Urban Guerilla Activities in Greece








 Technological Forecasting & Social Change, 72, No 1, 2005






     Logistic fits are made to the populations of attacks by urban guerilla groups in Greece, under the assumption that these organizations grow like species.  The analysis is mainly based on data that cover the attacks of the two major urban guerilla groups, 17N (Revolutionary Organization November 17) and ELA (Revolutionary Popular Struggle). We conclude that urban guerilla activities in Greece may have been triggered into existence by the military junta but probably had their roots deeper into the earlier political system in Greece that repressed leftist movements. Our analysis shows that the life cycle of political violence in Greece had already been completed when the police finally began cracking down on the 17N, which takes away the shine from the police’s achievement.


Keywords: Urban guerilla groups; Terrorism; 17 November; E.L.A.; May 1st; Logistic; Life cycle


1. Introduction


     In the beginning of the new millennium political violence seems to be concentrated around the notion of terrorism. In the United States political violence focuses on singular large-scale attacks on the population such as the Oklahoma bombing and the World Trade Center.  In Europe political violence is expressed for decades now through a multitude of smaller-scale less singular acts committed by self-proclaimed revolutionary terrorist groups such as the german Red Army Faction (RAF), the italian Red Brigades, the french Action Direct, and the greek Revolutionary Organization 17 November (17N). These groups target individual symbols of the social system they are trying to undermine.  A slowly evolving pattern of action over the years makes such groups behave like distinct “species” inside the “ecology” of politically violent organizations. For this reason we employ the term “urban guerilla groups” to distinguish them from other organizations that employ violence as a means of achieving their goals.  

     Greece’s urban guerilla activity finds itself at the end of a life cycle. The recent dismantling of the 17N group brought to light detailed data on numbers, dates, and types of attacks.  Such data are amenable to a diffusion analysis. The evolution of 17N and similar groups during the last 30 years in Greece can be analyzed as that of a “species” growing in competition along a logistic pattern with a pre-determined potential for action. Cesare Marchetti originally employed this approach when he studied the attacks of the Red Brigades.[1] Here we apply it individually to the 17 November and other urban guerilla groups in Greece. Finally we consider the origins of all political violence in Greece, expressed through the composite action of all urban guerilla attacks in Greece put together.

     In analyzing the activities of a guerrilla group as a species growing in competition we have generalized the concept of competition. As the group realizes its potential by accumulating hits, competition appears among potential hits for the group’s choice. This is not unlike competition among words for an infant’s attention while the infant acquires vocabulary.[2]

     Our approach emphasizes the logistic character depicted by the evolution of the data. Subsequent attempts to attach political events to dates or “explain” things, either by us in this paper or by the reader, unavoidably carry the uncertainties inherent to “interpretations” and are consequently subject to debate.



2. The data


     Attacks by urban guerilla groups have been reported in daily newspapers, books, and the Internet. But the data are not always reliable and with good reason. Besides those hits for which some urban guerilla group publicly claims responsibility, there are the hits that police attributes to a group without proof, and also the hits confessed by captured members of the group who later retracted their confession. Two of our sources are quite reliable. One is a book on 17 November written by G. Kassimeris,[3] and the other is a series of articles containing a discussion of each and every event related to 17 November published in the daily newspaper “To Vima”.[4]  Finally, we have cross-checked and enhanced our data on the other large urban guerilla group in Greece, the Revolutionary Popular Struggle (ELA) from the following sources:

·         The weekly newspaper, Kyriakatiki Eleftherotypia, issue of August 11, 2002, pages 6-7

·         The daily newspaper Eleftherotypia, “Known and Unknown Resistance to the Junta”, issues of April 21 and 22, 1997 (in Greek).

·         The daily newspaper, Kathimerini, issue of August 18, 2002, page 7

·         http://www.enet.gr/online/online_p1_obj.jsp?pid=62940560,26&tp=A,T&id=56280624

     From the analysis point of view the most robust data set perhaps is those hits that led to an assassination and were subsequently followed by a declaration to the press accepting responsibility for the attack. Involving unambiguous and well-defined acts with political motivation, this data set gave the most complete overlap with a logistic curve, hallmark of a species growing into a niche.



3. The analysis


     Our assumption is that a well-defined urban guerilla group with members sharing political ideas and the belief that violence can be employed as a tool to achieve this goal behaves effectively like a biological species.  Therefore its growth pattern should be logistic. In analogy to a species population, the population density of its hits can measure the growth of an urban guerilla group. The assumption underlying this thesis is that the group has a finite capacity for action, due to limited material resources, skills of its members, and a hostile non-supportive environment. Therefore the total number of hits an organization has to its credit by a certain date shows how far the organization has advanced toward realizing its capacity for action, i.e. its growth potential. By fitting a logistic to the cumulative number we expect the group’s activities to undergo a life cycle and eventually cease. Considering that we are dealing with a group, this is more reasonable than fitting a logistic to the population density (as one may have done with a rabbits population), which would imply that the group in question would remain in operation forever.



