Drones and Special Forces

The Second Karabakh War was more than a war between two belligerents; essentially, it represented a war between two strategic paradigms: one belonging to the twenty‑first century and the other a relic of twentieth‑century military thinking. Azerbaijan’s achievement of air supremacy with the help of drones was not only one of the crucial factors that decided the outcome of that particular war; it will almost certainly contribute to the further development of the military art in armed conflicts of various scopes across various theaters in the time ahead. At the same time, referring to this war as a “drone war” does not reflect reality. In this, as in much else, impressionable journalists got it wrong: this could be a proper designation only if the drones had fought also on the ground instead of soldiers. Certainly, the use of drones was decisive; but the laurels of victory in this war do not belong solely to this contingent of the Azerbaijani military: the mobile groups of special forces were also decisive. Each played a critical part in the country’s military achievement. 

We begin with an examination of the area of hostilities and the respective dispositions of the parties to the conflict before getting into a discussion about some relevant aspects of Operation Iron Fist—the official designation used by Azerbaijan for the Second Karabakh War. We conclude with remarks of a more geopolitical nature. 

Applied Geography

We define the war zone on the day of the commencement of the hostilities on 27 September 2020 as consisting of Azerbaijani sovereign territory occupied by Armenian forces—that is to say, the bulk of the former NagornoKarabakh Autonomous Oblast (NKAO) and the seven surrounding districts. Throughout this essay, the term “Karabakh” will be used interchangeably with the term “war zone” to refer to this area, which is characterized by a complex relief with a predominance of mountainous and high‑altitude terrain. Indeed, the average elevation is 1,100 meters above sea level, with insignificant flat‑hilly terrain in the southern part along the Aras River valley. The conduct of war in mountainous terrain is considered as a fighting activity under special circumstances in which the capacity of the attacking side to use military equipment and artillery, as well as transfer reserves, is sharply reduced. Standard military textbook tactics for a classical frontal attach operation in such terrain indicate that the proportion of attacking troops to defending troops should be six to one. The arrangement of Armenian positions on the dominant peaks above the Azerbaijani positions created additional advantageous conditions for the defense of Karabakh by Armenian armed forces.

The Second Karabakh War represented a war between two strategic paradigms: one belonging to the twenty‑first century and the other a relic of twentieth‑century military thinking.

These geographical advantages were over time supplemented in various ways. Thus, after the First Karabakh War, Armenian authorities ordered the construction and then the strengthening of three successive lines of defense commonly designated the “Ohanyan Line” (named after Seyran Ohanyan, who served from 2000 to 2007 as a “minister of defense” of the breakaway ethnic‑Armenian statelet in Karabakh and from 2008 to 2016 as minister of defense of Armenia). This defense line consisted of a series of fortifications involving barrier fences, minefields and barbed wire, and cemented firing points. After the barrier strips, an around 7‑meter‑high earthen rampart was built. 

After the Four‑Day War that took place in April 2016, the Armenian command further reinforced its defensive positions. Thus, several rows of anti‑tank ditches were dug along the lines of defense, with widths of between to 4 and 6 meters and depths of between 3 to 7 meters. This was understood to be an additional deterrent against a conventional armored assault: in the event of an attack, tanks were expected to get stuck in these ditches and become sitting ducks to be fired upon with impunity.

Also, in the wake of the Four‑Day War, the Armenian military dug an entire network of secret passages and shelters behind each defense line. Throughout the defense line, round‑the‑clock surveillance systems were installed on the stocks, as a result of which the Armenian forces gained the capability to detect enemy movement at a distance of up to 350 meters.

In addition to reinforced defense lines, three fortified defense areas for infantry and artillery, featuring multi‑storey bunkers constructed with reinforced concrete, were constructed in the areas of Fizuli, Aghdam, and Aghdere.

The conduct of war in mountainous terrain is considered as a fighting activity under special circumstances in which the capacity of the attacking side to use military equipment and artillery, as well as transfer reserves, is sharply reduced. 

Planning for a large, combined arms military operation on the scale of the Second Karabakh War required Azerbaijan’s military‑political administration to precisely survey the Armenian forces’ protective capabilities, properly assess various dangers of a financial nature, carefully take into account the domestic political circumstances in both states, and accurately gauge the likelihood of constricting reactions by outside powers both in the neighborhood and more distantly. The course and outcome of the war shows that all of Azerbaijan’s calculations were correct: for instance, Russia did not effectively support Armenia whilst Azerbaijan received Turkey’s full military and political support.

