Level Two Negotiations: Helping the Other Side Meet Its “Behind‐the‐Table” Challenges

Early in his career, former New York City Police detective and hostage negotiator Dominick Misino faced a potentially explosive situation. On a sweltering summer night in Spanish Harlem, three hundred to four hundred people stood outside a crowded tenement building in which a young man with a loaded shotgun had barricaded himself. During the tense negotiations with Misino, the young man, a parole violator but not a murderer, told Misino that he wanted to surrender but could not because he would look weak. According to Misino (2002: 54),

This story illustrates a simple lesson of how a savvy negotiator helped his counterpart save face with an important constituency in a potentially lethal situation. In settings from labor relations to high‐level diplomacy, many negotiations display more complex versions of this same underlying structure: you (in this example: Misino) negotiate “externally” with your counterpart (here: José) who must somehow deal effectively with his or her “internal” constituencies (here: the crowd) — in order for you to be successful (here: to take José into custody, avoid a shootout, bloodshed, and wider risks to the police, crowd, and neighborhood).

A long analytic tradition has explored the challenge of productively synchronizing “internal” with “external” negotiations, with a particular focus on how each side can best manage internal opposition to agreements negotiated “at the table” (see, e.g., Walton and McKersie 1965 or chapter 17 of Lax and Sebenius 1986). Often implicit in much of this work is the view that each side’s leadership is best positioned to manage its own internal conflicts. Traditionally, a negotiator does this by pressing for deal terms that will overcome internal objections and by effectively “selling” the agreement to key constituencies. The ways that one side can help the other side with the other’s “behind‐the‐table” barriers (and vice versa) has been territory much less traveled by negotiation analysts.

At a minimum, helping the people on the other side calls for “putting yourself in their shoes” to understand their interests. Standard negotiating advice normally suggests using this mutual understanding to help devise a creative agreement that meets their interests while satisfying yours. Going beyond what are commonly viewed as the interests of the at‐the‐table negotiators, I argue that you should explicitly understand the other party’s “interests” to include how they could deal effectively with their internal, behind‐the‐table challenges (and vice versa).² This requires deeply probing the context in which they are enmeshed: the web of favorable and opposing constituencies as well as their relationships, perceptions, sensitivities, and interests. This understanding can be developed in a number of ways.

For example, U.S. Ambassador Stuart Eizenstat negotiated with Germany in the mid‐1990s over Holocaust‐era assets and compensation for slave labor used by the Nazis and German firms. Key issues in this tense, emotional process included how much survivors should be compensated and how to achieve “legal peace,” which would have ended any further claims against German companies following an agreement. Eizenstat and his German counterpart, Count Otto Lambsdorff, had known each other for many years and cultivated a relationship that meant, in Eizenstat’s (2003) words, that “we were able to share confidences with each other. We were able to share with each other what our constituencies were pressing us to do.”

In this article, I characterize “behind‐the‐table” or “Level Two” (Putnam 1988) barriers more broadly, provide several examples of how each side can innovatively help the other overcome these obstacles, and offer advice on doing so most effectively. To provide a fuller illustration of how a Level Two negotiator can help the other side manage formidable behind‐the‐table challenges, I pay special attention to the multifaceted approach taken by former U.S. Secretary of State James Baker who, working closely with President George H. W. Bush, played a key role in the end‐of‐Cold‐War negotiations with the then‐Soviet Union over German reunification within the North Atlantic Treaty Organization (NATO).

The term “Level Two” comes from Robert Putnam (1988), who developed the concept of “two‐level games” in the context of diplomacy and domestic politics.³ In Putnam’s conception, the “Level One” game focuses on traditional diplomatic agreements, while the Level Two game focuses on the formal or informal domestic ratification of such agreements. While Putnam described rich interactions between these two intersecting games, the simplest version of his approach is sequential: international Level One agreements are reached first, externally (or “across the table,”) and are then followed by each side’s domestic, Level Two ratification process, internally (or “behind the table.”) A stripped down version of this analytic structure, of course, involves two agents negotiating on behalf of their separate principals, each of whom must approve the deal struck by the agents (see, e.g., Pratt and Zeckhauser 1985; Lax and Sebenius 1986; and Mnookin, Peppet, and Tulumello 2000).