3.1.  The 17 November


     The logistic function has the form:

where M, a, to and C are constants. The last constant is introduced to account for cases where the logistic does not begin at zero level. This can happen because in the framework of the conspiracy that invariably underlies the formation of an urban guerilla group, the group’s potential begins to be expressed before the “official” first hit. Organization, planning, and even “rehearsal” of attacks (as revealed in testimonials of 17N members) all constitute means of expression. The results from the fits, including values for the correlation coefficients, are tabulated in Table I.

     Figure 1 shows 17 November attacks by two accounts. Whether we look at the Kassimeris data or at the enhanced data set from the newspaper “To Vima”, we obtain similar logistic curves (comparable midpoint and time constant—see Table I). Both sets are amenable to a fit of reasonable quality and both sets show complete growth processes. The only difference is in the level of the respective ceilings, because the “To Vima” data also contain a number of criminal acts (mainly armed robberies) that had not been associated with 17N until the arrest of its members.  Some of these acts are not yet fully attributed to individuals and there have also been retractions of some confessions.  Nevertheless the two data sets give consistent results on the logistic’s midpoint and time constant, which enhances our confidence on the approach. The few last data points in each set indicate a tendency to deviate from the logistic pattern, something that can be interpreted as “abnormal” behavior. There have been documented cases where a pronounced deviation from a natural-growth pattern is met with resistance.[5,6] The intensified efforts of the police during 2002 can be interpreted as evidence for such resistance that eventually culminated with a crackdown.

Figure 1.  Logistic fits on two sets of data for hits by the 17 November. The growth processes are complete. Notice the tendency of the data to deviate from the natural trend at the very end right before the group was dismantled.


     The set of hits involving a political assassination is a more appropriate data set for studying the evolution of 17 November because these events better reflect the ideas behind the group’s existence (many of the robberies and bank hold-ups may have been instigated somewhat arbitrarily by lower-level members in the organization). Indeed, the logistic description of the assassinations data set turned out to be textbook-like, see Figure 2.

Figure 2. Assassination hits by 17 November. The lower graph is the derivative of the gray line in the upper graph and depicts the life cycle of the phenomenon.


     The data seem to cover the entire growth process. The last victim, Brigadier General, Steven Saunders, assassinated in June 8, 1999, brought the total number of deaths to the 96% level of the logistic’s ceiling. This makes the dismantling of the organization in 2002 a foregone conclusion because it is generally accepted that a species’ extinction can be considered as a “natural” death whenever the growth process exceeds the 90% completion level.

     But of interest is also the fact that the nominal beginning of the logistic (the 1% level) points to April 1973 a time period that witnessed several pertinent events. On February 14, 1973 the police charged on 1500 students assembled at the National Technical University of Athens (NTUA), in the first of a series of violent crack-downs on student protests that followed, arresting some 100 students. On February 21-23, 1973 an unprecedented two-day occupation of the Law school by 4000 students culminated with an exodus of the students, a violent clash with the police, and many arrests. It is understandable that such events solidified political violence against the junta.  Resistance to the military junta culminated with the student uprising and occupation of the NTUA on November 14, 1973. When the junta’s police stormed the university grounds three days later (November 17), there were many deaths. The event marked the beginning of the process that lead to the junta’s downfall. It also provided a raison d’être and name to the 17N urban guerilla group



3.2.  E.L.A. and May 1st


     In addition to the 17N there are two other significant urban guerilla groups, the Revolutionary Popular Struggle (E.L.A.) and the May 1st. The latter has only a handful hits to its credit whereas the former has over one hundred. Moreover, both organizations have declared in the press their merging into a single group in 1990, which justifies treating them as one “species”. However, the timing of the appearance of May 1st warrants special attention.

     The activities of May 1st are concentrated between 1987 and 1989 in the wake of a long quiet time stretch 1982-1985 during which urban guerilla group activities in Greece were rather dormant.  In fact 17N in one of their press communiqués declared a “grace period” between 1981 and 1983 suspending all activities. This action was taken out of respect for the electorate’s will, aiming to give the newly elected socialist government the opportunity to put its program into effect. It is likely that the other groups also adopted this notion of a “grace period. We therefore distinguish two waves of hits before and after the grace period: The first wave is dominated by E.L.A. The second wave consists of E.L.A. and May 1st taken together, See Figure 3.

Figure 3. Punctuated by the long quiet stretch 1982-1985 the two waves of activity are amenable to logistic descriptions. All activity of these two groups ceased abruptly in 1995.


