On 20 July 2015, an Islamic State (ISIS) suicide bomb attack killed 33 and injured more than 100 mostly pro-Kurdish young activists in the majority Kurdish town of Suruç in south-eastern Turkey. That same day in nearby Adıyaman province, an alleged attack by the Kurdistan Workers’ Party (PKK) killed a Turkish corporal. This marked the breakdown of a two-and-a-half-year ceasefire between the PKK – listed as a terrorist organisation by Turkey, the U.S. and the EU – and the Turkish state. It was also the start of a violent cycle that has taken at least 2,981 lives, about three times more than during the July 2011-December 2012 escalation, when Crisis Group confirmed almost 1,000 deaths.
Among the deaths confirmed through Crisis Group’s open-source data collection, nearly half were PKK militants (1,378), followed by state security force members (976) and civilians (408). The remainder (219) were “youths of unknown affiliation”, a category created to account for confirmed urban deaths, aged 16-35, who cannot be positively identified as civilians or members of the PKK or its urban youth wing.
Violence peaked between February and May 2016 when fighting erupted in some urban districts of south-eastern Turkey for the first time in the conflict’s 33-year history. The PKK had built up an armed presence in the region during the 2012-2015 peace process. Around one third of all deaths occurred in Hakkari province’s Yüksekova district, Şırnak province’s Cizre and Silopi districts, Şırnak’s provincial centre, Mardin province’s Nusaybin district and Diyarbakır province’s Sur district. In June 2016, the conflict moved back to its traditional rural arena. Since then, around 90 per cent of all deaths, as tracked by Crisis Group, occurred in rural south-eastern districts.
The PKK or its affiliates have not carried out any major attack in the country’s urban centres and the west of Turkey since December. U.S. pressure, intense operations by the Turkish military and PKK’s strategic considerations appear to have contained its attacks. Nonetheless, Ankara is alarmed by the boost of PKK’s self-confidence especially following the U.S. decision to arm the YPG, the PKK’s Syrian affiliate, for the Raqqa offensive. Once the Raqqa offensive ends, the likelihood of military confrontation may increase if U.S. engagement wanes or if Turkey decides to strike the YPG in north-west Syria. Moreover, the Kurdistan Freedom Falcons (TAK) – a PKK offshoot widely believed to maintain links to the organisation – threatened new attacks on Turkish cities and tourist sites in a statement issued 6 June. These dynamics could herald an increase in violence in the coming months.
What follows are five main conclusions derived from the data collected over the last two years by Crisis Group. The analysis focuses on the last eight months of the conflict (December 2016-July 2017), but also looks at broader trends.
1. PKK attacks on local ruling party officials have intensified Turkish nationalist feelings
The recent killing of Justice and Development Party (AKP) political figures and civilians in the south east has heightened nationalist feelings in Turkish society. Since March 2017, when the Turkish military intensified its operations against the PKK, there have been seven attacks on political figures and civilians in the region. All are widely assumed to have been carried out by the PKK, though so far it has only taken responsibility for three of them.
Attacks claimed by the PKK:
On 9 June, a group of militants attacked the official car of Kozluk mayor (Batman province) Veysi Işık, killing 22-year-old music teacher Aybüke Şenay Yalçın, who was walking by.
On 16 June, Necmettin Yılmaz, a primary school teacher, was abducted while driving his car which was found burned in Tunceli province. The PKK announced that Yılmaz was “penalised” for collaborating with Turkish security forces. His body was found in a nearby river on 12 July.
On 1 July, AKP deputy head of Diyarbakır’s Lice district Orhan Mercan was shot dead in front of his house. The women’s branch of PKK’s urban youth wing, YPS-Jin (Civil Protection Units-Women), claimed responsibility, alleging Mercan was spying for the state and trying to recruit Kurdish youths as spies.
Attacks attributed to the PKK:
On 9 March, gunmen wounded Tayfun Ayhan, AKP head of Esendere town in Hakkari province’s Yüksekova district, and killed his brother, Murat Ayhan, while they were carrying out campaign activities for the presidential system referendum.
On 15 April, the motorcade of AKP Muradiye district head Ibrahim Vanlı was attacked in Van province. Vanlı’s nephew Adnan Vanlı, himself a village guard, was killed.
On 1 July, AKP deputy head of Van province’s Özalp district Aydın Ahi was shot dead.
On 8 July, two lorries and two cars were attacked in Hakkari’s Yüksekova district. Four civilians, Harun İbişoğlu, Dursun Doğan, Sadık Aktaş, and Hüseyin Kartal, were killed.
The PKK’s targeting of AKP political figures is probably an attempt to show the state, the AKP and its local supporters that it can still carry out dramatic attacks, despite an intense military crackdown. But this could backfire: instead of weakening the government and the president, these attacks are strengthening AKP support. Reports of slain security officials, political figures and civilians dominates the national media, justifying, in the eyes of many, harsh anti-PKK operations. Fuelling nationalist fervour over the past two years has allowed Ankara’s political leadership to consolidate support for its agenda, strengthening its alliance with the Nationalist Movement Party (MHP), which helped it win the April 2017 presidential system referendum. Meanwhile, PKK attacks also allow Ankara to justify the prosecution of some opponents by labelling them “terrorists” or “collaborators”.
As the AKP continues mobilising nationalist segments of society and banks on its alliance with the MHP, it is unlikely that the political leadership will return to a more constructive agenda addressing Kurdish demands or resuming peace talks in the medium term for two reasons:
Following the April referendum, the AKP still relies on its alliance with the MHP – which opposes concessions to address Kurdish aspirations – to pass new internal parliamentary regulations and other adjustment laws.