Employing this concept, but venturing beyond its diplomatic origins, I use the terms Level One and Level Two to refer to the international/external and domestic/internal negotiations, respectively. Figure One below is a standard representation in which side A negotiates with side B “at the table” to reach a Level One agreement that is then subject to Level Two ratification on each “side” of the table.

My special focus here is on the Level Two domestic constituencies or other “internal” factions — whether in the public or private spheres — that can support or block Level One deals. In Figure One, Level Two internal/domestic opponents of the A–B deal are represented by A– and B–, while supporters are indicated by A+ and B+. I am intrigued by how Level One negotiators on one side can help (or hurt) their counterparts on the other side to manage their Level Two challenges and vice versa.

Many kinds of barriers, from psychological to cross‐cultural to structural, can block the path to desirable agreements (see, e.g., Arrow et al. 1995 and Lax and Sebenius 2006). While Level One negotiators may see significant joint gains through agreement, various factions on each side of the table may act to block such deals. Such Level Two factions may represent broader populations but frequently act on behalf of small but influential minority interests.

For example, an American president may reach a provisional, Level One deal with the Speaker of the U.S. House of Representatives over the federal budget or debt ceiling increase. While a majority of constituents on each side of the table (Level Two) may favor the deal, powerful factions on one or both sides may serve as blocking coalitions. Such factions may be potent because of their greater intensity and cohesion or their institutional position. As such, the broader national good may be frustrated.

To take another example from Israeli–Palestinian negotiations cited by Robert Mnookin and Ehud Eiran (2005), the Level Two behind‐the‐table challenges may be even greater than the Level One across‐the‐table ones. Settlers and their political advocates on the Israeli side, as well as militant factions and diaspora Palestinians, may, for separate reasons, effectively prevent leaders from making — or even publicly proposing — generally desirable deals for which the leaders would be unable to gain sufficient public support. In another realm, generally beneficial trade negotiations are notorious for being blocked by the intense opposition of relatively small factions, even when they have been offered special compensation. Such behind‐the‐table blockages are hardly limited to the public and diplomatic arenas: a rogue union faction may derail a worthwhile labor contract, or an entrenched management may effectively block a merger that promises otherwise high benefits to shareholders and other constituencies.

Suppose that the negotiating principals want to strike a Level One agreement but expect to face significant Level Two opposition from domestic constituencies and/or other internal factions. Can such likely opposition — the identities of the skeptical parties along with the substance of their interests, concerns, and perceptions — be mapped in advance for each side? In addition to forging a Level One deal, could the principals on each side also focus on developing Level Two agreements or understandings on specific measures that each side could take or avoid to help the other side(s) with its domestic and other constituency challenges?

For example, Figure Two illustrates one type — there are many variants as we will see later — of Level Two negotiation in which side A negotiates with side B to help B strengthen B’s internal supporters (B+) in B’s internal negotiation to overcome B’s internal opponents (B–). In the case illustrated by Figure Two, A and B could be said to form a tacit coalition to overcome the potential of B– as a blocking coalition in B’s internal negotiation. (Of course, Level Two negotiations could run from B to A as well as in both directions simultaneously. Other intriguing dynamics are also possible: e.g., A + and B + could join forces to help A and B overcome A– and B–.)

Traditionally, negotiators deal with Level Two barriers in two ways. First, Level One deal terms can be crafted to meet the interests or to overcome the objections of enough internal players to permit the creation of a winning coalition (i.e., enough of the right parties to enable a deal to be reached and, ideally, implemented and sustained). For example, trade deal provisions may be designed to compensate “losers” who might otherwise block the broader agreement, or at least the deal may be structured to make one or both negotiators “look good” to the bosses who need to approve the agreement. Second, once a Level One deal is agreed, negotiators can be tasked with “selling” the deal to constituents via a sustained persuasion campaign or by arranging side payments on unrelated issues (which, analytically, can be regarded as linked versions of the first method.)