There is some dispute in expert circles about the precise number of military equipment and weapons used by the parties in the war. We do not have the ambition to fill in all the blanks, but we can assert with confidence that although Azerbaijan lagged in the number of operational tactical missile systems, it had an advantage in the quantity and quality of military hardware.  

Moreover, the quantitative proportion in manpower looked as follows: all power structures were involved in fighting on both sides, and partial mobilization was announced. During the war, Azerbaijan had around between 130,000 and 140,000 military personnel in active service, whereas Armenia’s strength was estimated to be between 60,000 and 65,000 military personnel. Apart from Azerbaijan’s quantitative superiority in the field, there was also a qualitative disbalance: the number of professional contract servicemen in the Azerbaijani armed forces was higher. 

Moreover, Armenia could not develop all the necessary countermeasures for the defense of the territories they held in Karabakh on the basis of lessons (they should have) learned in the wake of the Four‑Day War. This is when Azerbaijan first put into practice a strategy of deploying mobile groups of special forces from the rear to the front and also practiced the use of kamikaze drones. One reason for the Armenian failure to learn from this experience may be that its military leadership had been trained in Soviet and Russian military institutions that teach outdated maneuvers. Whatever the reason, in the Second Karabakh War they were unprepared to respond to innovative tactics involving the use of drones and mobile groups—although the Armenian forces did establish reconnaissance and assault battalions, which were staffed mainly by contract soldiers. But this was not enough. Moreover, presumably due to its modest economic circumstances, Armenia also failed to upgrade existing military hardware and purchase modern high‑precision weapons systems. In addition, available financial resources were allocated irrationally. For example, Armenia purchased several Su‑30SM fighters—this sort of air power is not muchneeded for countries with a small territory to defend; on the other hand, Yerevan did not purchase either night vision systems or upto‑date means of secure military communication equipment.

The 2016 Four‑Day War proved the correctness of the concept of using tactical mobile groups of Azerbaijani special forces against fortified platoon strongholds in mountainous areas. Relatedly, we can say that Israeli multifunctional missiles of the Spike family and kamikaze drones also proved to be effective weapons of war for Azerbaijan. Moreover, certain conclusions were made after the successful operation of the Turkish army in Syria (Operation Olive Branch), during which Bayraktar attack drones were massively used. Azerbaijan purchased attack drones and electronic warfare systems. Various communication systems were also purchased to create a single information field. Ground and air hardware was modernized for the use of high‑precision weapons. In short, by the beginning of the Second Karabakh War, the concept of using drones and mobile groups and all its components in an integrated fashion was ready for execution. 

By the beginning of the Second Karabakh War, the concept of using drones and mobile groups and all its components in an integrated fashion was ready for execution. 

As a parenthetical remark, we can state that the military disposition of the parties prior to the Second Karabakh War can also be understood in light of ratings produced by globally authoritative indices. Here we can refer to two such ratings. First, the Global Firepower (GFP) military index, which ranks each country’s potential war‑making capability across land, sea, and air fought by conventional means. It incorporates values related to manpower, hardware, natural resources, finances, and geography, broken down into over 50 individual factors, which are then used in formulating the finalized GFP ranks. GFP’s 2020 edition ranked Azerbaijan sixty‑fourth and Armenia one hundred and eleventh.

Second, the Global Militarization Index (GMI), which is published by the Bonn International Centre for Conflict Studies. It depicts the relative weight and importance of the military apparatus of one state in relation to its society as a whole and uses a number of indicators to represent the degree of militarization of a country. Three overarching categories are employed to determine the results: military spending in relation to GDP and health spending, the relation of military personnel to the total population and physicians, and the number of an armed forces’ heavy weapons in relation to the total population. In 2020, Azerbaijan’s GMI was ranked third in the world whilst Armenia’s GMI was ranked fifth in the world. 

Azerbaijan’s first strike in the Second Karabakh War destroyed up to 30 percent of Armenian artillery and up to 60 percent of Armenian air‑defense systems.

Operation Iron Fist

Aside from combat operations, the Second Karabakh War also had a simultaneous cyberspace component. Back in July 2020, a group of hackers begins to publish photos and passport data on Azerbaijani hacker forums of several hundred Armenians, including employees of the Karabakh National Security Service. However, beginning in September 2020, the number and quality of attacks rose to unprecedented heights. Azerbaijani hackers also published leaflets on Facebook about the inventory of the military units of what its proponents call the Karabakh Defense Army. 

