By Ethan Holden
In December 2007, the United Nations High Commissioner for Refugees (UNHCR) convened to discuss the topic of refugee protection and international migration. Three main points were made: firstly, that in the current stage of globalization, migration is inevitable, secondly, that border control or policy alone is ineffective in managing migration, and thirdly, that more comprehensive measures which integrate various aspects of migration are required to provide the safety necessary to those who migrate. Against this backdrop, attempts to define migration are not merely theoretical exercises; they function as legal and policy filters that determine access to rights, including freedom of movement and international protection.
Several definitions of migration have emerged during the post-cold war period; however, many have been controversial as the reasons why individuals and groups migrate can change drastically based several variables such as: circumstance, volition, and resources available with direct implications for how protection is granted or denied.
This piece examines four influential and widely used approaches to defining migration—forced and voluntary migration, proactive and reactive migration, inclusive definitions, and transnationalism—and assesses their respective strengths and limitations. It explores how these frameworks operate in practice, particularly in determining access to rights and international protection, and highlights the risks inherent in rigid or overly expansive categorisations. The analysis argues that migration definitions are only meaningful when situated within their historical and political contexts and contends that transnationalism offers a more flexible and legally responsive framework for understanding contemporary patterns of mobility.
The terms forced and voluntary migration gained popularity during the post-cold war era due to a rapid increase of cases where large populations were involuntarily relocated. Forced migrants were characterised by a lack of choice and associated with forms of movement such as exile, the slave trade, and mass expulsions. Voluntary migration was described by a combination of push and pull factors in which the migrant had an active choice. This understanding closely aligns with the legal framing established in the 1951 Refugee Convention, which defines refugees as individuals compelled to flee due to a well-founded fear of persecution, thereby embedding lack of choice as a central criterion of forced displacement. Subsequent regional instruments, most notably the 1969 OAU Convention and the 1984 Cartagena Declaration, further expanded the refugee definition to include flight caused by generalized violence, foreign aggression, and serious disturbances to public order, reinforcing a broader conception of forced migration. In contrast, voluntary migration has been described as the outcome of a combination of push and pull factors in which migrants retain a degree of agency and decision-making capacity. The forced–voluntary dichotomy has been widely employed in sociological analyses of migration, framing movement as a response to negative pressures encouraging departure from particular locations (push factors) and positive conditions attracting migrants to others.
One perceived advantage of the forced versus voluntary migration distinction is that it creates a clear administrative boundary between categories of migrants, enabling governing bodies and international organisations to prioritise certain groups for protection. In legal terms, this distinction is anchored in the aforementioned refugee definition established by the 1951 Refugee Convention and operationalised by UNHCR, which predates the contemporary conflicts often used to illustrate forced displacement. This framework has been applied in high-conflict contexts, including displacement arising from the Afghan civil war in the early 1990s and the period of Taliban rule until 2001.
While the forced v. voluntary definition seems to work well in specific cases and on short time spans, it is harder to justify it empirically on larger time spans, the decision to migrate in pursuit of improved living conditions is often only one among several possible strategies, such as: continuing education, applying for better paid jobs, or diversifying income. All these options can carry varying degrees of constraints depending on the specific situations and contexts of migrants. Nonetheless migrants that migrate for the above-mentioned reasons are still categorised as voluntary migrants, and cases such as extreme poverty are not seen as affronts to humanity and therefore deserving international protection as it would be if they were categorised as forced migrants.