A closely related approach involves reaching agreement on elements of the negotiation process itself, which can send a valuable signal to Level Two players. For example, a prominent labor negotiator once described to me a simple, if cynical, effort to sway skeptical union constituents. In this case, given economic realities, both union and management negotiators clearly understood the feasible deal terms from the outset, but reaching agreement too quickly would have raised union members’ suspicions that their interests had not been vigorously advocated. So the two negotiators agreed to lock themselves into a room from mid‐afternoon until the wee hours of the morning. Those outside the room heard angry shouts and tables being pounded. Inside, the reality was congenial: with nice meals ordered in, plenty of alcohol, friendly reminiscences, and knowing chuckles as the two sides periodically made loud theatrical sounds to dramatize the “battle” being “fought” – for the benefit of outside constituencies. Finally emerging, haggard, in predawn hours, the two sides’ “hard‐won” agreement had a far greater chance of acceptance among union members because the process mollified their suspicions of a sellout, without altering the terms of the negotiated contract itself.

The higher the stakes, however, the more daunting this Level Two challenge can become. As Putnam observed, “Level One negotiators are often badly misinformed about Level Two politics, particularly on the opposing side.” He cited the conclusions of a series of international bargaining cases analyzed by Glenn Snyder and Paul Diesing: “decision makers … only occasionally attempted such assessments, and when they tried they did pretty miserably …” (Putnam 1988: 452).

In an example of inadvertent negative handling of Level Two issues, consider the Geneva Accord, a prominent, unofficial effort to craft an Israeli–Palestinian peace deal. After an important negotiating session, a key Israeli participant sought to indicate progress to key Israeli constituents. He was quoted to the effect that the “Palestinians had given up the right of return.”⁴ This claim was negatively received by many Palestinians, and Palestinian leaders denied it nearly instantly, damaging prospects for wider support of this important initiative.⁵

More broadly, leaders on each side make statements in Arabic or Hebrew about peace talks or agreements that are intended for “domestic consumption.” Typically, such statements rapidly find their way to the other side, generating suspicion and undermining what may be genuine progress at the table. In the media‐intensive Internet era, efforts to achieve “acoustic separation” by separately conveying contradictory messages to different publics often prove futile.

Despite this discouraging record, savvy negotiators can skillfully address Level Two barriers in creative ways. William Ury (1991: 122) has observed that “your counterpart’s constituents may attack the proposed agreement as unsatisfactory. So think about how your counterpart can present it to them in the most positive light, perhaps even as a victory.” Ury offers the following example from the Cuban missile crisis, which was in addition to President John F. Kennedy’s tacit agreement to remove obsolete U.S. missiles from Turkey and a private ultimatum Soviet Premier Nikita Khrushchev (Allison 2012):

Ury (2007) has counseled side A to consider helping to equip side B to write B’s “acceptance speech” — in a manner that meets A’s interests — directed to B’s constituencies. As a tool to help craft the other side’s acceptance speech, he has suggested making a chart listing several key considerations (that I have elaborated):

• precisely who B’s constituencies are along with their likely interests and perceptions of the negotiation;

• the key themes and framing of the “acceptance speech” that will make it persuasive;

• the most likely criticisms and questions such as “What exactly did you give up and why?” “You never should have made that concession, which gives away our vital interests!” “That makes us look weak and sets a terrible precedent!” and “You should push back hard rather than giving in!”; and

• the best anticipatory and subsequent responses to the most important such criticisms.

If A has probed and understood B’s interests, perspectives, and constituencies in enough depth to help craft a credible acceptance speech for B, the range of feasible deals should be much clearer to A. Obviously, the easier it is for B to imagine “selling” the deal to his or her constituencies, the more likely B is to make a deal with A. This approach should, paraphrasing the words of Italian diplomat Daniel Vare, permit “B to have A’s way” (Ury 1991).

At first glance, Ury’s advice might be understood as simple face saving, framing, and tacit issue linkage, but Level Two actions can go much farther to directly affect the other side’s Level Two process. For example, during the preparations for the 1978 Bonn economic summit, there was significant internal American opposition to oil price decontrol, a policy strongly favored by the U.S.’s key economic partners as part of a package involving German and Japanese stimulus, policies that were themselves opposed by powerful German and Japanese factions. According to the most conventional interpretation, ultimate international agreement on these decontrol and stimulus measures, which were actually implemented in each country, resulted from mutually beneficial tradeoffs in a package deal (Putnam and Bayne 1987; Putnam 1988).