The biggest cyberattack carried out by the Azerbaijani special services took place on 27 September 2020—the day the war began. The mobile operator operating in the occupied lands suffered massive interruptions, which significantly complicated mobile communications and impeded internet access. Viruses also blocked the work of most computers with IP addresses originating in the territories under Armenian occupation. Azerbaijani hackers also hacked into many official ‘Artsakh’ websites.

Operation Iron Fist, which consisted of a coordinated air‑ground assault campaign, covered the entire front line, with several directions identified for offensive action: the main was the southern one whilst two auxiliary ones were in the north (Murovdag Ridge) and in the northeast (the village of Sugovushan).

In the morning of 27 September 2020, Azerbaijan’s missile and artillery units launched a massive strike on more than 500 reconnoitered targets along the entire line of contact and into the depths of the territory occupied by Armenia. The greatest concentration of fire was in the main southern direction. Strikes were carried out on Armenian military positions, equipment storage facilities, and ammunition depots. This first coordinated strike destroyed up to 30 percent of the Armenian forces’ artillery. 

Integral to Iron Fist was a critical operation conducted by the Azerbaijani Air Forces to suppress Armenian air‑defense systems and establish Azerbaijani air supremacy. Thus, a complex plan was executed with the express intention of forcing Armenian forces to put into operation all their anti‑aircraft missile systems. 

This plan consisted in launching an air offensive that had the appearance of a massive air strike in which various types of aircraft, helicopters, and drones were used. For instance, Azerbaijan launched inexpensive light An‑2 aircraft into the zone of destruction of the Armenian air‑defense system. Each had been equipped with bombs so that an Armenian failure to shoot them down would have allowed these aircraft to be used as kamikaze drones, for each had been assigned a specific target. But shooting them down also resulted in an Azerbaijani military advantage, for the idea was to force the Armenian forces to waste expensive missiles attached to their Tor‑M2KM anti‑aircraft missile systems, which they had in limited quantities, on what were effectively false air targets. (This tactic was so cumulatively successful that in some stages of the war, the Armenian forces could not make any use of their anti‑aircraft systems due to a lack of missile availability.)  

Azerbaijan acquired and maintained air supremacy over virtually the entire war zone from day one.

The wave of An‑2s was quickly followed by the deployment of Harop, Skystriker, and Orbiter‑1k kamikaze drones, together with Bayraktar attack drones, which all moved in to target pre‑selected Armenian air‑defense equipment.  

To support the drone offensive, various forms of electronic warfare (EW) were also put to active use to jam and otherwise interfere with the functioning of Armenian anti‑aircraft missile systems. Moreover, false electronic targets were created to further confuse Armenian air‑defense systems. In parallel, strikes were carried out on these same air‑defense systems by Spike NLOS missiles mounted on helicopters, which had a range of up to 30 kilometers. 

Thus, by the end of the first day of the Second Karabakh War, up to 60 percent of Armenian air‑defense systems had been destroyed. At virtually no point during combat operations was Armenia able to control the sky above its defensive positions: Azerbaijan acquired and maintained air supremacy over virtually the entire war zone from day one.  

Here we must break off the narrative to make the following observation: some experts and analysts have claimed that the dominance of drones in the air was due to the weakness of Armenian air‑defense systems. This is quite simply incorrect. Rather, it was the effective application of a tactically sophisticated and innovative plan by the Azerbaijani Air Forces that quickly and decisively incapacitated a potentially capable air‑defense system, including, inter alia, four technologically up‑to‑date vehicle installations brought in from Armenia for the mobile launch of Tor‑M2KM missiles. That being said, even had the Armenian forces been equipped with only the latest air‑defense systems, the agony of their defeat would only have been prolonged—ultimately, the result would still have been the same. 

This is also a good place to note that the Azerbaijani Armed Forces also received out of area intelligence support by the Turkish Air Force. Thanks to the constant duty of AWACS aircraft and Bayraktar drones flying near the Turkish‑Armenian border, Azerbaijani command posts received realtime information on all troop movements from Armenia to Karabakh.

After the massive artillery barrage, Azerbaijan’s ground units began breaking into the first line of Armenian defense in the northeastern auxiliary and southern main directions. In the northern direction, special forces groups began to liberate strongholds on the high‑altitude Murov Ridge. There were also simultaneous attacks in the central Aghdam direction—the shortest distance to Khankendi (the town is still called Stepanakert by Armenians—a name imposed in 1923 by the Soviet authorities in homage to Bolshevik revolutionary Stepan Shaumian, nicknamed the “Caucasian Lenin”). Only later, after the war—i.e., once the mine maps of the liberated Aghdam region were transmitted to the Azerbaijani side—was the decision not to advance directly in that direction vindicated, notwithstanding the expressed intentions and plans of some Azerbaijani generals. 

