Faculty of Arts University of Helsinki Comparing Military Cultures: Warfare in the Aegean Region from the Fourth Crusade to the Early Fifteenth Century Juho Wilskman DOCTORAL DISSERTATION To be presented for public discussion with the permission of the Faculty of Arts of the University of Helsinki, in Porthania 674 (70), on the 4th of December, 2021 at 12 o’clock. Helsinki 2021 Doctoral Program of History and Cultural Heritage Supervisors: Björn Forsén, Heikki Mikkeli, Hannes Saarinen Pre-examiners: Kelly DeVries, Jukka Korpela Opponent: Jukka Korpela ISBN 978-951-51-7704-9 (paperback) ISBN 978-951-51-7705-6 (PDF) The Faculty of Arts uses Urkund system (plagiarism recognition) to examine all doctoral dissertations. Unigrafia Helsinki 2021 ABSTRACT There are arguably two ways of understanding the fundamental processes of military history. One emphasizes unchanging principles in warfare and that conduct of war is largely dictated by the physical realities. The other claims that the ways of war are more-or-less cultural constructs. I approach this “military culture debate” by analysing the land and naval warfare in the area that surrounded the Aegean Sea, which had belonged to the Byzantine empire in the late twelfth century. The period discussed extends from the Fourth Crusade (1202–1204) to the early 1420s. The region and era in question are suitable for testing different understandings of military history. It is a geographical and geopolitical region where two major groups of conquerors arrived in the thirteenth century. The Catholic Christian Latins represent the western European ways of war of the time and the military traditions of the Muslim Turks originated from the steppe and from the east. Both settled in the region and founded realms. Therefore, they had to adjust themselves to the environment. Similarly, local Orthodox Christian, mostly Byzantine, rulers had to adapt to the situation. I compare the composition of armed forces and willingness to engage in battles of the above-mentioned three groups fighting in the region. My study analyses what influence their cultural background had on their conduct in these matters. I provide new information about the warfare of the study era and increase our understanding of why the Ottoman Turks were able to conquer most of the region. I found that the “culturalist” understanding of military history usually had the best explanatory potential in the scope of my research. The belligerents tried to construct their armed forces and fight their wars in a manner guided by their cultural background. To a certain extent they had to adapt to the local conditions, but in the issues dealt with here it was mainly a question of adapting to the social structures and the adversaries, rather than to the climate and geography of the region. The armies shared a growing number of similarities as time went by, but the influx of troops from outside the region and religious doctrines contributed to differences in the composition of armed forces and thus how wars were waged. The Ottomans apparently made the best use of the military potential of regions they controlled. This was probably one of the key reasons why they prevailed in the study area. For cultural reasons, the Christian powers tended to neglect the use of certain groups of people in the military. Thus their forces were more often supplemented with mercenaries and allies whose reliability was uncertain and this had adverse consequences, particularly for Byzantium. 1 As regards engaging in battles, an important issue was how reliable the people living in fortified towns were perceived to be, which influenced how necessary it was to fight in the open. Transporting of siege-equipment, on the other hand, hampered the possibility of avoiding battles. The Latins showed more battle-willingness in their warfare than the other belligerents in the conflicts of the region. There were cultural reasons behind this. It should nevertheless be noted that there was variation in the battle-willingness of the Byzantines depending on commanders, circumstance, period, and adversary. Much of the Turkish and steppe nomad warfare consisted of raids in which battles were usually avoided. However, battle-willingness was more apparent in the conflicts of the great Muslim dynasties, and when the Muslim leaders faced one another, or when they repulsed invasions of Christian powers in their own regions. This battle-willingness of the Turks could come as surprise to the latter. 