EU Aid Policy in the Middle East and North Africa: Politicization and its Limits

Some elements of EU assistance to the countries of MENA have become more politicized in recent years. This is largely because sharper differences of opinion have emerged over EU funds being used to limit migration and for security aims. Contestation has intensi ﬁ ed horizontally between different EU institutions and European governments. However, this politicization varies across different policy areas and different parts of the MENA region, and in some instances remains relatively muted. After outlining the general increase in salience and wider institutional engagement; we then contrast this with the more measured increase in the third dimension of politicization, namely polarization over the EU ’ s MENA aid policies. We offer three examples of the way that internal policy-making dynamics have softened polarization and discuss implications for the broader analysis of EU foreign and security policy. We conclude that different kinds of dynamics both drive and limit politicization within the EU ’ s external policies.


Introduction
In recent years the EU has used its external funding in the Middle East and North Africa (MENA) region in increasingly instrumental ways to further a range of foreign and security policy aims. The EU has increased aid in particular for initiatives related to counter-terrorism and migration control; it has equally sought to deploy aid as leverage to respond to thorny strategic challenges within the region. This instrumental use of development funds has brought with it a degree of politicization. In examining EU aid policy in the MENA region, we analyse the three dimensions of politicization outlined in this special issue's introduction: salience; polarization; and an expansion of actors (Hackenesch et al., 2021). Our analysis is based on secondary literature and interviews conducted with key EU officials from different EU bodies between September 2018 and June 2020.
Our empirical results indicate that the salience of EU funding in the MENA region has increased markedly and a wider range of institutional actors has come to engage proactively in these aid decisions. On polarization we find more mixed evidence. Sharper differences have certainly opened up between different institutional actors in the EU on the use of external funds in the MENA region. Yet, a number of factors have also kept the overall level of polarization within relatively manageable parameters.
We detail the different inter-policy balances, institutional processes and geopolitical factors that have acted to dilute polarization. We find that opinions have become more polarized around some elements of the EU's MENA aid policy but also that this is partially off-set by general agreement between different actors on the instrumental use [Correction added on 29 March 2022, after first online publication: The copyright line was changed.] of European aid in this region. This finding adds nuance to other articles in this special issue that also examine security-related aid (Kiratli, 2021;Lauwers et al., 2021). The main argument of the paper is that politicization has become a second-order influence, rather than a definitive block on or determinant of EU aid policy in the MENA region. We stress that the three dimensions of politicization have evolved in different ways.
In this article, we first outline the general increase in salience and wider institutional engagement; we then contrast this with the more measured increase in the third dimension of politicization, namely polarization over the EU's MENA aid policies. We offer examples of managed polarization and discuss the implications of these dynamics for the broader analysis of EU foreign and security policy.

I. Analytical Framework
We adopt the analytical framework outlined in this special issues' introduction (Hackenesch et al., 2021) and apply it to the specific case of the EU's aid policies in the MENA region. This framework follows the broadly shared understanding that politicization involves different combinations of three elements: increased issue salience; a polarization of opinions; and the expansion of actors and audiences involved in EU issues (De Wilde, 2011;De Wilde et al., 2016). We assess the increase in these three dimensions of politicization triggered by a more instrumental use of EU external funds.
Of the three dimensions we home in most tightly on polarization. We understand polarization as the existence of holding extreme positions and political conflict over an issue (De Wilde et al., 2016;Hackenesch et al., 2021). We explore how far the greater strategic instrumentalization of aid has driven a divergence of opinions between different EU institutions and member states. We look here for diffusion from the security to the aid sectorreflecting the key dynamic of horizontal politicization that the special issue's introduction proposes. Polarization is so key because if internal divergences become too pronounced decision processes can fail and the flow of EU aid dry up. Given this, we are concerned to assess crucial signs that strategic considerations have in select ways worked to soften polarization and open new directions in EU external funding. We conceive polarization over aid issues in the MENA as a broad, second-order influence rather than an intense and focused impediment to policy agreements.
To shed light on this dynamic, we draw from analytical frameworks that have sought to show how politicization can be blunted as it spills-over across multiple policy areas. Analysts have offered useful pointers to the kind of dynamics that can contain the extent of politicization. In such cases a particular issue is nested within broader debates with multiple cross cutting concerns at stake, rather than there being one hard, all-defining set of internal divisions (Antentas, 2017, p. 470). Disagreement can be dampened where there is 'poly-cleavage' in the sense of various different issues becoming salient at the same time with different divides in each respective issue area (Zeitlin et al., 2019, p. 966).