     The first wave reached 94% completion and can be considered exhausted, but the second wave reached only 77% completion. We can consequently conclude that this urban guerilla “species” suffered an unnatural death. Another interesting aspect of the two waves has to do with their beginnings. With the logistic function determined we can look back to estimate the nominal beginning of the process, namely, the 1% level. We thus see that the first wave has its roots in 1969 and the second in 1974. Both waves have 8 to 9 early “missing” hits that represent activities possibly conceived and planned but not realized.

     It is worth noting that 1969 marks the most intense resistance to the junta as expressed through bomb attacks by a variety of groups (see Figure 5 further down) active during 1967-1974.  Moreover, this point in time comes in the aftermath of the May 1968, a crucial period in modern history that inspired many leftist groups.

The beginning of the second curve—1974—is evident as it coincides with the fall of the junta following the Turkish invasion of Cyprus in July 1974. It is likely that a new generation of ELA members, politicized after the fall of the junta, took over from the older generation that had resisted the junta. 

The abrupt end of ELA’s second wave—a “premature” death—could be attributed to a fortuitous breakthrough by the Greek police. By some journalistic accounts Greek police exploited information originating in the records of the former Eastern Germany secret service Stazi.




    3.3 All urban guerilla groups put together


     Besides the groups mentioned so far, there have been several others of lesser importance that came into action after 1985.  In addition there have been several bomb attacks in the period 1967-1974 in the general framework of resistance against the junta. There is a qualitative difference between violent acts directed against a junta and violent acts directed against a democratic government. Nevertheless it is worthwhile combining all violent activities together in order to get a picture of the long-term evolution of political violence in Greece.  Figure 4 shows all attacks reported in Greece since the beginning of the dictatorship in 1967.

Figure 4. The smooth lines show logistic fits on the corresponding sections of the overall data curve. The data represent all attacks, including unsuccessful ones. The data cover only the later part of the large logistic curve (thin black line) that reached more than 99% completion. The indicated 1% levels point to the nominal beginnings of the other two logistic curves (in light gray and dark gray respectively). The levels of completion are also indicated.


During the years of the military junta we see much activity by numerous urban guerilla groups in Greece. In Figure 4 one can discern a number of steps—each one a good candidate to be fitted by a small logistic curve—probably representing the comings and goings of various groups. (Post-1985 groups include “Epanastatikoi Pyrines” and “Antikratiki Pali” while some 10 other groups were active against the junta between 1967and 1974.  For the sake of simplicity we fit only three logistic curves. The first curve, of which we see only the later part, reaches more than 99% completion in the early 1980s. The curve has its origin (1% level) around 1919.

     The second curve (light gray line) describes activities mainly attributed to E.L.A. The curve reaches 90% completion in late 1983 and has its roots in early 1965. The third curve (dark gray line) describes activities mainly attributed to 17N. It reaches completion (99%) in mid 2002 and has its roots in mid 1978. 4. Discussion of results in a historical perspective


     The major thrust of political-violence attacks in Greece has deeply seated roots in the early decades of the twentieth century. This was a time period influenced by the Russian revolution, as well as a period during which the Greek royalty disappointed the public with disastrous undertakings (Asia Minor catastrophe). Moreover, in the wake of World War II, a civil war in Greece (1944-1949) between the defeated left and the conservatives fueled politically violent activities further.  Lacking adequate documentation for such individual activities we have lumped them together into the general civil war conflicts. I It remains to be seen whether the first sigmoid of Figure 5 that has its origin in 1919 becomes corroborated with additional data whenever and if they become available. It is not surprisingly that among the documented activities attacks proliferate during the junta years.

     More recently we distinguish two waves of activities by urban guerilla groups in Figure 5, which echo the two waves of E.L.A. in Figure 3. The first wave has its roots in 1965, which was a year of intense social unrest that led to the takeover by the military junta two years later. In 1965 the centrist government of George Papandreou collapsed when several of its members defected into a new government scheme aligned with the then king. Besides the shift to the right, 1965 is marked with the unpunished killing during a demonstration of a young student of economics, Sotiris Petroulas, member of the left youth organization “Lambrakides”. “Lamprakides” took their name from Gregory Lambrakis a prominent parliament member of the left. A well-known sportsman, Lambrakis had been several-times champion at the Balkan Games and was a physician and professor at Athens University. He was assassinated in 1963. Petroulas’ death as well as that of Lambrakis two years earlier became highly emotive events among the revolting Greek youth. Mikis Theodorakis composed a song for Petroulas and Costa Gavras directed the film “Z” recounting Lambrakis’ story. Incidents of violent resistance accompanied the junta right from its inception.

     However the first “planned” assassination by an urban guerilla group came only in 1975, after the junta fell and elements from the resistance groups evolved toward a different type of armed struggle, this time not against the junta. It is estimated that some 25 out of more than 200 attacks against the junta, were carried out by individuals that would later continue as members of the urban guerilla groups active during 1975-2002.

