AMBIGUITY IN TALIBAN-CONTROLLED AFGHANISTAN

In 2023, for the first time since the Taliban takeover, we traveled back to Afghanistan to obtain official permission to conduct new research. For ethical and safety reasons, we had restricted our pool of interviewees to Taliban elites and bureaucrats and determined that asking the authorities for approval and access was the best way to protect them from potential reprisals. We assessed the appropriateness, necessity, and feasibility of conducting this research (Ibrahimi and Malejacq Reference Ibrahimi and Malejacq2024); designed safety measures; and applied for visas.Footnote ³ After exhausting formal channels, we turned to personal networks and obtained visas through one of the few diplomatic missions controlled by the Taliban. Our interlocutors confirmed that obtaining a research permit was mandatory and would allow us to pass through checkpoints unencumbered; however, the permit could be acquired only in person at the Ministry of Foreign Affairs (MFA) in Kabul. And so we headed for the Afghan capital.

Upon arrival at the MFA, we were informed of a different procedure that required first obtaining permission from the Afghanistan Science Academy. At the Academy, a Taliban official rebuked us as spies and lectured us about Islam and the achievements of the new regime, but ultimately appeared to approve our request. The next morning, we were summoned to the MFA only to be bounced from one waiting room to the next, criticized for procedural errors, and eventually told that a separate committee would make the decision. The following day, we were notified that our request had been rejected and we were “invited” to leave the country immediately. At the airport, we received another message from our MFA contact stating that a permit could be issued rapidly if we provided more details on our research. Worried about staying without authorization, we left the country anyway. We submitted the required documents later that evening, but our request was denied again. By that point, we could not determine which authorities or rules were relevant to our request and whether any decision had been made.

Arguably, what we identified as ambiguity could have been amplified by our positionalities as foreigners and our relative inability to identify existing patterns in such a timeframe. We are keenly aware of the value of immersive ethnographic research and the cultivation of long-term, personal relationships, which we have adopted and reflected on in our respective work (Koehler, Bhatia, and Rasool Mosakhel Reference Koehler, Bhatia and Mosakhel2022; Malejacq Reference Malejacq2019; Weigand Reference Weigand2022). Yet, ambiguity was already a feature of governance in Afghanistan before the change in regime, characterizing both the international engagement (Kühn Reference Kühn and Lemay-Hébert2019) and the insurgency (Tapscott and Urwin Reference Tapscott and Urwin2024). For example, the concept of amr-e-sani, which loosely translates as “until further notice,” is mentioned frequently by Afghan bureaucrats to refer to matters in which the outcome is uncertain and for which they have discretionary authority. “Sometimes,” explained a former bureaucrat, “it is also used to obscure a lack of clear vision within certain branches of the public administration.” Ambiguity was even prominent in Taliban policy making when they first ruled the country (1996–2001). For instance, many articles of their constitution “left certain offices or articles vague and subject to interpretation by the judiciary and leadership” (van Linschoten and Kuehn Reference van Linschoten and Kuehn2018, 210).

The fact that ambiguity has been a long-lasting feature of Afghan governance further suggests that Taliban policy making—or lack thereof—cannot be understood by capacity deficit and institutional immaturity alone. Moreover, our initial insights were validated by the conversations and interviews we held with interlocutors who provided congruent interpretations of our ordeal or had experienced similar ones. These exchanges confirmed that the ambiguity we encountered was not merely a function of ignorance, confusion, or bureaucratic incompetence. The next section unpacks the different mechanisms through which ambiguity is produced and experienced, and alternative explanations are revisited in the conclusion.

A TYPOLOGY OF AMBIGUITY POLITICS

The research-permit episode alone does not allow us to draw conclusions about the Afghan state apparatus as a whole, but it served as a prompt for exploratory research (Stebbins Reference Stebbins2001). Building inductively on our experience and expertise and deductively on existing literature, our typology is a theory-building endeavor aimed at guiding analysis beyond the Taliban case (George and Bennett Reference George and Bennett2005, 239–44). We notably drew on diverse bodies of scholarship that demonstrate the pervasiveness of ambiguity politics in various contexts (Natter, Norman, and Stel Reference Natter, Norman and Stel2023), not only as a feature of fragile institutions but routinely present in all types of settings—from Hafez al-Assad’s Syrian regime (Wedeen Reference Wedeen1999) to the US Congress (Page Reference Page1976) to the International Monetary Fund (Best Reference Best2012).