Two significant elections are scheduled for 2019: local and presidential. The AKP and the president will continue to rally nationalist constituencies to mobilise support.
2. An escalatory cycle of increased IEDs against security forces and intense Turkish military operations
Around 60 per cent of all security force fatalities since July 2015 were caused by improvised explosive devices (IEDs). During June and July of 2017, the number of IED attacks in rural areas increased, mostly along roads leading to military bases, including an attack on 17 July in Hakkari’s Yüksekova district that injured seventeen soldiers.
There are several reasons why militants use these devices:
IEDs are relatively cheap and easy to employ;
They can inflect greater casualties by targeting a group of security personnel with a single explosive;
Since IEDs can be detonated from a distance, militants can avoid direct armed confrontation, limiting their own risks.
This occurred amid intense military operations against the PKK in rural south-eastern districts which most recently focused on Lice district in Diyarbakır, Çukurca district in Hakkari and Başkale and Çaldıran districts in Van. The objective is both to target militants and destroy PKK ammunition depots.
Beginning in March 2017, the Turkish military carried out what it described as its most intense operations in years, deploying about 7,000 soldiers, special forces, police officers and village guards. In March, at least 79 PKK militants were killed, up from 23 in February, thirteen in January and six in December that Crisis Group could confirm.
3. Fewer PKK attacks prior to the April referendum
The number of security force members killed was relatively low in the run-up to the April referendum. While seventeen security force members were killed in PKK attacks in the two-and-a-half-month period prior to the referendum, this number almost quadrupled with 67 security force fatalities in the two-and-a-half-months after the referendum. There were no such fatalities in February 2017, the first month without security force being killed since July 2015. Aside from two urban IED attacks in Diyarbakır in January and April, the PKK or its affiliates did not carry out major, dramatic attacks in urban centres in Turkey and in the west of the country in the months prior to the referendum.
Sources close to the PKK told Crisis Group that they made the strategic decision to hold back in order to avoid generating further nationalist support for the “Yes” vote. As noted above, increased military operations by Ankara and U.S. pressure likely also played a role in curbing attacks. Regardless of the reason, this suggests the PKK has the ability to determine the timing and intensity of its attacks as well as to control its militants on the ground.
In the run-up to the referendum both the PKK and the government used the threat of violence to rally their supporters. Cemil Bayık, a PKK leader, warned on 9 April that if the “Yes” camp won, “the war” would intensify. AKP officials raised the spectre of PKK terrorism in their campaign for a “Yes” vote. President Erdoğan on 2 March equated voting “No” with supporting Qandil (a reference to the PKK’s headquarters in northern Iraq). Other officials said a “Yes” vote would put an end to all terrorist activity in the country whether carried out by the PKK or by what the government calls the Fetullahist Terrorist Organisation (FETÖ), allegedly led by Fethullah Gülen, an exiled cleric living in the U.S. who the government blames for masterminding the 15 July coup attempt.
Ankara saw the referendum results as vindication of its hard-line approach: it interpreted the fact that the “Yes” camp received 10 per cent more votes in the south east than the AKP got in the 2015 elections as demonstrating support for its strategy to “eradicate” the PKK. In turn, a strengthened president mobilising nationalist support made it easier for the PKK to legitimise in the eyes of its supporters resort to violence.
4. Violence more dispersed in the south east
Although security force and PKK militant deaths have remained largely concentrated in the south-eastern provinces of Hakkari, Şırnak, Mardin and Diyarbakır, violence was more dispersed in the south east over the past eight months. From December 2016 to July 2017, around 50 per cent of all confirmed fatalities occurred in these four provinces, compared to around 70 per cent in the previous sixteen months (July 2015-November 2016). Fatalities resulting from military operations rose in other south-eastern provinces, in particular Tunceli (28 confirmed militant deaths in February) and Bitlis (38 confirmed militant fatalities in May and June). Crisis Group also confirmed 29 militant fatalities in northern Iraq in March and April as a result of cross-border airstrikes by the Turkish military.
Confirmed fatalities in Tunceli, Bitlis and northern Iraq resulted from the military’s decision last spring to intensify its efforts to track down militants in mountainous areas and conduct cross-border operations in northern Iraq. At the same time, the military has been winding down operations in urban districts (such as Cizre, Sur, Silopi, Nusaybin, Yüksekova) located mostly in the provinces of Hakkari, Şırnak, Mardin and Diyarbakır, another factor possibly explaining the extension of deaths to other provinces.
5. Kurdish village guard deaths have increased since April 2017
Fifteen members of the state-funded Kurdish Village Guard were killed between April and July 2017, a slight increase compared to the first four months of the year. Ankara has ramped up recruitment for these paramilitary forces since January 2017 when it retired 18,000 guards over the age of 45 in a force that totals about 50,000. It plans to recruit 25,000 new paid guards, between 22 and 30 years old. Turkish media outlets reported that the newly-hired guards would be equipped with heavy weapons and take part in operations against the PKK.
Civilians interviewed by Crisis Group in Nusaybin in early 2017 confirmed that Kurdish-speaking security personnel – most likely village guards – participated in anti-PKK operations. Guards receive two weeks of basic military training immediately after joining plus supplementary training once a month.
The government recently revitalised its system of “neighbourhood guards” in urban areas, probably in response to the PKK's urban tactic last year. The government plans to place neighbourhood guards in urban stations around the country to assist police and the military in maintaining “public order”. T