     The first wave of post-junta attacks was completed (to 90%) by late 1983. There followed the quiet period 1982 to 1984 known as the "grace” period mentioned earlier. The second wave of attacks, however, does not have its roots at the end of the grace period but in 1978, see Figure 5. Student unrest in Greece showed up ten years later than in other western countries for “technical” reasons (the military dictatorship). The period 1978-1979 is marked by the occupation of universities, extreme student protests, and the proliferation of many small leftist groups sometimes resulting from the splitting of larger political organizations. Recruiting new members for urban guerilla groups was all too easy in the politically charged environment of the time.

     The first urban guerilla group E.L.A. has its roots in 1969 during the harsh early years of the repressive military regime, and one year after the worldwide revolution-inciting events of 1968, see Figure 3. Its birth coincides with the year of the most intensified violent resistance against the junta.  The first wave of hits by this group was completed (to 94%) by 1984. Following the grace period there was some reshuffling among the various groups. Some E.L.A. members went to 17 November while E.L.A. absorbed all of May 1st. The emerging new “species”, however, appears to have its conception deeper in the past—in 1974—just as does the 17 November. The tragic events of November 17, 1973 at the NTUA, where people were killed and the military government violated human rights, seem to be at the base of both urban guerilla groups. Finally and abruptly E.L.A. ceased activities in 1995 having accomplished only 77% of it potential. This “unnatural” death conceivably reflects a precautionary decision on the part of the group given that the police had purportedly obtained crucial information on its activities, as mentioned earlier. Alternatively an arrangement such as a merger with the 17 November cannot be excluded.

     The 17 November by all accounts had completed its life cycle in 1999. On the assassinations graph, Figure 2, we see the 96% level (normally scheduled for mid 2002) all ready achieved in June 1999. From Figure 1 we could surmise that some of the recent robbery-type attacks were “going too far” in the sense that they brought the hits curve well above the corresponding logistic curve. The most recent of these “unnatural” acts has proven detrimental to the group’s survival.



5. Conclusion


Urban guerilla groups in Greece began organizing themselves  shortly after the military junta came into power but a deeper need for violent revolution seems to have its origin much further back in the age-old repression of leftist movements. The most celebrated urban guerilla group, 17 November, survived a generation of unchecked activity but was finally silenced as it reached the end of its life cycle. Other organizations such as E.L.A. and May 1st had all ready gotten quiet in the 1990s. All acts of political violence have now ceased in Greece.

Concerted diminishing activities by guerilla groups take away the shine from the triumphant claims of recent police achievements. Our analysis shows that political violence in Greece as measured by the population density of hits by urban guerilla groups, has followed natural-growth curves that were completed when the police finally stepped in. Arguably an “aging” terrorist “species” displayed weaknesses that played in the anti-terrorists’ hands, and the police finally killed a dead mouse!

Alternatively, one could say that the achievements of the Greek police were not all concentrated in the summer of 2002 but that continuous struggle over the last decades steadily chipped away on the fitness of the urban guerilla groups, making their final eradication possible.

Table I


Data set






17 November all hits







17 November all hits

(“To Vima”)






17 November only assassinations






E.L.A. 1st wave






E.L.A. 2nd wave (includes May 1st)






All attacks: large-scale curve






All attacks: 1st wave






All attacks: 2nd wave








[1] C. Marchetti, On Time and Crime: A quantitative Analysis of the Time Pattern of Social and Criminal Activities, invited paper to the Annual Interpol Meeting, Messina, Italy, October 1985. Also report WP-85-84, November 1985, International Institute of Applied Systems Analysis, Laxenburg, Austria.

[2] T. Modis, Predictions, Simon & Schuster, New York, 1992.

[3] G. Kassimeris, Europe’s Last Red Terrorists - The Revolutionary Organization November 17, Hurst & Company, London, 2001.

[4] S. Theodorakis, The End of History- The Hits of November 17, Insert in the greek weekly newspaper “To Vima”, (32 pages) July 28, 2002 (in greek).

 [5] C. Marcheti, Fifty-Year Pulsation in Human Affairs, Analysis of Some Physical Indicators, Futures, vol. 17, no. 3, 376–88, 1986

[6] T. Modis, Predictions – 10 Years Later, Growth Dynamics, Geneva, Switzerland, 2002.

[1] Athanasios G. Konstandopoulos is the founder and head of the Aerosol & Particle Technology Laboratory of CERTH/CPERI in Thessaloniki Greece. E-mail: agk@cperi.certh.gr.  http://apt.cperi.certh.gr

[2] Theodore Modis is professor at DUXX Graduate School of Business Leadership and the founder of Growth Dynamics, an organization specializing in strategic forecasting and management consulting. http://www.Growth-Dynamics.com.