Ambiguity is understood commonly as the quality of an idea or a situation that has multiple meanings or potential interpretations, leaving “the person or situation that is the object of [it] in a state of cognitive impasse” (Mumford and Carlucci Reference Mumford and Carlucci2023, 198). March (Reference March1979, 17) influentially argued that, within organizations, “[t]he problems of ambiguity are partly problems of disagreement about goals among individuals, but they are more conspicuously problems of the relevance, priority, clarity, coherence, and stability of goals in both individual and organizational choice.” We extend March’s (Reference March1979) classic understanding and define ambiguity politics as a mode of governance in which institutions—formal and informal policies, rules, norms, hierarchies, and authorities—are unspecified, contested, or misunderstood, thereby creating space for multiple interpretations and discretionary implementation of the “rules of the game.”

Ambiguity is adjacent to but distinct from at least four related concepts, as follows:

• • Vagueness concerns underspecified or noncommittal discourse, policies, and decisions whose content is unclear (Praprotnik and Ennser-Jedenastik Reference Praprotnik and Ennser-Jedenastik2024), at times fostering multiple interpretations and therefore ambiguity.

• • Uncertainty refers to situations in which the rules are clear but their outcomes are still unknown (Matejova and Shesterinina Reference Matejova, Shesterinina, Matejova and Shesterinina2024).

• • Ambivalence pertains to “the simultaneous existence of conflicting feelings, wishes, and thoughts” (Bauer Reference Bauer2021, 18).

• • Unpredictability is the experiential consequence of ambiguity, whereby actors cannot anticipate which interpretation, rule, or authority will prevail (Stel Reference Stel2020, 8; Tapscott Reference Tapscott2021, 10).

Ambiguity provides greater analytical leverage because it directs attention to the mechanisms that produce and sustain multiple interpretations of the rules of the game, at both individual and aggregate levels, ultimately generating experiences of unpredictability.

Our typology maps these different and often coexisting mechanisms and the forms of ambiguity politics that they produce. The quadrants in table 1 represent ideal types (Weber Reference Weber, Shils and Finch1949) determined by (1) the level at which ambiguity is generated (elite/administrative); and (2) the intentionality through which it is produced (strategic/accidental). Our typology functions as a heuristic device for the analysis of imperfect cases of governance practices that may combine several of these forms of ambiguity rather than classifying cases into mutually exclusive types. By isolating the underlying mechanisms of ambiguation, the typology moves beyond surface descriptions to offer a theoretical framework that explains the pervasiveness of ambiguity politics across institutional settings. The four ideal types are described in detail in the following discussion.

Table 1

A Typology of Ambiguity Politics

Table 1 Long description

The table is structured with two main columns under the heading Intentionality: Strategic and Accidental. These are intersected by two rows under the heading Level of Governance: Elite and Administrative.

* Top-Left Quadrant (Elite and Strategic): ACCOMMODATING. Elites deliberately create or maintain ambiguity to negotiate power sharing and or avoid commitments.

* Top-Right Quadrant (Elite and Accidental): CONTRADICTORY. Contradicting decisions and or a lack of coordination lead to ambiguous or contradictory institutions from elite sources.

* Bottom-Left Quadrant (Administrative and Strategic): DISCRETIONARY. Bureaucrats strategically obfuscate to increase personal power or discretion.

* Bottom-Right Quadrant (Administrative and Accidental): UNANTICIPATED. Bureaucrats face complex or unanticipated situations without clear guidance, creating ad hoc ambiguous practices.

At the elite level, institutional development and authoritative decision making can foster dissent and fragmentation. Elites intentionally may create or sustain ambiguity to facilitate power sharing and avoid politically costly commitments. We phrase this as accommodating ambiguity. Tapscott (Reference Tapscott2021, 5), for instance, showed how some states “cultivate an institutional environment structured to accommodate numerous authorities with overlapping and contestable jurisdictional remits.” In Afghanistan, this form of ambiguity has allowed the Taliban to maintain cohesion amid competing poles of power, both during and after war (Malejacq and Terpstra Reference Malejacq and Terpstra2023). The lack of clear policy vis-à-vis foreign visitors—which we experienced firsthand—could be interpreted in this way. Whereas accommodating ambiguity is particularly consequential in newly established regimes, wherein power is still being consolidated and unity is paramount (Lyons Reference Lyons2016), it remains present in most institutions and organizations—particularly when decisions concern politically sensitive issues. Ethiopian Emperor Haile Selassie, for instance, was notorious for using “unclear and ambiguous” words, “especially when he did not want to take a definite stand on a matter that required his opinion” (Kapuściński 1983/Reference Kapuściński2006, 8).