A closer look, however, reveals that actions were taken by each side to help the others with their Level Two domestic challenges. For example, Putnam (1988: 429) reported that, to overcome potent U.S. domestic opposition to oil price decontrol, “American negotiators occasionally invited their foreign counterparts to put more pressure on the Americans [at home] to reduce oil imports.” Ultimately, such interventions aimed at influencing (Level Two) U.S. opponents proved successful.

Similarly, external pressure for economic stimulus in Germany and Japan that was orchestrated by internal advocates and willingly supplied by foreign counterparts overcame opposition and tipped the internal balance. As Putnam (1988: 428–429) described it, “Within Germany, a political process catalyzed by foreign pressures was surreptitiously orchestrated by expansionists inside the Schmidt government. … Publicly, [West German Chancellor] Helmut Schmidt posed as reluctant to the end. Only his closest advisors suspected the truth: that the chancellor ‘let himself be pushed’ into a policy that he privately favored. …” And in Japan, “without the external pressure, it is even more unlikely that the expansionists could have overridden the powerful MOF [Ministry of Finance]. ‘Seventy percent foreign pressure, 30 percent internal politics,’ was the disgruntled judgment of one MOF insider. ‘Fifty‐fifty,’ guessed an official from MITI [Ministry of Trade and Industry]” (Putnam 1988: 429).

These examples illustrate how a Level One negotiator can help with his or her counterpart’s Level Two challenges. As we will see via the extended example in the next section, however, these methods hardly exhaust the repertoire of such devices.

A more elaborate episode involved the delicate diplomatic situation between the U.S. and the then‐Soviet Union over German reunification within NATO after the fall of the Berlin Wall.⁶ Soviet President Mikhail Gorbachev faced powerful internal opposition to his reform policies (perestroika) in general as well as to his increasing willingness to go along with American advocacy of German unification and German membership in NATO. The KGB (Soviet spy agency), the Politburo (the central political body of the Soviet Union), conservative politicians, and many Soviet military officers believed that Gorbachev was conceding far too much to the West. With almost four hundred thousand Soviet troops stationed in East Germany and the rights that the Union of Soviet Socialist Republics (USSR) retained as one of the victorious “Four Powers” at the conclusion of World War Two, the Soviets had potent means to obstruct German reunification within NATO.

Wanting perestroika to succeed and Germany to be reunified within NATO, President George H.W. Bush and Secretary of State James, Baker proved themselves to be extremely skilled Level Two negotiators in at least four ways:

1. They consciously avoided actions that would have caused domestic problems for their reformist Soviet counterparts.

2. They helped the Soviets craft a convincing domestic explanation of the direction that negotiations over Germany were taking.

3. They chose not to escalate in response to inflammatory negotiating statements made by the Soviets for domestic consumption.

4. They directly worked with their Soviet at‐the‐table counterparts to help the Soviet reformers overcome their powerful domestic opponents.

Their actions in this important, even singular, case carry broader implications. First, as the Berlin Wall fell, Bush and Baker realized that the American response could exacerbate already huge domestic problems for Gorbachev and his Foreign Minister Eduard Shevardnadze. Robert Zoellick (2000: 25), counselor to Baker and himself a key American negotiator during the reunification talks, cited the value to the process of “Gorbachev’s [correct] belief that [President] Bush would not exult” or convey any sense of what came to be known as “triumphalism.” Baker (2012) observed that

Baker (1995: 170) reported a later encounter between Bush and Gorbachev at which Bush noted the stinging public criticism in the U.S. that Bush had taken for seeming to lack “the vision thing” in the context of German reunification. Bush stated,

Second, as Robert Zoellick (2000: 19) explained, “We even helped our Soviet counterparts to develop a public explanation of how the outcome took account of Soviet interests and sensitivities.” Baker (1995: 251) elaborated how this was done, describing deliberate Western actions on security, political, and economic issues: “We had already planned to take all these steps individually, but by wrapping them in a package and calling them the ‘nine assurances,’ we greatly enhanced their political effect and assured the Kremlin that it would see their full impact. The package was designed so that … the Soviets would not be handed an abject defeat. Above all, it was an effort on our part to stand in Gorbachev’s shoes and help frame the issue so that he would have a domestic explanation.”