We take on board these suggestions and examine their relevance to the relationship between the different dimensions of EU policies in the MENA regionand in particular the contrasting aims to which external funding is supposed to contribute. Our approach builds on observations that politicization is likely to vary considerably across policy domains and institutional arenas (Zürn et al., 2012;Kriesi, 2016;Costa, 2018). To the extent that work on politicization has an external focus, it has tended to look at how security issues are becoming more high profile and controversial in a very generic sense (Schmidt, 2019), but less at how this trend manifests itself in disagreements over how policy tools are deployed on the ground. In order to address the latter issue, we use interview material and detailed process tracing related to specific, second-order aid-related policy debates (George and Bennett, 2005;Collier, 2011;Bennett and Checkel, 2015).
We are also concerned to understand how the politicization of EU aid policy resonates with broader and longstanding theoretical debates about EU foreign and security policy. Academic debates on EU security policy have long centred on internal differences of opinion and the effects of member-state primacy. Yet, recent academic debates on EU security policies have pointed to the emergence of more common strategic identities that serve to temper differences between member states and different EU institutions. Analysts commonly argue that EU security policies are not just the sum of inter-governmental bargaining but increasingly reflect a common set of identities that mitigate divergences (Sperling and Webber, 2019). Some insist that ontological insecurity has most recently played a particularly notable role in offsetting fissiparous politicization (Hegemann and Schneckener, 2019;Lucarelli, 2019). Moreover, these debates have had a particular flavour in the southern Mediterranean: for years, the academic literature has posited EU-Mediterranean relations as an almost emblematic case of a distinctive and common European approach to external relations (Youngs, 2016).
In terms more specifically related to this volume, these trends of collective securitization have been expressed in a so-called security-development nexus. During the 2010s the EU developed a policy framework stressing the way that security and development policies ostensibly work together (Youngs, 2008;Keukeleire and Raube, 2013). Similar dynamics applied to the Union's longstanding promise to address migration challenges through development-led approaches (Lavenex and Kunz, 2008). A crucial question is whether the more pressing strategic challenges that have arisen in recent years have given extra weight to these 'nexuses' or rather placed them under greater strain in a way that deepens politicization. We show how in practice there is evidence of both these dynamics centrifugal and centripetal. In short, we find these broader foreign-policy frameworks relevant in accounting for both polarization and its limits.

II. Politicized Strategic Aid
In the last decade, the EU's overarching strategic priorities in the MENA region have evolved and this has in turn produced important changes to the Union's aid funding across the region. For a brief window after 2011, the EU sought to support those pushing for democratic change in Arab states. As authoritarian regimes pushed back, as conflict spiralled in Syria and then Libya, as these conflicts produced the post-2015 surge in migrant and refugee flows, and as radical jihadist groups gained ground and were linked to attacks inside Europe so the EU adjusted its strategic policies in an attempt to deal with these challenges (Soler i Lecha and Tocci, 2016;Mogherhini, 2019). A core part of this adjustment was the commitment to use external funding in a more strategic or instrumental manner.
The changing context prompted the Commission and member states to increase aid flows to the region. These aid policies in the MENA region have assumed particularly strong strategic salience. The top five recipients of EU aid in 2018 were, in order: Turkey, Iraq, Afghanistan, the Occupied Palestinian Territories and Morocco. Ethiopia and Somalia (now joined with several EU initiatives in North Africa) as well as Syria were also in the top ten (Concord, 2019). The MENA took 26 per cent of Commission aid in 2018, up from 19 per cent in 2013. These high volumes of aid make the wider MENA region an important case study for the politicization of EU development policies. The combination of increased aid flows and high-priority security imperatives has injected this area of EU external action with a high salience.
The most striking trend is how salient these increased aid flows have become to beefing up Arab states' border controls and counter-terrorist capacities (Raineri and Strazzari, 2019). In one of its most high-profile external actions in recent years, the EU launched an Emergency Trust Fund for Africa with an express remit to stem migration from Africa and the Middle East. The EU pressed third countries to sign 'migration compacts' under which the EU offered increased aid in return for recipient countries' cooperation on controlling migration and counter-terrorism. In 2017, European leaders agreed a new 200-million-euro initiative to train and equip the Libyan coastguard to intercept and turn back migrants embarked for Europe. Spain pushed for and won 140 million euros more EU aid for Morocco for migration controls in 2018. Many of the external funding initiatives focused on migration controls have increasingly also incorporated counter-terrorism programmes as the line between migration control and counter-terrorism has blurred (Tsourapas, 2019).