Alternatively, ambiguity can emerge “by accident” when fragmented authority or weak coordination produces inconsistent or competing elite signals. We call this contradictory ambiguity. For instance, Nassar and Stel (Reference Nassar and Stel2019) documented how elite-level incoherence in Lebanon leads to contradictory policies. In Afghanistan, the Taliban have been molding an authoritarian movement governed through informal institutions and an extreme ideology into an existing, seemingly Weberian state apparatus, thereby creating a dissonance of rules that—combined with intra-elite competition—fostered a high degree of ambiguity. Tribal elders and religious leaders who were asked about their experience with state institutions referred to the government as “disjointed.” As one scholar noted, “the government in Kabul is so powerless they can’t buy tissues without [Taliban Emir Haibatullah]’s approval.” Although the opacity of the decision-making process prevents drawing clear inferences, our experience with Afghan institutions also could be understood as a case of contradictory ambiguity, whereby Taliban leaders display different views on the presence of foreign researchers. Similar tensions notably exist regarding the Taliban’s international engagement with Western countries (Smith and Simon Reference Smith and Simon2022).

At the administrative level, ambiguity may be used strategically to maximize discretionary power. Lipsky (1980/Reference Lipsky2010, 149), for instance, showed how “street-level bureaucrats sometimes cope with their jobs by privately modifying the scope of their authority.” We call this discretionary ambiguity. This often results in either the development of clientelist networks—when certain individuals are provided with services and others are not—or Kafkaesque logics of unending bureaucratic postponement (Sutton and Vigneswaran Reference Sutton and Vigneswaran2011). Our experience with the Afghan MFA is an example of this form of ambiguity, which also may be caused by elites purposely leaving space for interpretation. This is notably the case in Afghanistan, where the Taliban Emir’s rule by decree has provided space for piecemeal policy making and a diversity of practices at the local level (Jackson Reference Jackson2021; Jackson and Amiri Reference Jackson and Amiri2019). The French colonial order was maintained similarly through broad presidential decrees and the arbitrary power of colonial administrators (Badie 2014/Reference Badie2019, 105), and the German asylum regime still rests on an “organised ambiguity towards the acceptance of refugees…deliberately holding asylum applicants in limbo” (Thränhardt Reference Thränhardt, Finotelli and Ponzo2023, 277).

Conversely, we consider unanticipated ambiguity to be another administrative response to a lack of clear guidance from elites. Such ambiguity is not created intentionally by bureaucrats but often is sustained by their lack of action. In these situations, they simply avoid making decisions because they do not know which rules apply and they want to avoid responsibility for ruling on ambiguous matters. “When there is a problem, [the Taliban] postpone, as if the problem will resolve itself,” a former Special Representative to Afghanistan told us. In this context, “everyone is afraid of taking an initiative” confessed an Afghan interlocutor who was reflecting on his experience with Taliban bureaucracy. Mirroring our experience, Sneha et al. (Reference Sneha, Sinha, Varghese, Durani and Patel2021, 55) found that indecision in Indian administration is driven by “a form of rational self-preservation exercised by bureaucrats from the various legal and extra-legal risks to their person, careers, and reputation.” Such ambiguity may be particularly useful for bureaucrats in authoritarian and newly established regimes, not least to avoid blame for making contentious decisions (El-Shaarawi Reference El-Shaarawi2015). As a scholar with extensive experience in post-2021 Afghanistan explained: “What produces ambiguity is more often a reluctance to take sides in unresolved intra-elite debates and a fear of being held personally responsible for politically sensitive decisions.” However, this logic should continue to operate in all types of regimes, institutions, and organizations. Similarly, ambiguity may create unpredictability (Tapscott Reference Tapscott2021, 5) but it also opens up space for agency, which enables ordinary people to negotiate in their daily interactions with bureaucrats.

Ultimately, our typology departs from existing scholarship in two important ways. First, we conceive of ambiguity as both intentional and unintentional rather than only strategic.Footnote ⁴ Ambiguity can be engineered to manage power and risk, but it also frequently occurs “by accident” due to fragmented authority and coordination failures. Second, by distinguishing the different levels at which ambiguation takes place, the typology enables us to unpack empirically interwoven phenomena and to analyze interlinkages. For example, strategic ambiguity at the elite level—such as intentionally vague decrees and policy deferrals—can create space for discretionary ambiguity among bureaucrats. This was notably the case in pre-genocide Rwanda, where the gap that existed between the national and local levels fostered forms of “local engagement with authority [that] was largely experienced as an ambiguous, yet functional, way to manage one’s existence” (Desrosiers Reference Desrosiers2020, 558). Conversely, accumulated ambiguity at the administrative level can generate pressure for elites to clarify or reassert authority, particularly as regimes consolidate. For example, the Taliban Emir’s decision to keep girls’ secondary schools closed—on the same day that the Ministry of Education had planned several school openings (Jackson and Weigand Reference Jackson and Weigand2023)—could be interpreted as an attempt to reaffirm his authority by ending ambiguity on a highly contentious issue.

