significant exception to this generalization consists of a group of controversial species, mostly listed as endangered or threatened. In some of these cases, state land management agencies actively reduce the population, by poisoning black-tailed prairie dogs or shooting coyotes or blackbirds. In the Northern Rockies, gray wolves, grizzly bears, and bison have been the most controversial species—wolves because of a risk to livestock, bears because of seasonal livestock risks and potential danger to humans, and bison because they can infect domestic cattle with brucellosis.Footnote 53 State and provincial agencies generally side with livestock owners and wish to extirpate these species from jurisdictions under their control.
In contrast, adjacent federal land-management agencies generally want larger animal populations even in the case of controversial species. In a source-sink setting, these differences of goals between federal and state agencies can facilitate transboundary cooperation that would not occur if the agencies had similar preferences in a source-sink setting.
Bison in Yellowstone National Park (YNP) present an interesting case, though difficult to classify. Bison are essentially nomadic, though terrain tends to cause recurring patterns of seasonal movement over time, varying in part with the weather. For our purposes, we can characterize the population as a mixture of “migration” from higher to lower elevations each year and “source-sink” relationships as a healthy bison population tries to disperse outside the park. Further complications ensue because some of the herd remains in YNP while some does not, so TBC issues do not always arise. Behavioral differences between herds and solitary bachelor bulls complicate matters further. When boundaries do stretch across the bison herd's movements, the park's neighbors tend to oppose the entry of bison onto their land, fearing that bison will spread brucellosis to grazing cattle, making live animals largely unsaleable. The NPS wants a healthy bison herd, but to minimize local political opposition to the park it regularly restricts bison emigration out of the park. As a result, YNP's neighbors have less-type preferences over both dispersing and migratory bison. The park itself has more-type preferences over all bison, wanting migratory bison to return each summer while it wants would-be dispersers to remain in the park.
TBC has been very difficult in these settings but the relevant agencies did reach agreement in 2002.Footnote 54 This agreement sought to reduce the brucellosis threat with a mix of techniques: hazing bison to keep them away from cattle grazing areas, fencing some areas, vaccinating some bison, physically moving some bison, and “management-related mortality” of several hundred bison a year.Footnote 55 Each agency sacrificed important parts of its management goals in order to achieve a cooperative solution to the problem. After regional tribes successfully asserted old treaty rights to hunt bison, the state of Montana also opened up a limited hunt for both Indians and non-Indians.Footnote 56
To the south, Jackson Hole bison have posed similar issues for Grand Teton National Park (GTNP) and the National Elk Refuge (NER). A small herd lives in GTNP, in an area with some private inholdings and adjacent to ranches that depend on livestock or tourism for income. The herd has discovered supplemental elk feeding on the NER, and now migrates to take advantage of this winter fodder. As the herd has grown, it has also dispersed into adjacent areas, include the town of Jackson itself. Dispersing bison are controlled or eliminated, so the TBC problem essentially concerns migration between the NER and GTNP.
The NER has traditionally wanted a lower bison population because it defines its mandate largely in terms of elk. The Wyoming Department of Game and Fish (WDGF) also wants fewer bison, to prevent the spread of brucellosis. GTNP wants a sustainable herd but also wants a herd dependent on natural processes and not on supplemental feeding. Achieving “natural processes” requires a change in policy that will effectively lower the size of the herd, giving the park less-type preferences. These preferences create the possibility of TBC by allowing the park to accept a smaller herd. The ultimate result was a 1997 interagency plan, which provides for a self-sustainable population, reducing supplemental feeding, and minimizing contacts between bison and domestic livestock.Footnote 57
Aside from the analysis here, there do not really exist explanations for this cooperation. It stands out as rather contrary to the usual recommendations of conservation biologists to manage habitat for larger populations. Yet, ironically, these agencies have succeeded in this controversial case but have been unable to achieve TBC in many less-controversial species. The theory here provides a plausible explanation for this variation.
International, Migratory Species, Homogenous Goals: Cooperation
In contrast to the international source-sink problem discussed above, management of migratory species has been fairly easy across international boundaries. The oldest Canadian-American wildlife treaty still in effect is the 1916 Migratory Bird Treaty (MBT), signed during a Progressive era that also saw the 1908 Inland Fisheries Treaty and 1911 North Pacific Fur Seal Convention.Footnote 58 The MBT addressed the problem that hunters killed migratory birds along their flyways and in their habitats at either end, thus reducing the total population.