In some areas, artillery duels quickly ensued. Thanks to its use of drones in the war zone, Azerbaijan had better situational awareness and destroyed Armenian artillery with accurate strikes. Still, the advance of regular Azerbaijani ground troops was very slow at times for a combination of reasons: the mountainous terrain, reinforced defensive positions (e.g., earthen ramparts, anti‑tank and anti‑personnel minefields, engineering barriers), and the order by headquarters to protect the lives of soldiers as much as possible. 

Nevertheless, there were gaps in the Armenian defensive positions in all directions, and Azerbaijan’s mobile special forces groups began to penetrate them. They infiltrated deep into Armenian defenses, secretly entered their rear, and effectively destroyed their strongholds, all of which created the conditions for the advance of regular ground units. The first line of defense in the southern direction began to collapse, which allowed Azerbaijani assault battalions to liberate several frontline villages. On 3 October 2020, for example, it became possible to liberate several villages in Boyuk Marjanli, a village that had been itself liberated a bit earlier, after which the Armenian forces put up virtually no more resistance in that part of Karabakh. Thus, Azerbaijan launched an offensive in two directions: along the Araz River valley and in the direction of the city of Fizuli.

Although mobilization was announced in Armenia, it was not successful. Some of the men of draft age were abroad, while others chose to evade the draft. With the advance of the Azerbaijani army deep into the war zone, members of the Armenian forces began deserting—and from units that were located not just in Karabakh, but in Armenia (i.e., not in the war zone per se). By the end of the Second Karabakh War, the number of desertions had grown to 10,000—a huge figure for a small army.

As the situation on the ground began to deteriorate for the Armenian forces, the Union of Armenians of Russia (a diaspora organization headed by Ara Abramyan, who is simultaneously both the head of the World Armenian Congress and small political party in Armenia) together with VOMA (a paramilitary training organization whose acronym stands for “The Art of Survival”) reportedly worked through social media networks in Russia and elsewhere to recruit mercenaries and volunteers to participate in hostilities. When recruiting, preference was given to snipers and operators of anti‑tank missile systems. And not only persons of Armenian citizenship were recruited—the case of a Russian citizen, Eduard S. Dubakov, is a case in point. The exact number of people who were recruited in such a manner and fought in the war on the Armenian side is unknown, but there is video footage of several VOMA battalions on the internet that shows them participating in the Second Karabakh War. 

The most difficult breakthrough of the front line was in the northeastern auxiliary direction of the war zone, taking several days. Between 2‑3 October 2020, Azerbaijani troops managed to liberate the villages of Talysh and Sugovushan. Soon after the approach of Azerbaijani subdivisions to the town of Aghdere, the battle took on a positional character. To demoralize the Azerbaijani population, the Armenian side began to strike Terter and Barda with artillery (i.e., Azerbaijani cities not in the war zone); in this sector of the front, Armenian artillery positions were sheltered in enclosed concrete bunkers, which prevented them from being effectively counterattacked and destroyed. Artillery shelling from both sides lasted with varying intensity almost until the end of the war. 

However, attack drones operating in this direction destroyed most of the Grad and Smerch rocket artillery systems, and in late October 2020 kamikaze drones tracked down and destroyed two Scud missile launchers at the border with Armenia. On 4 October 2020, Armenian forces began heavy ball and rocket attacks on Ganja, Mingachevir, Gebele, and Kurdamir—Azerbaijani cities all outside the war zone—which resulted in civilian casualties. In response, the Azerbaijani military in the following days staged a punitive counterattack along the entire front line against Armenian military targets.

After ensuring the breakthrough of the front, Azerbaijani artillery units transferred fire to the depth of the war zone. Defense nodes, command and control posts, and support roads were hit. A missile strike by the LORA operational tactical missile system damaged the main bridge bisecting the Lachin corridor.