Third, because they had developed an understanding of the other side’s political situation, the American negotiating team was able to show restraint at the bargaining table in the face of apparent provocation and backsliding. As the internal tug‐of‐war between the Soviet reformers Gorbachev and Shevardnadze and their conservative opponents heated up, at‐the‐table progress stalled. For example, the crucial (positive) turning point in the reunification negotiations occurred at a White House meeting during which Gorbachev agreed to respect German sovereignty after reunification and to permit Germany to choose its alliance, knowing that the Germans would choose NATO. Weeks later, however, in Berlin talks, Shevardnadze made a lengthy, confrontational statement in which he harshly repudiated these core concessions. Baker suspected that there had been a reversal in Moscow against the reformers. Choosing to respond firmly, but not to escalate and force the issue, which could have led to a damaging stand‐off, he sent his top staffer, Dennis Ross, director of the State Department’s policy planning staff, to find out what happened. Ross privately confronted his counterpart, Sergei Tarasenko, with whom he had established a close “back channel” relationship. “This is a total reversal,” Ross said. “You guys just screwed us. What the hell is going on?” (Baker 1995: 256).

Ross learned that Shevardnadze had been forced to present a Politburo‐prepared document, which could not be reversed – it was “frozen” – at least until the end of the upcoming Party Congress. It soon became apparent to Baker “that [Shevardnadze] was posturing for the benefit of his military, and that what he was saying really wasn’t what he believed” (Baker 2012). At this point, however, in Baker’s eyes, Shevardnadze was “as beleaguered as I’d ever seen him, … the domestic situation was clearly overwhelming him,” and Shevardnadze “couldn’t predict” whether Gorbachev would be able to maintain his status as party general secretary (Baker 1995: 257).

Fourth, given this perilous situation, Bush and Baker took extraordinary negotiating measures. They worked directly with Shevardnadze to equip him and Gorbachev with ammunition to meet their upcoming Party Congress challengers. Partly for this purpose, Bush and Baker negotiated internal U.S. government agreement on strong, specific measures — arms control and strategic nuclear doctrines — that would increasingly transform NATO into more of a political than a military alliance. As Baker (1995: 257) stated, “I told Shevardnadze that we were proposing the adoption of a declaration at the London NATO Summit that would highlight the alliance’s adaptation to a new, radically different world.” He described (Baker 1995: 258) the unorthodox process and objective of this action:

Not only did Baker lead the negotiations for NATO members to adopt this document in London, he coordinated the process closely with his Soviet counterparts: “To help Shevardnadze, I sent him a draft of the declaration, hoping to put the reformers a step ahead of the reactionaries as the Party Congress heated up” (Baker 1995: 259).

Zoellick (2000: 19, 25) later explained that this process had been “extremely helpful, Shevardnadze went on to say, because it would enable him to pre‐empt opponents like [Soviet military leader] Marshall [Sergey Fyodorovich] Akhromeyev … And that is precisely what he did. We had progressed to the point where the American and Soviet foreign ministers could plan secretly how to use tentative NATO language to persuade the Soviet Union to accept a unified Germany within NATO.”

Confirming the effects of these Level Two actions after the Party Congress, Shevardnadze told Baker, “Without the [London NATO] declaration, it would have been a very difficult thing for us to take our decisions on Germany. … If you compare what we’re saying to you and to [German Chancellor Helmut] Kohl now with our Berlin document [the basis of Shevardnadze’s earlier apparent hardline reversal], it’s like day and night. Really, it’s like heaven and earth” (Baker 1995: 259).