The MBT has provided a framework under which US and Canadian land mangers acquire land and manage existing units to further the goals of the act. The treaty's implementing legislation in the United States, the Migratory Bird Conservation Act (MBCA) of 1929, helped provide the foundation for a system of national wildlife refuges to protect migratory waterfowl. Presidents have used their executive powers under the MBT and MBCA to establish many NWRs on international flyways, and migratory birds dominate the US National Wildlife Refuge System.Footnote 59 Cooperation with Canada in the MBT proved to be essential for US policy because the constitution's interstate commerce clause proved to be a weaker legal instrument for wildlife preservation than the president's treaty powers that lay behind the MBT.
Standard neo-liberal theories of international cooperation would seem to explain the MBT case well; it is symmetric and both parties are better off cooperating than not cooperating. There is some risk from defection if one country fears the other will harvest all the animals first. The theory here provides a similar explanation, one that is more satisfactory in that it also explains cases of non-cooperation, such as international source-sink problems. It can also be extended to other, non-international problems, as in the next section.
Domestic, Migratory Species, Homogenous Goals: Cooperation
In contrast to the politically controversial species discussed in several cases here, elk are politically popular. For that reason, elk managers generally have more-type preferences, though on some overpopulated ranges managers would like elk to disperse to adjacent lands. As a migratory species that moves between winter and summer ranges in the GYA, we would expect successful TBC as all managers have similar interests. Again, there is a small complication in that many elk remain inside YNP for the winter, but the herd in the south of the GYA all migrates to the NER and poses a TBC problem.
Such TBC occurs in several fora. The Commission on the Conservation of Elk of Jackson Hole, Wyoming, ran from 1927 to 1935, including both government agencies and non-government participants. After two decades of less formal coordination, the agencies formed the Jackson Hole Cooperative Elk Studies Group in 1958, which is still in operation.Footnote 60
At the ecosystem level, the Greater Yellowstone Coordinating Committee (GYCC) consists of the relevant federal land managers in the GYA, as well as the relevant regional directors of the NPS and USFS (that is, the land managers' bosses). Since its founding in 1964, it has met twice yearly to discuss policy coordination, share information, and agree on action. Success has varied significantly by species, but elk management (along with cutthroat trout and bald eagles) is one of its claimed success stories. The GYCC's Greater Yellowstone Elk Working Group and its predecessors have overseen a quadrupling of the elk herd in the region.
That very success has led to widespread concern that there are now too many elk, and some agency preferences have moved toward less-type preferences instead of more-type preferences. Though the transition has sometimes been difficult because different agencies changed their interests at different paces, renewed TBC has resulted, focused around a newer set of goals. The USFWS and WDGF drew up an agreement in 1975 to manage a smaller herd. WDGF was responsible for reducing the herd to 7,500 through hunting licenses, while the USFWS's NER would continue to provide supplemental feeding in a more dispersed forage system.Footnote 61 This agreement did not reduce the herd, which grew as large as 11,000 head. This failure largely reflects the fact that the two agencies had different goals. The WDGF did not really want to reduce the herd but to increase hunting opportunities outside the NER. The agreement also failed to resolve the key underlying issue, that supplemental winter feeding on the NER provides a breeding ground for brucellosis, which is found at only very low levels in other elk herds. To force a better solution of this issue, the NER denied WDGF permission to vaccinate the herd on NER land in the winter of 1998.
Cooperation among federal agencies, in contrast to state-federal cooperation, has proceeded apace. In recent years the USFWS and NPS have drawn up a new plan for managing elk on the NER and the adjacent GTNP.Footnote 62 Interestingly for the theory here, both agencies' preferences have changed to less-type preferences, in that each would like to see a smaller elk herd. The current proposal involves a modest reduction of the herd by reducing supplemental feeding on the refuge and changing various practices elsewhere in the ecosystem. Since the agencies continue to have the same preferences as one another, TBC has continued through the shift in preferences over time. When compared to cases of unsuccessful elk TBC such as Glacier-Blackfeet, the variation in this case adds plausibility to the overall analysis here.
Domestic, Migratory Species, Heterogenous Goals: No Cooperation
The eastern half of Glacier National Park was purchased from the Blackfeet Indians in 1895, though the tribe reserved hunting, fishing and timber rights as long as the land remained “public.” When these lands became a national park in 1910, the US government maintained that they were no longer “public” because they had been withdrawn from public registry and sale. The Blackfeet never accepted this claim, and continued to hunt in what was now a national park.Footnote 63 The meat long remained important to winter survival, and hunting continued despite regular prosecutions for poaching.
In this environment it is not surprising that the tribe and NPS found it very difficult to cooperate on elk management. In terms of the theory here, the outcome depended on different preferences—the tribe wanted a smaller population moving from winter grounds on the reservation to summer grounds in the park, while the NPS wanted a larger population in both seasonal movements.