I n ground battles, meanwhile, the Azerbaijani special forces had infiltrated to the rear of the Armenian lines, flanking them whilst also relying on reconnaissance units. Constant aerial reconnaissance and a developed unified network of data exchange turned all attempts of counterattacks by the Armenian army into a “fire bag.” The largest of these, which was announced by Armenian prime minister Nikol Pashinyan on 7 October 2020, was to take place in the direction of Horadiz. Its stated purpose was to encircle the Azerbaijani troops that had broken through the southern front line—a repeat, it was hoped, of what Armenian forces had accomplished in the First Karabakh War. But this time, the Azerbaijani forces had laid a trap, with the result that about 1,000 Armenian soldiers found themselves surrounded before suffering defeat. According to various sources, the losses amounted to between 600 and 700 Armenian servicemen. In fact, all three major attempts by the Armenian military during the Second Karabakh War to counterattack, close the gap, or encircle Azerbaijani units that had broken through the defensive lines failed. 

Moreover, in fierce fighting, the Azerbaijani military began to move north of the Araz, in the direction of the former NKAO itself. Azerbaijani units approaching the Fizuli fortified defense area surrounded the city from three sides. Combat reconnaissance indicated that a frontal attack would lead to largescale losses in both material and manpower. Thus, two developments took place in parallel: part of the ground attack forces continued their advance towards the town of Jabrayil. 

At the same time, Azerbaijani Su‑25 specially modified attack aircraft were deployed over Fizuli—at the time of production, these aircraft had been optimized for combat use at low and medium altitudes; later technical work made it possible for the Azerbaijani Air Forces’ Su‑25 to fly at high altitudes and carry laser‑guided bombs jointly produced by Turkey and Azerbaijan. During the war, Su‑25s performed more than 600 combat sorties. Mi‑17 helicopters equipped with Spike‑NLOS and LAHAT missiles also worked on ground targets. An analysis of amateur video footage of the fighting showed that Azerbaijan’s Mi‑24G attack helicopters (i.e., South Africa’s modernized version of the Mi‑24) were rarely used.

Azerbaijani troops managed to enter the rear of the main Armenian military groups in the war zone. The opposing side swiftly began to lose its ability to fight once the liberation of cities in the south and the advance in the northeast began to bear fruit. Many settlements, including the town of Jabrayil, were liberated on 4 October 2020. After fierce fighting, Agoglan was liberated on 9 October 2020. This was a key turning point: after the fall of Agoglan—demoralized and having problems with both command and support—the Armenian side was able to organize only focal nodes of resistance on dominant peaks near large settlements. Here we can underscore that the conduct of the “Yarasa” special operations forces unit belonging to the Foreign Intelligence Service of Azerbaijan contributed to the growing disorganization, creation of panic, and general chaos in the ranks of the Armenian forces. Around this time, the failure of the Armenian information campaign to explain the video reports of the Ministry of Defense of Azerbaijan showing the liberation of various towns in Karabakh, as well as the secret forays of Azerbaijani special forces units to take “selfies” in the cities under the control of Armenians, became apparent.

In the sky above Karabakh, Azerbaijan’s drones continued to dominate and collect their tribute from the battlefield. Outstanding results were achieved by Bayraktar reconnaissance and strike drones. Thanks to their high‑precision ammunition, they destroyed hardware and manpower, enhanced the precision of artillery attacks, carried out target designation for bombs dropped from Su‑25 attack air craft, and served to guide missiles of the TRLG‑230 multiple‑launch rocket system. Directly with its ammunition, Bayraktar drones destroyed some 50 to 60 percent of their targets—quite an impressive figure.  

We underline that the Second Karabakh War was the first war in which drones were successfully used so massively. Aside from its military effectiveness, the use of drones also inflicted psychological pressure on the Armenian forces. There is more than one instance of a drone dropping a single bomb on a single tank in a large convoy that resulted in the crews of all the other tanks in the convoy quickly getting out of their fully functional vehicles and running away on foot. 

We would also like to note the advantages provided by a unified information field system—an encrypted tactical communication system with the ability to broadcast video online, such as the NATO Command, Control, Communications, Computers, Intelligence, Surveillance, and Reconnaissance (C4ISR). As a result of this technological edge, Azerbaijan’s entire military was able to maintain close coordination in real time, which made it possible to immediately respond to a change of scenery and halt emerging dangers.

We underline that the Second Karabakh War was the first war in which drones were successfully used so massively. 

Here is a good place to say something more about the decisive role played by the mobile groups of special forces in securing Azerbaijan’s victory in the Second Karabakh War. In the context of Azerbaijan, the “special forces” are composed of the following: the special forces of the Azerbaijan Land Forces (the army), a separate special purpose brigade located in Nakhchivan, and small special units belonging to the Ministry of Internal Affairs and the State Border Service. 