For years under the Euro-Mediterranean Partnership and then Union for the Mediterranean funding did not exhibit a high level of salience. While the EU funded projects related to migration and some security-sector training in the MENA states, debates over the distribution of EU aid funds did not elicit high level strategic engagement or priority. As the strategic salience of these EU external funds increased during the 2010s, so more politicized dynamics emerged. The foremost experts in migration highlight how far this issue has now become a mainstream pillar and priority of external relations debates, well beyond episodic crisis responses (Geddes, 2018).
As this issue salience has risen, so more EU institutional actors have engaged in decision-making over EU aid policy in the MENA region and have done so with far stronger commitment than was previously the case. Moves towards the more strategically instrumental use of aid resources have given rise to a more political and less technocratic series of debates within and across the EU institutions, including the European External Action Service and different parts of the Commission like DG DevCo and DG NEAR the directorates that oversee general development assistance and Neighbourhood-South funding instruments, respectively.
The range of institutional engagement has widened especially in regard to the EU's policy towards Turkey. Until a few years ago, there was tacit agreement on the pre-accession assistance to Turkey and the funds flowed without much debate among EU institutionseven if Greece and Cyprus were sometimes frustrated and sought to use such funding for strategic bargaining. Politicization of these funds has intensified horizontally between different EU institutions and the European Parliament has engaged much more systematically in this debate. In response to Turkey's authoritarian drift, the Parliament has stepped up its engagement in an effort to press the Commission to take into account the latest developments in Turkey when deciding on funding (European Parliament, 2017). The Parliament became increasingly engaged through successive resolutions on accession. This has forced various parts of the Commission and EEAS, as well as member states, to engage with this issue, as the question of whether funding should continue to Turkey rooted itself as a high priority and recurring issue on the EU agenda.
The third element of politicizationpolarizationhas also become apparent, as differences have emerged between these different institutional actors over external funding decisions in the MENA region. As funds were re-directed in more instrumental directions from around 2016, so policy-makers working on development, civil society and good governance issues were uneasy over funding going to security and migration-management programmes. These tensions were especially marked in internal debates about the new Trust Fund. In these negotiations, policy-makers in DG DevCo and DG NEAR baulked at aid funds being used for migration and counter-terrorism initiatives. They have also been especially concerned at development support being made conditional on third countries' acquiescence to so-called readmission agreements (Interview #3).
The most significant axis of polarization has been between those officials mainly concerned with development programming, on the one hand, and those policy-makers with a wider foreign and security policy mandate, on the other hand. These thematic cleavages are pronounced at the EU level, but also exist within member states, and indeed cut across national divides. This kind of internal bureaucratic politics is hardly new within the EU (Hartlapp et al., 2013). The security-development nexus encountered internal obstacles from its conception (Furness and Gänzle, 2017). Yet recent developments have given such internal bureaucratic politics a higher-level strategic tone. Divergences exist on this axis despite the EU's standard line that support for development, security, migration and governance aims all tie together in a mutually reinforcing manner.
DevCo officials have been uneasy at development finance being presented as an easy way to 'stop migration' (Roman, 2018, p. 6). A certain cultural difference exists between development policy officials concerned with the long-term economic and social roots of stability and diplomats in the European External Action Service oriented towards immediate security imperatives; DG Home is even more tightly focused on short-term reductions in migration (Interview #2). In sum, there has been a clash between what might be referred to as the 'security community' and 'development community'. In institutional terms this has manifest itself in DG DevCo and DG NEAR in particular holding different outlooks from EEAS diplomats, and all of these traditional external-policy players expressing concern that DG Home has gained too much influence over aid programming because of its lead role in migration-and border-related projects in the MENA and Sahel regions.

III. Managing Polarization
While this all represents a significant politicization, some internal policy dynamics have helped mitigate the harshness of internal discrepanciesand helped the EU as a whole reach positions broadly agreeable to the array of different institutional actors. This dimension of the three dimensions of politicization, namely polarization, has shown the most mixed directions of change. While internal disagreements have grown, on some crucial dimensions of policy they have been kept within manageable proportions. Polarization has increased, but not dramatically or to an extent that has prevented broadly common EU policies being decided and implemented. Three examples can be offered of dynamics through which the EU institutions have sought to manage politicization.