Of course, the American negotiating strategy was not limited to helping Soviet reformers with their behind‐the‐table challenges, although that is the focus on the present article. As Ross has emphasized, a complementary series of American actions was intended to “leave no doubt that it would be futile and counterproductive [for the Soviets] to try to prevent reunification” (Ross 2007: 41). And it would be the height of misinterpretation to imagine that Bush and Baker were motivated by altruism or a primary concern for the other side. Rather, these Level Two actions and understandings were aimed at accomplishing a central goal of American foreign policy at the Level One table. Baker (1995: 259–260) emphasized this during a particularly contentious moment during the negotiations over NATO’s acceptance of its extraordinary declaration:

To help one’s counterparts with their behind‐the‐table challenges requires first and foremost that the negotiator understand the other side and the barriers it faces. As James Baker stressed in his remarks upon receiving Harvard’s 2012 Great Negotiator Award, “If there was a single key to whatever success I’ve enjoyed in business and diplomacy, it has been my ability to crawl into the other guy’s shoes. When you understand your opponent, you have a better chance of reaching a successful conclusion with him or her. That means paying attention to how he or she views issues and appreciating the constraints they face” (Baker 2012). This ability helped Baker not only in the negotiations over German reunification but also in others as well. “This approach,” he said, “helped us build the Gulf War coalition that ejected Saddam Hussein from Kuwait in 1991. Effective U.S. leadership depended on our ability to persuade others to join with us. That required us to appreciate what objectives, arguments, and trade‐offs were important to our would‐be partners” (Baker 2012).

In part, this meant developing direct understanding via high‐level personal diplomacy, backed by the work of expert staff members who in turn drew on regional experts. But carefully cultivating close back channel relationships — such as the one between Dennis Ross and Sergei Tarasenko (Shevardnadze’s chief assistant and confidant) — proved vital. Similar back channel relationships were consciously developed between Americans and Germans: Robert Blackwill at the National Security Council with Horst Teltschik, Chancellor Kohl’s national security advisor, and Robert Zoellick with Frank Elbe, right‐hand man to Hans‐Dietrich Genscher, Germany’s foreign minister (Ross 2007).

The admonition to understand the other side is, of course, standard negotiation advice, but its typical objective is to figure out a creative deal that will meet each side’s interests. The actions of Baker and his team with respect to German reunification, however, illustrate a fuller rationale for developing such an understanding: to help the other side overcome its constituency barriers. Ross put it this way: “I would coordinate with Tarasenko before the meetings to avoid surprises or to find out where there were problems that would have to be managed. … [these] made it possible to understand a Soviet move and how U.S. or German responses might affect the maneuverings in Moscow … it also permitted us to design the words and actions that each of us could use to help the other” (Ross 2007: 44). To successfully craft actions for this purpose, one side cannot limit its knowledge of the other to the interests of at‐the‐table negotiators. Rather, one side must deeply understand the context in which its counterpart is enmeshed: the web of favorable and opposing constituencies as well as their relationships, perceptions, sensitivities, and interests. Recall the value for this purpose of direct, trusting relationships such as that nurtured by Eizenstat with Lambsdorff in negotiations over Holocaust‐era assets and slave labor.

Armed with this understanding, it becomes possible for one side to help the other side with its Level Two challenges. Beyond tailoring the content of the Level One deal for this purpose (e.g., with “compensation provisions”), one side can help the other, and vice versa, via a number of devices: by the form of the negotiating process (to send a useful signal to constituencies); by avoiding (or making) statements that inflame (or mollify) the other side’s internal opponents; by constructive actions at the bargaining table informed by knowledge of the other side’s internal conflicts; by providing content and a useful frame for the other side to craft a domestic explanation or even an internal “acceptance speech”; by coordinating external pressure on the other side’s internal factions; and even by directly negotiating with one’s Level One counterparts to design measures to thwart their Level Two opponents.

This list hardly exhausts the possibilities, but it begins to flesh out the underappreciated dynamic of “Level Two negotiations”: how each side can help the other with its internal conflicts. The full range of such measures as well as a systematic analysis of the conditions under which they can be effective await further research. After all, Level Two challenges can range from simple to complex: from an agent getting an “OK” from his boss, to a union or the U.S. Senate ratifying a deal, to interest groups and nongovernmental organizations tacitly going along with an agreement, and to disorganized and faction‐ridden publics collectively deciding to support or oppose a high‐level deal via Twitter, Facebook, and other social media. Developing effective and legitimate means for dealing with such varied Level Two barriers poses an ongoing challenge to negotiators and researchers alike.