The elimination of predators such as wolves and grizzly bears in the 1930s ultimately changed NPS preferences. Eventually, elk overpopulated the range and caused significant damage to other park resources. The NPS decided to reduce the fall herd. This gave the park the same interest as the Blackfeet, who wanted to harvest this herd. The peak of cooperation came in 1962, when the park hired four tribal members as seasonal rangers to cull elk. These seasonal rangers “accidentally” harassed a large number of elk across the park boundary, where a group of Blackfeet hunters happened to be waiting. After the hunters killed 86 elk that day, a shadow fell over future cooperation. The explanation for subsequent non-cooperation makes sense in terms of both neoliberal theory and the theory here: with a small shadow of the future, the Blackfeet had an incentive to defect, sparking NPS retaliation in the form of future non-cooperation.
By the 1980s, native predators began to return to the region. As elk numbers returned to a naturally regulated level, the incentives for TBC again disappeared. The interaction between preferences and habitat type neatly explains the variation in this case.
International, Migratory Species, Heterogenous Goals: No Cooperation
There are apparently few cases in which managers have different goals over a migratory species that crosses international boundaries. Though it lies outside the US-Canada domain of the other cases, the case of monarch butterflies in the US and Mexico fills an otherwise-empty cell in table 2.Footnote 64 Monarchs are multigenerational migrants.Footnote 65 They concentrate in a few locations at certain times, such as overwintering sites along the California coast and oyamel-pine forests in central Mexico.
Land managers' goals differ considerably. US public land managers generally want more of these butterflies. In contrast, much of the Mexican forest is owned by collectives (ejidos), many of which want to harvest the high-value forest regardless of the effects on the butterflies. Some landowners and non-governmental organizations take action on behalf of the butterflies in both countries, while others do not. International cooperation has so far consisted only of an international symposium held in November 1997, which reached the conclusion that the needs of butterflies, loggers, and local people were in conflict.Footnote 66 This conference achieved, at most, an exchange of information. Beyond this, conservation actions are entirely local and there has been no TBC to protect habitats. This is consistent with the theory.
International, Dispersing Species, Heterogenous Goals: Cooperation
The geographic domain lacks a good international case of a dispersing species over which managers across an international boundary have different preferences. The lack of cases reflects the fact that less-type preferences tend be uncommon. When free to look across many domestic problems, I did find such cases, but when constrained by the necessity of finding problems across an international boundary these cases are rare.
The prairie dog complex consisting of black-tailed prairie dogs and black-footed ferrets provides a partial example. It is primarily domestic but has some international spillover. Prairie dogs live in complexes of burrows, called towns. Larger collections of towns make up a complex. In terms of the population model here, they are dispersers. They are the dominant prey for the highly endangered black-footed ferret, which cannot survive without prairie dog towns.Footnote 67 Most ranchers believe that prairie dogs compete with cattle for forage, though this claim is challenged by some. It is certainly true that cattle may injure their legs by tripping in prairie dog burrows, and then need to be destroyed. To protect their herds, ranchers favor poisoning prairie dogs, and this deprives black-footed ferrets of their prey.
Managers' goals vary considerably. Badlands, Theodore Roosevelt, and Wind Cave National Parks in the United States, like Grasslands National Park in Canada, generally protect prairie dog complexes and have more-type preferences. Grasslands National Park borders non-park land in the United States owned by the Bureau of Land Management (BLM), the state of Montana, and private ranchers. Thus, this part of the Saskatchewan-Montana border represents a potential case of international TBC. Canadians have proposed an International Peace Park at this location, on the model of Glacier-Waterton International Peace Park, but both US and Canadian ranchers are strongly opposed. States and provinces largely side with the ranchers, as does the BLM. I have not found any examples of cooperative management among these units across the international boundary, and there is also no cooperative management within national boundaries except for actions required under the ESA (which falls outside TBC as defined here). Thus, the presence or absence of an international border does not affect the absence of cooperation in this case. Consistent with the theory, the presence of an international boundary does not affect the TBC problem.
Conclusions
I have analyzed international and interagency cooperation in tandem to explain variation in cooperation on wildlife issues. I have focused narrowly on the presence or absence of TBC, as opposed to the substance of such cooperation. Certainly the details of elk and bison TBC in the Greater Yellowstone Ecosystem have attracted significant controversy on all sides, but that is a question for a future article. In addition, I have limited attention to government actors. Public-private partnerships such as the Quincy Library Group became increasingly important in the 1990s, but are excluded from the analysis here.
Variation in decision rules and species type point to a poorly-understood issue in the theory of international cooperation, and by extension also in the analysis of interagency coordination: variation between symmetric and asymmetric settings. Agencies managing migratory animals face a symmetric problem as both must work together to guarantee survival of the migrating population. When one agency manages a source population of a dispersing species, and another agency manages a sink, the situation is asymmetric.