The Trust Fund and Rise of Multi-Sector Aid Packages
First, a large part of the recent increases in EU aid have been allocated through multi-sector packages that include various combinations of security, migration, development and governance initiatives. These inter-issue balances have helped underpin a degree of collective securitization. A significant part of the overall increases in funding have been for general development purposes, on the ground that better economic development should stem migration flows and radicalization (Lauwers et al., 2021). This balance has helped contain internal differences between different parts of the EU machinery and between member statesfrom security-focused France at one end of the spectrum through to development-oriented Sweden at the other end.
In the most significant concrete example of this, the Emergency Trust Fund for Africa, was designed in such a way as to attract support from both the development and security communities, with a wide mandate that included both sets of goals (Castillejo, 2016). This fund was divided into four categories of aid, mixing developmental, governance, conflict and border-control work. There is broad acceptance within different EU institutions over the need for a balance between these four categories of funding (Interviews #2, #3, #4, #5). Disagreements have generally been about second-order programming priorities. Discussions on programming priorities tend to be on-going and iterative, with officials finding compromise balances (Interview #2). By 2019, the Fund was worth 4.8 billion euros, 88 per cent of this coming from the Commission. Across 26 eligible countries, 2 billion euros had been spent in the Sahel and Lake Chad, 1.6 billion euros in the Horn of Africa and 0.8 billion in North Africa. From the Fund's thematic categories, 31 per cent had been spent on migration management, 24 per cent on resilience, 21 per cent on economic issues and 21 per cent on governance. Within these rather imprecise category definitions, in practice around two-thirds of the Fund's four billion-euro allocation had been spent on border control projects, the rest for development and governance programmes. (European Commission, 2020).
Very specific internal processes facilitated and reflected the management of polarization. The initiative for the Trust Fund came from DG DevCo and this directorate took the lead in its design; this gave the Fund a development slant from the beginning. The most politicized moment of negotiation came when member states had to agreewithin comitology procedures -their contributions and this triggered debates about what should be understood by the 'root causes of migration' (Interview #12). DG Home then engaged and pressed in particular for the possibility to attach funding to high-level migration dialogues with third countries and to devote important resources to the return of irregular migrants. It was agreed that DG NEAR should then play a primary role to cohere these contrasting perspectives at a political level. DG DevCo and DG NEAR established a memorandum of understanding with the EEAS and DG Home; the balance between different types of funding was then to be addressed at internal Quality Review Meetings that involved the signatories to this memorandum (Interview #12). A group made up of DG DevCo, DG NEAR and Nordic states in particular negotiated to ensure that the number of standard development, economic and civil society programmes increased as programming was set for 2017 and beyond within these internal processes (Interview #9).
Linked to but broader than the Trust Fund as such, other very specific examples also illustrate the kind of policy balances that have become common. At the end of 2019, the EU allocated a new 389 million euros package to Morocco. After long internal negotiations, this was made up of 101 million euros for border control technologypushed most strongly by Spain -and the rest divided between a new development programme aimed at the most vulnerable part of the population and some human rights projects, to win the support of the Commission and some member states uneasy with the more securitized funds (El Pais, 2019). In similar vein, an inter-institutional working group was set up on Egypt that ensured funding there included development, civil society and more security-oriented projects. After initially polarized differencesbetween those wanting to cut funding and those advocating security-first cooperation with the Egyptian regime -this mix was agreed between the different directorates in particular when the Commission negotiated its new support framework for Egypt in 2017 (Interview #9).
Integral to these multi-sector aid packages, those policy-makers responsible from running programmes on economic development, social issues, civil society and governance began to reframe and fine-tune their programming to align it more meaningfully with EU security imperatives. Most in the development community have gradually accepted the basic logic of external funding being linked more integrally to the security and migration agendas. Policy-makers describe this as a 'difficult process' but one necessary to the new funding getting through the labyrinth of internal comitology (Interview #12). Development officials have sought to harness this link to some degree as a positive opportunity, adjusting to a different geopolitical context by framing development dynamics in a more strategic way. Officials acknowledge they have incorporated more strategic logics, links and rationales at the operational level of their development programmingand that this has helped to sell many civil society, governance and social programmes to the Commission leadership (Interview #3).