Now then, apart from the first days of the war (and on other rare occasions) when it was necessary to tear through the defensive line solely with regular troops, in front of Azerbaijani ground forces were special forces mobile groups that attacked Armenian forces, liberated territory and settlements (villages, towns, cities), and secured captured defense lines.

We can provide a few examples. After the capture of the city of Agoglan, several mobile special forces groups launched an offensive from there in the direction of the city of Fizuli. They took back strategic heights and villages and were then reinforced by two motorized rifle brigades that took full control of those liberated territories. 

Special forces groups coming from the rear of the Fizuli fortified defense line liberated the city on 17 October 2020. 

A group of special forces trekking along the Araz River valley advanced towards the city of Zangelan. During this period, Armenian assault aviation flights reached their maximum number. Yet in this period five Armenian Su‑25 attack aircraft went down in that part of the war zone: one collided into a mountain on 28 September 2020 and four others were in shot down by Azerbaijani air‑defense systems. Here we underline the accurateness of the information and contrast it to what was propagated by various bloggers, including Van Hambardzumyan, and biased military experts from Russia. 

We can skip ahead to something that happened a little later, when troops in the southern direction, having freed the city of Zangelan on 20 October 2020, reached the state border with Armenia, where they were met with a hastily erected tent of Russian border guards. In this context, it may be useful to provide an excerpt from the statement of the former Chief of the General Staff of the Armenian Defense Forces, Colonel‑General Movses Hakobyan, on 19 November 2020 at a press conference in Yerevan, “Russia provided the maximum it could have. We should be grateful. It gave us that which we could only dream about, from the very first days of the war.” 

These “dreams” included the transport of weapons from Russia territory by aircraft via several air bridges. Later, the facts of Russia’s transfer of weapons—in total worth more than $1 billion—became known to the public. Transferred weapons included the Kornet‑D portable anti‑tank guided missile systems. Azerbaijani intelligence detected the transfer of a large batch of these weapons from Armenia to the warehouses of the military unit stationed in the vicinity of Khankendi, which were destroyed with a precision missile attack. This is the context in which Armenia’s official spokesperson during the war, Artsrun Hovhannisyan, had spoken about the roar of explosions that lasted all night long.

According to Turkish investigative journalist Fulya Öztürk, weapons were flown into Armenia from the Russia‑operated Khmeimim Air Base in Syria as well as from the Syrian capital by five civilian aircraft, a proxy militant, and similar sources. Syrian servicemen and ethnic‑Armenians from Syria also played a part. Somewhere in mid‑October 2020, when the Armenian air‑defenses had been almost completely destroyed, several Tor‑M2 anti‑aircraft missile systems arrived in Karabakh after having been transferred from Russia. The air‑defense unit operating this equipment was staffed by Russian servicemen. All this is based on Öztürk’s reporting. 

Around the same time—specifically, on 15 October 2020—the Caspian Flotilla of the Russian Federation conducted unscheduled exercises. In a press release, its defense ministry reported that these exercises took place in the central part of the sea, north of the Absheron Peninsula (the location of Baku) and included four ships equipped with the Kalibr‑NK cruise missile complex. Each missile has a range of up to 2,000 kilometers and is designed to strike at ground targets. From the start of these exercise until the beginning of November 2020, Russian combat aircraft carried out numerous violations of Azerbaijani airspace in the northeastern part of the sea border.

Not everything was so unambiguous in the case of the Russian Mi‑24 helicopter, which was shot down on 9 November 2020 just a few hours before the signing of the tripartite statement that ended the Second Karabakh War. This helicopter was flying at low altitude at night between 1 and 1.5 kilometers from the Nakhchivan section of Azerbaijan’s state border with Armenia.

But to come back to the main thrust of our narrative. Azerbaijani troops in the southern direction turned north towards the city of Gubatli, which was liberated on 25 October 2020. After this, the offensive continued along the Khakari River valley towards the city of Lachin. The Armenian units were hit hard with anti‑tank systems and mortars, which made it impossible for their defensive operations to continue. More and more units of the Azerbaijani army were introduced into the battle, one of which began to advance in the direction of the village of Krasny Bazaar and another in the direction of the city of Khojavend.

The decisive Battle for Shusha featured an audacious operation by Azerbaijani special forces. 