2 CONTENT Abstract 1 Acknowledgements 3 Content 5 A note on names and transliterations 7 A note on calendars 8 1. INTRODUCTION 9 1. 1. Aim of the study 9 1. 2. Primary sources and methodology 17 1.2.1. Methodology and the use of historical narratives as sources 17 1.2.2. Other kinds of sources 38 1. 3. The military culture debate 46 1. 4. Earlier research 57 1. 5. Physical context and constrains 63 1.5.1. Geography 63 1.5.2. Logistics 71 2. COMPOSITION AND RELIABILITY OF ARMED FORCES 80 2. 1. Latin political entities 80 2.1.1. Fourth Crusade expeditionary corps 80 2.1.2. The Crusader States 89 2.1.3. Venice and Genoa 105 2.1.4. Fourteenth century developments 115 2.1.5. The Hospitallers and the later Crusades 126 2. 2. Byzantine and other Orthodox political entities 137 2.2.1. Armies of resistance 137 2.2.2. Re-establishing military institutions 149 2.2.3. Native versus foreign, “public” versus “private” in the military forces of Theodore II Laskaris and Michael VIII Palaiologos 161 2.2.4. Natives and foreigners during the reigns of Andronikos II and his successors 170 2.2.4. The military forces of the tekfurs 182 2. 3. Steppe Nomad and Muslim political entities 2.3.1. The armies of the Steppe nomads and Selçuks 5 196 196 2.3.2. The Anatolian emirates and the gazi question 209 2.3.3. Ottoman military institutions 226 3. WILLINGNESS TO ENGAGE IN BATTLES 243 3. 1. The thirteenth-century 243 3.1.1. The founding and checking of the Latin Empire 243 3.1.2. Wars of the Successor states up to 1259 258 3.1.3. The re-established Byzantine Empire and the following conflicts 274 3. 2. The fourteenth and early fifteenth centuries 295 3.2.1. Continuities in warfare and the Turkish conquest of the western Anatolia 295 3.2.2. The Catalan Company in Europe, Turkish sea-raids, and civil wars in Byzantium 311 3.2.3. The rise of the Ottomans 326 3.2.4. Warfare between Christian powers during the Ottoman ascendancy 346 4. CONCLUSIONS 360 BIBLIOGRAPHY 373 Primary sources 373 Secondary sources 391 6 A note on names and transliterations With personal names of West Europeans and their non-Hellenized descendants I use the name forms indicating the ethnic background, for example Guillaume for the princes of Achaia from the French origin family of Villehardouin, but Guglielmo for the marquis of Monferrato in Italy. I treat the nobility of Flanders as French-speakers. For the Catholic churchmen (Hospitallers not included) I use Latin names or the Anglicized forms, if these have become common in English. I use the terms Catholic and Orthodox Christian (or Church) in the modern sense. For Turkish names I follow the modern Turkish forms and for Arab and Persian ones, the model of the Encyclopaedia of Islam. With the Greek names, I use The Oxford Dictionary of Byzantium as a guide; the Greek names are thus transliterated without macrons or accents. An exception is made concerning names that have become common in English in Anglicized or Latinized form (thus, for example, Constantine, rather than Konstantinos). If a modern Greek scholar has chosen some other way to transliterate his/her name in studies published in languages using Latin alphabets, I retain it throughout the text (e.g. Savvides and not Sabbides or Savvidis). For Slavic languages, I follow the “modified Harvard system” that is used in The Oxford Dictionary of Byzantium. I use the Byzantine toponyms (taking the Oxford Dictionary of Byzantium as a model) for my study area, i.e. the region around the Aegean Sea that belonged to the Byzantine Empire c. 1200. For the rest of the world, I employ modern toponyms. Thus I speak about Nicaea and not Iznik, but Konya and not Ikonion. In this way, I emphasize how the cultural and ethnic frontiers differed from the modern ones, or the ones in the minds of many historians. For the sake of convenience, I use the same name forms throughout the text. The other reason is that the changes in the ethnic and cultural composition of a population are harder to demonstrate and slower to take hold than changes in political entities. The relations of various ethnic groups vis-à-vis different political entities were complex. I have decided that when speaking about the Greeks, Slavs, Latins (or Franks), Italians, Albanians, Cumans, and Vlachs I am referring to them as members of ethnic groups, which are mainly defined by language, religion, and customs. The political affiliations are secondary. When I use, for example, the terms Byzantines, Bulgarians, Serbians, Ottomans (or a similar name of realm derived from the ruling dynasty), Zagoritas (an Albanian tribe), men of the Principality of Achaia (or another regional realm), or Venetians (or other city-states), I am referring to political affiliations and the ethnicity is secondary. 