These officials have been at the forefront of developing a narrative around 'hybrid security' risks that uses politicians' focus on migration and terrorism as an opening into more efforts to redress under-development and bad governance (Interview #4). In particular in the wake of the 2016 Global Strategy these officials used the concept of 'resilience' to justify the need for more development and governance aid (Interview #9). The wider development community, including donor agencies and NGOs, have been concerned to harness increased resources related to migration by ensuring that these are spent in accordance with 'local needs' and rights-based objectives (Concord, 2017(Concord, , 2019. The instrumental logic behind aid has given rise to more thinking in recent years on how development drives security, in some ways making it more important as a policy domain (Interview #3). This is an especially prominent dynamic under the new Commission that took office in December 2019 and that has defined itself as 'geopolitical' in intent.

The Development-Humanitarian Nexus around the Syria Conflict
A second polarization-dampening dynamic is the way the EU has combined strategic, development and humanitarian goals in the spill-over impacts of the Syria conflict. The EU allocated large increases in humanitarian aid to the region from 2015. A part of these increases has been channelled through multilateral bodies; member states promised an extra 200 million euros for the World Food Programme and the UN Refugee Agency to manage refugees within the Middle East. With a high political profile, the EU allocated six billion euros to the Turkish government to host refugees in Turkey. Gaining broad agreement from different parts of the Commission on this package was possible because the funding was ring-fenced for vulnerable refugees and did not go directly to Turkish authorities (European Commission, 2019).
Yet perhaps the most innovative new instrument was the EU's Madad Fund that was set up specifically to channel support for Syrian refugees with an original set amount of one billion euroshalf of this from the European Commission, half from member states. The Madad Fund's mandate is to support refugees in the states around Syria, especially Jordan and Lebanon. Since 2015 it has provided social assistance to these refugees and basic developmental services like access to water and health facilities. Its focus has increasingly homed in on strengthening the social resilience of host countries. The fund does not support core security functions, so is different from the Trust Fund for Africa; nor does it offer standard humanitarian aid, as ECHO does.
The Madad Fund is of particular interest here because the internal processes through which it was created and the way its operational scope was defined helped manage polarization. The fund was created as a new instrument to join together existing country aid envelopes and to bypass recipient governments' control over these. Italy was the fund's strongest supporter and an influential architect of its unique remit. In contrast, a majority of member states initially did not support the creation of a new instrument and a small number of them sought actively to block it. They were concerned the fund might divert attention from core EU strategic aims or cut across other finding. After internal refinements, these doubters came on board. By 2020, a full 21 member states had made bilateral donation to the Madad Fund, in addition to the UK and Turkey. By this stage, 2.2 billion euros had been committed to the fund, with 1.7 billion euros contracted -well above the original target. Many of those states that originally opposed the fund had by now topped it up on more than one occasion (Interview #9).
This management of internal differences occurred due to a number of factors. The fund was initially driven by DG DevCo; DG NEAR was then handed primary responsibility to manage its programmingthe fund was formally located within this directorate. The EEAS was given a role in relation to the ensuring the coherence of the fund's strategic orientation, while the humanitarian office ECHO was also brought in on programming decisions. In this way those managing the fund have ended up playing what they describe as a 'bridging' role between different parts of the EU institutional machinery. Each has had an incentive to be engaged positively. DG DevCo has gained a presence in middle-income countries like Lebanon and Jordan. The EEAS has been able to tie its strategic oversight to concrete funding on the ground. ECHO has been able to ensure its emergency aid dovetails with longer-term development aid and fuller engagement from security officials. In addition, several member state diplomats were posted to the fund to help the interface with national governments and ensure their concerns have been taken on board (Interview #9).
The fund has functioned in a way that is complementary to other more politicized strands of funding. Alongside Madad, ECHO has continued its standard humanitarian work and DG Home its migration-control programming. In this way, internal processes ensured the fund would not trample other funding lines' autonomy. A key factor in overcoming member states' doubts was the decision not to operate inside Syria itself. In this way the fund helped overcome several member states' concerns that humanitarian relief might empower the Assad regime. Another helpful move came in 2015 when the fund's remit was widened to include Iraq and the Balkans; this responded to some member states' rising strategic concern with instability inside Iraq and with other member states' concerns over the Balkans migration route into Europe. In this very concrete way, the fund's core developmental work aligned itself with governments' core security priorities.
In all this, the increasing prominence of a 'development-humanitarian nexus' has helped bring in a wider range of actors beyond the longer-present security-development nexus (Interview #9). In this way the EU has found a common line within the context of differences over the Syrian conflict itself. In the particular context of the MENA, the development-humanitarian fusion has enabled the EU to engage in addressing the ramifications of the Syrian war even if there was no appetite for military intervention. Concretely, it created an area of commonality to offset differences between the French government's push for more funding to the Syria opposition and the more cautious German line against such political intervention (Interview #1).