Heavy fighting was taking place across the width of war zone, and the location of the Armenian side’s positions on the dominant heights in the mountainous and wooded terrain greatly complicated the advance of even regular mobile infantry groups. Azerbaijani drones switched to destroying military vehicles delivering reinforcements and supplies along mountain roads, which became a growing problem for the Armenian command. By 22 October 2020, advancing Azerbaijani troops were already six kilometers from Lachin, the gateway town to the eponymous land corridor between Armenia and the former NKAO. In an attempt to somehow weaken the attack on the Lachin corridor, Armenian forces counterattacked using infantry units in mountainous forest areas whilst setting up ambushes along the main supply routes in the area used by the advancing Azerbaijani forces. One success of these ambushes was the defeat of an Azerbaijani advance party consisting of one T‑72 tank, four BMP‑2 armored infantry vehicles, and five Sandcat armored vehicles.

The Battle for Shusha

It is within this context that began the decisive battle for the city of Shusha: through an audacious operation by Azerbaijani special forces. Operating in small groups, these mobile units began to seep through the mountains and forests on foot through territory occupied by Armenian forces and concentrated around Shusha. In a coordinated way, they took control of the settlements and roads located near the city. The Armenian army, shackled by battles elsewhere, was unable to transfer reserves to reinforce the more than 2,000 troops that had been present in the city. Late in the evening of 5 November 2020, Azerbaijan special forces mobile groups had reached the Lachin corridor road and, in combination with the destruction of a key bridge over the Khakari River (as mentioned above), were able to block the arrival of Armenian reinforcements trying to help defend Shusha. At the same time, advanced Azerbaijani ground units were breaking through to the village of Dashalti, which provided access to the road leading to Shusha. During this period, unfavorable weather had come to Shusha (e.g., fog, low clouds) and for three or four days Azerbaijan could not use drones in that part of Karabakh. Drone flights were also curtailed because of Armenian electronic warfare systems: Yerevan had purchased several types of electronic warfare equipment from Russia in the wake of the Four‑Day War and was able to make effective use of these at that point. Here we can note that throughout most of the Second Karabakh War, Azerbaijani radio engineering reconnaissance was able to detect and guide its electronic warfare systems to destroy such equipment. 

However, during the Battle for Shusha, the situation had become more complex. Azerbaijani troops had encountered problems with communication interruptions, leading to the losses of two Bayraktar drones (depending on the operational situation, such missions were carried out with those drones or with Spike‑NLOS missiles or Harop loitering ammunition) for reasons having to do with a temporary inability to suppress Bayraktar control channels. This was due in part to the Armenian use of the Pole‑21 system of numerous small‑sized sensors installed on antennas and cell towers that jammed the frequencies of GPS positioning systems. Although is not critical for drone operation, it did become more challenging to pilot them with precision and to determine the accuracy of coordinates for target designation.

As it turned out, this was not an effective deterrent for the liberation of Shusha. Onto the scene stepped forward Azerbaijani commandos, armed only with light weapons. Arriving from three directions under the cover of night on 6 November 2020, they began to climb the steep cliffside that serves as a natural fortress‑like defense for Shusha, which was commonly understood to be an unassailable city. Undetected, they entered the city and began to engage in close combat street fighting with the large Armenian force. The Azerbaijani commandos managed to destroy several Armenian tanks and infantry fighting vehicles using grenade launchers and portable anti‑tank guided missiles. The Battle for Shusha eventually fell in a handto‑hand combat operation.

Right after Shusha’s liberation, Armenian forces attempted to retake the city on three occasions before the end of the war. Assault units armed with heavy weapons were preceded by rocket artillery strikes in which the Armenians used all the artillery missile systems in the arsenal, including the TOS‑1 heavy flamethrower systems and the Iskander‑M missile systems.

By 8 November 2020, Armenian forces had lost their strike potential and been definitively pushed back to Khankendi. In addition to the liberation of the city of Shusha, the Azerbaijani Armed Forces had taken control of a junction of roads in its vicinity. The Armenian forces had no strength left for an effective defense of Khankendi, in the area of which armed clashes were already taking place. In short, the Armenian military defeat at Shusha led to the collapse of their entire defenses. 

End of the War and its Continuation by Other Means?

Unquestionably, the Battle for Shusha decided the outcome of the Second Karabakh War. During this period, Russian diplomacy strenuously tried to stop the fighting in order to prevent the full liberation of Karabakh by Azerbaijan and thus the complete defeat of Armenia. The result was the signing of a trilateral statement on 10 November 2020 that is, in terms of scope, more than a narrow ceasefire agreement but less than a general peace treaty: strictly speaking, only its first article deals with the cessation of hostilities in Karabakh; others lay out various concrete measures, including the deployment of a “peacekeeping contingent of the Russian Federation […] in the amount of 1,960 military personnel with small arms, 90 armored personnel carriers, and 380 units of the automobile and special equipment.” 