The cases of Afghanistan and Pakistan are often treated as paradigmatic examples within discussions of forced and voluntary migration, reflecting dominant policy and scholarly perceptions rather than rigid legal categories. Migrants from Afghanistan have frequently been characterised as forced migrants due to prolonged conflict and well-documented human rights abuses since the 1990s; however, longstanding patterns of Afghan migration and mobility predate these conflicts, complicating the notion of a singular or exclusively forced form of migration from the region. By contrast, migration from Pakistan has often been framed as predominantly voluntary, particularly in relation to male labour migration on short-term contracts to the Gulf states. However, Pakistan itself has experienced cases of conflict since 2009 in southwest areas such as Baluchistan, and has also been affected by collateral damage linked to external ‘war on terror’ interventions. Most migrants move out of a certain amount of volition as well as some pressure on them to leave, neither can be entirely forced or voluntary and therefore this dichotomy has failed to identify protection needs in several cases. Therefore, despite the administrative appeal of having definite criteria for granting international protection, reliance on the forced–voluntary dichotomy risks obscuring cases where protection needs are present but not readily recognised, a reality reflected in the presence of substantial numbers of Pakistani refugees globally. These cases illustrate that most migration involves varying degrees of both agency and constraint, rather than fitting neatly into forced or voluntary categories. As a result, while the dichotomy may offer administrative clarity for the allocation of protection, it ultimately fails to capture the multiple and overlapping drivers of migration.
In the attempt to move away from a fixed dichotomy Richmond (1993) elaborated a new definition using terms proactive and reactive. Unlike voluntary or forced, proactive and reactive act as extremes on a spectrum where migrants can either be described as more proactive or more reactive, the boundary between the two categories is more flexible. Richmond complements the proactive/ reactive definition with another axis of structural facilitators and structural constraints. Proactive migrants share many aspects with the definition of a voluntary migrant, such as choice and pursuit of economic gain. However, among the voluntary migrants, spies, defectors and political actors are also considered proactive migrants if they are seeking to maximize a net advantage and have a level of structural facilitators. Reactive migrants, instead, align more closely with the criteria set out in the 1951 Refugee Convention, and have severely restricted amounts of freedom. However, Richmond does not exclude economic collapse and famine as reasons for reactive migration, which makes this criterium more inclusive of individuals deserving international protection, who would be excluded with the voluntary/ forced categorisation.
The separation of migrants and refugees is one of the main points of contestation in dichotomic definitions. International refugee law and many governmental and institutional frameworks require the crossing of an international border in order to recognise them as a refugee. This is not always an adequate description of the diverse routes of migration, especially in cases where international borders become contested. Unlike dichotomies such as forced and voluntary, the proactive and reactive approach to migration does not draw a fixed line between migrants and refugees. A scalable system which allows for different types of migration is ultimately a more effective instrument for dealing with the rights connected with migration.
While not creating a clear boundary between refugees and migrants can be seen as a beneficial attribute to the definition, it also makes it more difficult for governments and international organisations such as the UNHCR to prioritise certain groups of migrants who may need particular attention. While the term reactive migrant does agree with some of the definitions that the UNHCR requires to be a refugee, such as: war, persecution based on race, and natural disasters, it also includes a much broader amount of conditions which may include extreme poverty and climate change which the UNHCR does not consider as reasonable grounds to claim refugee status. Therefore, while a proactive and reactive definition to migration may be more adequate to describe the multifaceted reasons of migration, it is difficult to apply in practice as it does not clearly identify who requires specific protection.
The reactive and proactive definition also heavily relies on structural constraints and limitations to define migrants. This however ignores the migrant’s agency and shapes migration arounds structural constraints or facilitators such as state policy, labour incentives, anti-migration laws, or social sentiment being considered as the main reasons of migrations. Consequentially, all migration is then seen through structural restrictions, even though migration often does not occur if there is no form of agency.
The inclusive definition of migrants moves fully away from dichotomies and puts migrants and refugees under the same holistic umbrella without attempting a distinction between migrants and refugees. This definition states that any individual moving from their original place of habitation regardless of the circumstances that made them migrate is considered a migrant.
The benefits of having a broad and unique definition promotes a widespread change in priorities by acknowledging that every migrant potentially needs additional support, protection, or resources. This allows for a more realistic description of migration as migrants may, at times, need different degrees of additional aid. A single, encompassing definition seeks to reduce political bias by presenting migrants primarily as individuals in need of protection, rather than as mobile populations motivated by self-interest. International organisations such as the United Nations Department of Economic and Social Affairs, as well as news outlets such as the British Broadcasting Corporation, have moved towards the inclusive definition of migrants. A single definition avoids boundaries and therefore the risk of excluding specific migrants from potential aid or support is reduced.