EEAS diplomats have not wanted funding to go to the Assad regime in a way that cuts across their aim to create a political process that puts pressure on the Syrian government; the Madad Fund's remit represented a compromise that took this on board and ensured the development-humanitarian nexus contributed to the Union's broad stance on the conflict (Interview #1). Cyprus had threatened to block the six billion-euro package in Turkey but within talks on the Madad Fund was supportive of funds here going specifically to refugees in Jordan and Lebanon, not Turkey (Interview #7). Development officials and MEPs on the EP Development Committee sought to prevent funds going directly to arm combatants; the Madad Fund explicitly excluded such kinds of intervention (Interview #5).
In sum, funding around Syria has been aligned to longer-term development, civil society and governance programmes (Koenig, 2017). The humanitarian-development nexus has widened to include resilience building, with political, development and humanitarian actors working in tandem with each other (Anholt and Sinatti, 2020, p. 318). This approach to security is one that has commended relatively wide support from different actors within the EU (Riddervold, 2018;Palm and Crum, 2019). Even as they differed on many other questions in the region, member states, the EEAS and different parts of the Commission agreed that this kind of fusion would be central to the EU's 'comprehensive approach' to conflict resolution (Dany, 2015). There have certainly been sharper internal debates in the last several years over the right balance between short-term emergency relief and long-term development and capacity-building programmes (Bergmann, 2018). The Madad Fund was the EU's answer to these differences and a way of massaging them into a more or less merged approach.

Aid Cuts as Democratic Leverage in Turkey
In a third dynamic, the EU managed polarization by using select aid reductions in response to political developments in Turkey. Relations with Turkey have been a contested issue within the EU for many decades. Turkey's bid for EU membership has long attracted relatively polarized positions from member states and different EU institutions. Yet, decisions over aid allocations were generally not highly polarizing until the mid-2010s. Turkey has received pre-accession assistance since 2002. The funds are managed by diplomats and bureaucrats, and until recently had been shielded from political debates. Much of this has changed, however, in the last several years, at the EU level and within member states (Soler i Lecha et al., 2018). As elsewhere, the EU has used its different strands of funding in Turkey in an increasingly instrumental fashion.
In particular, it has sought to use its funding as leverage to reverse Turkey's authoritarian drift. Turkey received a hefty 1.3 billion euros from the Pre-Accession Financial Assistance between 2002 and 2006, a hugely increased 4.8 billion euros between 2007 and 2013 from the renamed Instrument for Pre-Accession Assistance (IPA) and 4.5 billion euros for 2014-20, the so-called IPA II tranche (Republic of Turkey Ministry of Foreign Affairs, 2019). However, in the last three years the EU has reduced pre-accession aid to Turkey as a response to the country's move away from democratic norms. The Commission cut Turkey's allocation of pre-accession funds by 253 million in 2018, 400 million in 2019 and has recently decided to cut another 483 million in 2020 (Interview #6). Overall, Turkey will receive only around 3.2 billion euros of the 4.5 billion euros of pre-accession aid that the EU initially allocated to it for the 2014-20 period. The EU has additionally diverted some support from the government into civil society (Zihnioğlu, 2019, p. 507). Aid has moved from being a relatively technocratic instrument to a tool deployed for political leverage.
The internal debates over these aid cuts shed light on the management of polarization. Until recently IPA amounts were set without strong debates and few actors or officials engaged with the technical details of how the funds were spent. Only Cyprus expressed serious reservations about so much aid going to Turkey (Interview #7). This permissive consensus began to crack as Turkey's government moved in an increasingly autocratic direction. After several years of incremental authoritarian drift, Turkey lurched more dramatically away from any semblance of democracy after the attempted coup of July 2016. President Erdogan introduced a state of emergency that allowed him to rule by decree and to restrict civic space by banning gatherings. The crackdown soon expanded into a purge across all sectors of society as critics were progressively silenced (Zihnioğlu, 2020, pp. 100-101). By 2019, democracy indices such as Varieties of Democracy, Economist Intelligence Unit and Freedom House listed Turkey as an outright autocracy.