Such asymmetries are ubiquitous in international environmental problems. Acid rain, for example, originates in the US and Germany but affects downwind forests and water in Canada and Scandinavia. Similarly, countries that share a lake face a symmetric problem while upstream and downstream countries in a river basin face an asymmetric problem. Extending the analysis here, managers with similar goals should find it easy to manage a lake together but difficult to manage a river together. By examining issue areas that vary in terms of symmetry and asymmetry, it would be easy to extend the analysis here to non-wildlife problems.
Moving beyond environmental issues, my central claim is that international cooperation does not occur between “states,” or even between “executives constrained by legislatures,” but between agencies. Empirically, the theory shows that interagency coordination and international cooperation represent similar processes and should be analyzed similarly. I have showed this with reference to cooperation in wildlife habitat management both within and across national boundaries. Variation in successful TBC reflects not the nature of the boundary, whether national, state/provincial, tribal or executive agency, but the nature of the wildlife populations being managed and the decision rules of the agencies managing them.
The analysis here assumes that we know what agency goals are, and that they remain constant during the period of analysis. I have not tried to explain where these mandates come from. Agency preferences will vary over time in response to presidential, congressional, and interest group pressure. Analysis of this problem would fit well with analytical approaches such as neoliberal institutionalism or constructivism, which tend to be agnostic about preferences. Realism, which posits homogenous preferences among security (or power) seeking states, would have greater difficulties incorporating these agencies into the theory.
Of course, no one study based on a specific mid-level theory of a single issue area can possibly overturn a large, state-centric literature. By choosing an issue that scales up and down across multiple levels of government, and by presenting a theory that uses simple notions of symmetry and asymmetry that are found in many other issue areas, I hope to have made the case that the central claim is plausible—plausible enough for future research using similar controls and analogous theories.
Furthermore, the argument contributes to a wider agenda in the field of international relations, one more oriented toward the politics of subnational units. Where once scholars saw a Realist world in which life was nasty, brutish, and short, much of the field of international relations now sees a world of cooperation, international regimes, communities, laws and norms. In this new world, international relations is not limited to sovereign states in an anarchic world but includes, say, the foreign policies of US governorsFootnote 68 and many other actors. The focus on agencies in this paper fits well with this tendency in the field.
Realists would object strongly. They would doubtless insist that the problems of anarchy and the resulting need for self-help in order to maintain a state's security make international relations fundamentally different from interagency cooperation in domestic politics. As discussed above, this objection rests on the nature of the security issue, and not on the nature of the relevant unit. This objection essentially grants my central point. The question is not whether defense policy is different from employment policy. Security problems may threaten the existence of the nation-state, so the military and some ministries rightly treat these very seriously. Some environmental policies, such as Israel's water management, have security implications. In those cases when trade or environmental policies have security implications, analysts should state explicitly the conditions under which those security concerns will or will not affect policy in other issue areas. Under normal conditions, however, there is no reason to assume that trade, transportation, or environmental ministries take the security problem as primary at all times. Conducting comparative tests within and across national boundaries, such as the case studies here, can help clarify the conditions under which security problems interact with other issues, as issues
The central question examined here is whether agencies that manage foreign policy are qualitatively different than agencies that manage domestic policy. I deny this latter point. If we look more closely at agencies as actors in international relations, many of our assumptions about the nature of international politics may not survive close scrutiny.
When Allison saw bureaucracies as central players, he emphasized how these agencies struggled over control of foreign policy in a single state. I would drop the “foreign policy” step entirely, and examine the relations between bureaucracies of different countries directly. It is certainly an open question whether US trade negotiations with China are affected by input from the Defense Department; if not, then looking solely at the US Trade Representative and China's trade ministry should suffice. If the Defense Department influences trade negotiations, it does so through an interagency policy process that we need to theorize about explicitly. Similarly, international interactions in European Union fora such as COREPER or European Policy Cooperation are essentially also interagency relations.Footnote 69
An alternative, equally domestic approach to cooperation would focus not on agencies' supply of cooperation but on societal actors that demand such cooperation. Many environmental groups organize to demand TBC because most public land units do not follow ecological boundaries. For example, the International Sonoran Desert Alliance and Yellowstone-to-Yukon Initiative lobby for transboundary preservation.Footnote 70 Such lobbying can affect the relevant land-management agencies, so a focus on agencies is probably best.
All such research points toward a need to break down the boundaries between “international” and “domestic” politics. Such a recasting of the field would even find resonance in security studies. For example, once the international/domestic divide is set aside, we can acknowledge that the study of civil war belongs solidly in the field of conflict studies, with no need for excuses such as the prevalence of “foreign” intervention in many civil wars. The field of EU studies has already taken significant steps to analyze “international relations” as consisting of domestic, national, and supranational actors.Footnote 71 The rest of the field can do likewise.