However, there are also negative consequences in using an encompassing definition. The inclusive definition does not acknowledge a variable degree of experience which migrants succumb to such as in Chile in 2018, when the Chilean government used the inclusive language of “safe, orderly and regular migration” which first originated from the IOM to disguise exclusive policy with the aim of restricting migrant mobility. By failing to acknowledge that certain migrants encounter more difficulties than others, inclusive definitions create the risk that certain groups fall victims of governmental policies elaborated for the international audiences with the aim of concealing national emergencies.
These three forms of definition, forced and voluntary, reactive to proactive, and inclusive, show that migration continues to be a complex and everchanging landscape. This is supported by a more recent approach to understanding migration called transnationalism. Transnationalism takes various elements from the three previous definitions. It incorporates aspects of the forced and voluntary dichotomy as it acknowledges that certain migrants require more support depending on the resources that they have at their disposal, or to their membership to certain vulnerable groups. However, it allows the flexibility of the proactive to reactive model as it accepts the salient nature of migration and the idea that a migrant can change definition over time, space, or circumstance. It offers an agency approach rather than relying purely on structure. Finally, it provides the opportunity for a more inclusive approach to viewing migration terminology, however, avoids the risk of being used as a disguise for exclusive policy because it acknowledges that certain migrants require specific care. From a refugee law perspective, this approach better reflects the reality of mixed migration flows increasingly encountered by protection regimes.
While the concept of mixed migration has become increasingly prominent in policy and scholarly debates, there is limited indication that it will lead to a fundamental shift in UNHCR’s refugee definition, given the legal and political constraints inherent in expanding protection regimes. Although refugees and other migrants frequently intermix in terms of routes, circumstances, and exposure to risk, any substantive redefinition would require broad international consensus among states, making such change unlikely in practice.
Regardless, there can be no single, universally adequate definition of migration. Dichotomies such as forced and voluntary are common in policymaking as they provide clear requirements to conditions, however they are often too general and clear-cut, risking the marginalisation of specific groups which require aid. Approaches such as proactive and reactive allow for a more nuanced understanding of migrations, however, because of their stress on structural limitations and their avoidance of clear-cut distinctions, they are of little benefit in the applied domains. Inclusive definitions aim to change the perceptions towards migrants and try to offer a fair potential of protection and acknowledgement across the phenomena of migration seen holistically, however we have seen that it is also easy to use inclusive definitions for excluding and marginalise some specific groups that need protection. Transnationalism takes the strengths of all three of the above-mentioned approaches and provides a potential solution to the definition of migration as it integrates various aspects of the three previous definitions and therefore effectively addresses grey areas by providing an encompassing view on migration where migrants’ needs and resources change over time and location. Although more demanding, to date transnationalism seems to provide governments and organisations such as the UNHCR, with the conceptual instruments to create tailored policies which avoid the marginalisation of migrants, or their exploitation due to a lack of protection, and are therefore more respectful of international human rights.
References:
Richmond, A.H. (1993) ‘Reactive migration: sociological perspectives on refugee movements’, Journal of Refugee Studies, 6(1), pp. 7–24.
Ethan Holden is a legal researcher and lecturer working at the intersection of international development law, human rights, and comparative family law, with a strong focus on the regulation of artificial intelligence. He holds an LLM in International Development Law and Human Rights from the University of Warwick, where his dissertation examined AI regulation from a legal and policy perspective, and a First-Class BA in International Relations and Applied Languages from Oxford Brookes University. His research interests include Islamic legal traditions and child protection, particularly the interaction between non-Western family-law institutions—such as kafala—and European legal and policy frameworks, as well as AI governance in contexts involving migration, access to justice, and rights protection. He has experience working with civil society organisations, legal clinics, and public-sector institutions, in the United Kingdom and in Pakistan, and currently combines legal research with university-level teaching.
The views expressed in this article belong to the author/s and do not necessarily reflect those of the Refugee Law Initiative. We welcome comments and contributions to this blog – please comment below and see here for contribution guidelines.