The cuts in pre-accession aid represented the EU's most concrete policy response to these political developments. In this case EU aid has become more strategically instrumentalized in a reactive way, because of developments inside Turkey. Initially there was significant polarization over this issue. There were clear differences between member states and also between different EU institutions over what steps the Union should take. Austria wanted to suspend membership talks altogether; France formally supported continued engagement with Turkey (Baczynska, 2016). A shift in German positions was key. While German officials had underlined the importance of continued engagement with Turkey, Angela Merkel in 2017 came out firmly against Turkish accession to EU (Rieger, 2017). After Turkey's arrest of Die Welt journalist Deniz Yücel, she noted in October 2017 that 'the entire rule of law in Turkey is moving in the wrong direction' and added that she would back a reduction in pre-accession funds (Barkin, 2017).
The EP passed resolutions in 2016, 2017 and 2019 calling for the accession process to be halted (European Parliament, 2016. The European Parliament's chief budget negotiator argued that the EU should not finance Turkey at all as long as it was 'drifting further away from European democratic standards' (Emmott, 2017). In 2019, the EP moved to suspend parliamentary exchanges with the Turkish parliamenta decision that was hotly contested within the chamber, reflecting intense debate over the best tactics to employ against an increasingly truculent regime (Interview #5).
Some actors called for a more emollient line. The European Commission has argued that funding remains an important means for influencing legislative and policy reforms (Interview #6). While he is at best lukewarm towards Turkish accession, French President Macron has underlined Turkey's role as an 'essential partner' specifically in the 'migration challenge and the terrorist threat' (Deutsche Welle, 2017a). Gradually a consensus took shape among member states around a reduction in pre-accession aid rather than any stronger measures against Turkey's abuse of democratic norms (Deutsche Welle, 2017b). Many MEPs, including Kati Piri, the Parliament's then-Turkey rapporteur, also supported aid cuts but not formally removing membership from the table (Gotev, 2017).
After the Commission's proposal for reducing the IPA funds, there was intense dialogue among different EU institutions (Interview #8). Trade-offs softened the emerging polarization over the EU's Turkey policy. The EP pushed for the aid cuts to be larger than the Commission initially proposed. Officials in the EEAS and DG NEAR worked to convince the hardest line MEPs to back away from supporting a complete withdrawal of aid. MEPs in turn insisted that funds to Turkish civil society and Syrian refugees were increased as aid to the government was cut (Interviews #2, #5). Cutting aid was the compromise between doing nothing and suspending accession negotiations.
In practice, paralysis in accession talks has anyway taken away part of the rationale for 'pre-accession aid'even if it is still useful in prompting certain modernization reforms. Given the relatively limited size of the IPA funds and cuts to them, diplomats saw this punitive conditionality as fairly non-intrusiveespecially as Erdogan himself had already been willing in de facto terms to turn away from the prospect of EU membership (Interview #11). The EU has not formally suspended the accession process, although a 2018 Council statement recognized that in de facto terms it has come to a standstill (European Council, 2018). The EU also decided on further cuts to pre-accession assistance for 2020 as a response to Turkish gas drilling in the East Mediterranean (European Council, 2019).
The cuts in pre-accession aid did not impede the 6 billion-euro migration-control aid package. Even those states most critical of Turkey's human rights record saw this as strategically necessary. EU officials stress how strongly member states and most parts of the EU institutional structure insisted on delinking the migration deal from democracy-related aid cuts (Martin, 2019). In 2019, Turkey asked for a further tranche of EU aid to support these refugees; most member states and the Commission were broadly supportive (Euractiv, 2019;Nielsen, 2019;Interview #7). Even as EU concerns have grown over Turkish military actions in northern Syria and Libya, member states have not cut this part of external funding to Turkey; this an agreed line from relatively tough France and the UK at one end of the spectrum through to Hungary at the other end (Brzozowski, 2019;Michalopoulos, 2019). On migration-control, Turkey has effectively imposed conditionality on the EU by regularly threatening to re-open its border (Niemann and Zaun, 2018). It eventually did just this in early 2020.