The Russian peacekeeping zone today consists of that part of the former NKAO that that had not been liberated by the Azerbaijani Armed Forces, plus the Lachin Corridor. According to another article of this document, Russian peacekeepers will remain in that part of Karabakh until at least November 2025. Their deployment “shall be automatically extended by a further five‑year period if none of the Parties [i.e., Armenia, Azerbaijan, or Russia] declares six months prior to the expiration of the period of its intention to terminate the application of this provision.” The document makes no mention of a further extension. 

But further considerations in this direction would enter the domain of speculation. What we can say with certainty is that the Azerbaijani victory in the Second Karabakh War brought the military conflict to an end. Nonetheless, an epilogue to this “hot phase” remains unwritten. For instance, Armenia has not yet officially announced the final figures of losses in manpower and military hardware. But the bottom line is that although we will hear a lot of interesting things about this war, no one can deny that Operation Iron Fist was skillfully designed and well‑executed. This remains beyond doubt. In the entire history of warfare, there have been very few instances in which a victorious attacking side suffered fewer casualties than a defending side in such difficult geographical conditions. 

In the entire history of warfare, there have been very few instances in which a victorious attacking side suffered fewer casualties than a defending side in such difficult geographical conditions. 

The fact that the fighting stopped before a seemingly small step to complete victory produced different emotions in Azerbaijani society, which had tasted the sweetness of victory. That being said, the price Azerbaijan would have paid for the liberation of Khankendi and the continuation of hostilities in what is now the Russian peacekeeping zone likely would have been high—perhaps prohibitively so. According to various sources, at least 7,000 civilians had not evacuated from Khankendi by the end of the Second Karabakh War (and the number of civilians in other parts of the Armenian‑occupied lands at that time was probably at least that much more). There is no doubt that some of these people would have died in the liberation of the city. The deaths of even a few hundred people would have been presented to the whole world as “genocide” or “ethnic cleansing” or something similar. We can recall the international outcry to the massacre of Bosnian Muslims in the Srebrenica enclave in July 1995 or the ethnic cleansing of the Krajina Serbs in August 1995. Moreover, it seems likely that the Azerbaijani Armed Forces would have had to get involved in some sort of hybrid or even explicit military confrontation with a Russian‑Iranian tandem in the event Khankendi had not been liberated but the fighting would have continued in Lachin or Kelbajar. It would be difficult to imagine how such a confrontation would have been in the national interest of Azerbaijan. 

Politics is the art of the possible and Azerbaijan achieved the maximum it could, given the geopolitical realities of late 2020. 

But Baku’s victory in the Second Karabakh War did not resolve the underlying conflict between Armenia and Azerbaijan. We can refer again to the tripartite statement as a whole, which lay out various concrete measures aiming towards a future predicated implicitly on the establishment of peaceful relations between two sovereign states: Armenia and Azerbaijan. On the other hand, we can again underscore that the results of the Second Karabakh War did not completely satisfy segments of Azerbaijani society while at the same time shocked and dealt a deep wound to Armenian society. The specter of Armenian revanchism hangs in the air, as do questions revolving around the restoration of Azerbaijani administrative control over the area encompassed currently by the Russian peacekeeping zone. And, of course, the state border between Armenia and Azerbaijan remains undelineated. 

Much unfinished business needs to be conducted; the weight of some important geopolitical variables is not yet fully known. In some circles, hopes for a final and complete resolution of outstanding issues between Armenia and Azerbaijan are rather low: one of the major external parties to the conflict, namely Russia, is perceived to conduct its policy toward the Silk Road region more in accordance with the principles of conflict management than conflict resolution. Still, the Second Karabakh War changed the military‑political balance of power in the South Caucasus: Turkey’s role as a regional power was strengthened while Russia’s influence was weakened. 

All things considered, the document that ended the Second Karabakh War is qualitatively better than the situation that existed previously. The prospect for a genuine, sustainable peace has never been greater—at least not in the period since both Armenia and Azerbaijan each regained their independence. This, in turn, would suggest that the prospect for reconciliation between not only the two states but also the two titular nations has also never been greater. War may very well be a mere continuation of policy with other means; but the outcome of the Second Karabakh War clearly indicates that the time for belligerence is past. Sure, obfuscation remains a distinct possibility. But ultimately, no good purpose can be served from now on by anyone embracing a doctrine predicated on the falsehood that policy is a mere continuation of war with other means.