A WAY FORWARD

Making sense of confusing or contradictory rules and procedures—the reasons for which are unclear—is extremely challenging. Methodologically, this study shows that “reflecting on experiences of getting access (or not) can tell us something important about the institutions we aimed to study” (Kalir, Achermann, and Rosset Reference Kalir, Achermann and Rosset2019, 6), thereby underlining the value of inductive fieldwork and the analysis of meta-data in producing theoretical and generalizable insights (Fujii Reference Fujii2010). Drawing from our encounters with the Afghan state, we argue that corroborating experiences and collecting additional data can help to uncover the reasons for ambiguity and some of the underlying logics of the institutions in question.

We suggest that researchers rely on a set of guiding questions to collect further data. Is the experience of ambiguity unique or widely shared? Is it reflected at the elite level, bureaucratic level, or both? Do the patterns persist over time or do they change with shifts in personnel and leadership? Who benefits from ambiguity and who bears its costs? Additional data may not allow researchers to make causal claims, but it may help to eliminate alternative explanations. In our case, further research revealed that our experience was not unique, that ambiguity was present at both the bureaucratic and elite levels, and that it has persisted over time (within the Taliban and Afghan politics more broadly). This evidence confirms that our experience cannot be explained simply by ignorance or incompetence alone.

There may be numerous explanations for the pervasiveness of ambiguity politics. For instance, the Taliban may simply be mimicking US-led efforts to make the state more legible without much success—in part but not strictly due to fragmentation and intra-elite competition (i.e., contradictory ambiguity). On the contrary, it might be engaged in a reflexive backlash against the legibility-building efforts of their former adversary (i.e., accommodating ambiguity). This is a possible yet unlikely explanation because many of the bureaucrats we encountered had worked under the former Republic. Perhaps the bureaucrats we encountered simply did not have the proper training to assist us (i.e., unanticipated ambiguity). This also is unlikely because our requests involved no specialized skills or any high level of organization. As a scholar who conducted research in Taliban-controlled Afghanistan explained, “[these] bureaucrats generally understand their responsibilities well enough.” In any case, these practices align with our framework as forms of ambiguity.

Our central argument is that ambiguity politics is not inevitably a temporary stage toward greater order and predictability (Tapscott and Urwin Reference Tapscott and Urwin2024) but rather a persistent mode of governance across various settings. Our typology functions as a heuristic device for developing a more flexible and fine-grained understanding of the different forms of ambiguity politics observed across these empirical contexts. For instance, we might expect that with fewer political stakeholders to accommodate and a greater willingness to resort to repression, ambiguity at the elite level is less necessary in authoritarian states than it otherwise would be in more pluralistic and less-coercive environments. Conversely, bureaucratic ambiguity may be more useful in authoritarian contexts, particularly wherein decision making is highly centralized and independent initiative is sanctioned or discouraged. Further research should investigate these dynamics relative to different types of actors and regime types, notably through the conduct of in-depth empirical case studies.

This analysis suggests other avenues for research. For example, if ambiguity is so advantageous, why do political actors frequently clarify positions? A likely answer is that competing pressures also exist—clarity and certainty also confer benefits. Some decisions cannot be delayed indefinitely—for example, the Taliban postponed announcing the decision on reopening girls’ schools as long as they possibly could but eventually had to take a stand when the school year began. Conversely, there are other potentially harmful decisions that the Taliban could have postponed but instead took decisive actions on, such as the ban on opium undertaken in 2022 (Sabawoon and Bjelica Reference Sabawoon and Bjelica2024). In this case, why did the Taliban not maintain ambiguity? Future inquiry should explore when and why political actors do and do not use or maintain ambiguity.

Another question future research must address is whether the advantages provided by ambiguity are time limited. We have noted that ambiguity, although prevalent in clientelist political systems, is not necessarily a cause nor a consequence of weak institutionalization. However, as March and Olsen (Reference March, Olsen, March and Olsen1976, 12) pointed out, it is conspicuous when organizations “are young or their environments are changing.” Ambiguity appears particularly useful in managing competing authorities and delaying contentious decisions in periods of political transition—including periods of democratization (Rothman Reference Rothman1978). Yet, ambiguity also is pervasive in well-established organizations. Future research should determine whether certain forms of ambiguity might be more prevalent, routinized—organized even—at different stages of institutionalization or whether some advantages might fade as regimes become more entrenched.

We recognize the inherent challenges in researching ambiguity politics, not least because the opacity of decision making makes intent almost impossible to ascertain—particularly in authoritarian and secretive regimes (Tapscott Reference Tapscott2021, 9)—and competing explanations difficult to rule out. As we experienced firsthand, research in this domain may be baffling, costly, and frustrating, which partly explains why the concept of ambiguity—although “implicit in much social theory”—remains greatly underemphasized, undertheorized, and under-researched (Best Reference Best2012, 86). However, studying ambiguity is possible, worthwhile, and even necessary to fully understand the political behavior of bureaucratic institutions and the individuals who comprise them. We intend our study to be a starting point.