In sum, the relatively smooth, technocratic use of EU funding within Turkey until 2017 has given way to a far more complicated situation in which aid has been dragged into the political tensions that have intensified between the EU and Turkey. To some extent, this intra-EU polarization has been curtailed because Turkey has become so deeply authoritarian and so clearly at odds with the EU on major strategic questions. The very severity of the foreign-policy problems related to Turkey has helped push member states and EU institutions towards a measure of agreement. In addition, the aid cuts were a half-way step between those in the EU institutions and member states reluctant to take any restrictive measures and those wanting to suspend all talks. Moreover, the area of broad agreement that was found within the EU was that aid cuts related to pre-accession preparations should not cut across aid increases for migration-control; significantly, member states attached less immediate strategic value to pre-accession aid than to migration-related aid. On this latter issue EU aid has become subject to a far more transactional logic (Dimitriadi et al., 2018). Curiously, managing polarization led the EU to cut pre-accession aid but to keep migration-control aid on a separate trackthe first responding to events in Turkey, the second an awkward response to the 2015-16 migration surge.

Conclusion
The article shows that politicization has crept into decisions about EU external funding across the MENA region. This involves the three dimensions of politicization: higher salience, polarization and wider actor engagement. The heavily instrumental use of external funding for strategic priorities has raised the salience of debates about development aid, made these debates more political and less technocratic, and generated polarized opinions on some questions between different EU actors. Debates over the way that EU aid is used in the MENA have gained in salience and relevance across a larger number of institutional actors as they are tied increasingly tightly to deliberations about high-level strategic aims related to the region. Aid to the Middle East and North Africa is no longer a low-politics matter but has been deployed with a more instrumental purpose for key security objectives.
In respect of polarization specifically, the article also reveals the dynamics that have served to limit or contain the severity of internal disagreements, both between different EU institutions and between member states. Polarization has been diluted or managed by a combination of inter-issue balancing and geoplitical adjustment in the use of aid. The perception of deep crisis in European security has helped temper internal disagreements. Polarization has increased but has not assumed overwhelming proportions. The high political salience of migration and other security issues has encouraged officials concerned with development aid projects to change the way they design and justify external fundingthis is the sense in which external funding has adjusted to a new geopolitical environment.
The more instrumental use of EU external funds in the MENA region has not entirely emasculated development policy, but has rather given external funding a more eclectic profile. This shift has been sufficient to engender signficant unease among many policy-makers within the EU institutions and some member states, but without polarizing positions to the extreme of decision-making paralysis. In Turkey the dynamic was different but also helped narrow polarization: here EU institutions and member states reached compromises on how to use aid to respond to political developments in Turkey, showing how the use of aid conditionality provided a tactical means of managing polarization. In this sense, our account supports the notion of multiple policy cleavages helping to soften polarization.
While other case studies in this volume (Lauwers et al., 2021;Plank et al., 2021) show how strong politicization in other policy domains leads to strong politicization in external funding, our account shows how different factors can rather mitigate some (if not all) elements of horizontal politicization. The choice between development and other funding-related aims is not a stark, binary either-or. Different perspectives have been accomodated within a rejigged external aid profile that includes traditional developmental initiatives and more security-oriented programmes. And in Turkey this was possible through the EU cutting pre-accession aid while increasing other, more immediately instrumental migration-control aid. This kind of counter-balancing is somewhat different to internal migration policy, where polarization over the ill-fated quota system brought the EU to such a standstill in this area that the scheme was abandoned. This kind of do-or-die polarization has not featured in the politics of external migration-related funding (Börzel and Risse, 2018). The security-development-migration nexus may be more aspirational than real, but it is not entirely absent from EU policy calculations.
All this suggests that the different dimensions of politicizationsalience, polarization, actor-engagementdo not necessairly go hand in hand with each other. An increase in salience can encourage member states and EU institutions to compromise on their differences and soften polarization's roughest edges. It can serve as one among many drivers of a wider dynamic of collective securitization in EU foreign policyit is in this way that debates over politicization and the wider analytical frameworks of EU external identity interact with each other. At the high political level, a relatively firm consensus has taken shape that the EU and indeed member states need to use external funding to contribute more tangibly to other strategic goals. Exactly how this is to be done is a tactical dilemma that has occasioned significance differences of opinion within the EU's different institutions and also at member-state level. Recent years have seen the deepening and containment of politicization go hand in hand with each other within the EU's various MENA policies.
In sum, politicization over EU funding has become more significant in the MENA region but acts upon policy in a second-order fashion: that is, mainly influencing tactical balances in the way that more strategically oriented policies are implemented. In some senses, this politicization may be overdue and positive. Yet, it means aid is now expected to fulfil very diverse functionsfrom helping development, to stopping migration, to enhancing security and defending democratic norms. Clearly, these functions commonly cut across each other, meaning that EU aid is at risk of losing its clarity of purpose, and even more so as the challenges coming from the Middle East become more complex and difficult to manage.