7 I also count as Byzantines the subjects of the “successor states” that were ruled by Greek-speaking Orthodox rulers (i.e. mainly the Nicaean Empire, the Despotate of Epiros). I can also speak of Turks both as an ethnic group and as realms that were ruled by Turkish lords. This practice is for the cases when it is a question of more than one realm, or it is uncertain which emirate is in question, or the presence of other ethnic groups is uncertain. I treat Cumans and other steppe peoples as ethnic groups, although they might have been more like political confederations, the exact nature and composition of which largely escapes us. Unless otherwise stated I am referring to Macedonia as a region, which includes both the modern Republic of North Macedonia and the Greek province of Macedonia. A note on calendars The Orthodox, Muslims, and some Catholics followed different chronologies than the modern one. The Byzantines usually counted years from the Creation, and the year began on the 1st of September. The Muslim calendar counts years from Muḥammad’s escape from Mekka (Hidjra) in 622 AD, and uses the lunar year; thus the New Year has no fixed place in the solar cycle. I give the modern AD year whenever possible. If only the Byzantine year is at our disposal, for instance 6902, I mark it 1393/1394; i.e., that event has taken place between 1 September 1393 and 31 August 1394. If only the Muslim year is known, I give the AD year and then the Muslim year in brackets, for example 1431/2 (AH 835). 8 1. INTRODUCTION 1.1. Aim of the study There are arguably two ways of understanding military history. Richard Abels has termed these as “scientific” and “culturalist”. The scientific approach emphasizes unchanging principles of warfare, while the culturalist approach claims that these are more-or-less cultural constructs.1 This so-called military culture debate has mainly been discussed on a general and theoretical level.2 I approach the issue in a more concrete fashion by investigating the composition of the armed forces and their willingness to engage in battles in medieval Romania from the Fourth Crusade to the early 1420s. In comparing the belligerents, I analyse what influence their cultural background had on their conduct in these issues. In other words, are scientific or culturalist understandings of military history closer to what we can say about the realities of the time for the analysed issues? In addition, I provide new information about the warfare of the study era and also increase our understanding of why the Ottomans prevailed in the region. With medieval Romania I refer to the geopolitical area surrounding the Aegean, which roughly corresponds the extent of the Byzantine Empire at the time of the Fourth Crusade.3 In the east, the frontier of my study area goes along the edge of the Plateau of Central Anatolia from the region of Sinope on the Black Sea to South-West Anatolia in today’s Turkey. The Aegean islands are included, and the Mediterranean Sea south of Crete is a self-evident limit. In the west is the Adriatic Sea, but the Ionian Islands on the coast of Greece are inside my scope. In the north, the frontier runs between Dyrrachion and Shkodër in today’s Albania, then through the region where nowadays the border of Serbia and North Macedonia lies. In modern Bulgaria, the limit goes along the southern edge of the Haimos- (Balkan) mountains to the Black Sea.4 Abels 2008: 1–4. In the citations I use the form xx: pp to indicate pages, and omit : when referring to verses or chapters. Jeremy Black speaks of “functionalist” instead of “scientific” (Black 2004: 232–234). Guy Halsall has used the terms “substantivist” instead of ”culturalist” and “normative method” instead of “scientific” (Halsall 2003: 6–9). 2 For the “military culture” debate and definitions of “military culture”, see chapter 1.3. 3 As often with geographical names in medieval times, the territory which comprised Romania, varies in different authors. The definition I am using was, however, the most common, and it is the most practical from the point of view of this study. For the various definitions of Romania see e.g. Bon 1969: 305–306; Maksimović 1995: 996–997; Thiriet 1959: 3–4; Balard 1978: 6–7; Bayrı 2020: 15–17. The area should not be confused with modern independent state of Romania. 4 I leave out, for example, the region of Trabzon, which was still part of the Byzantine Empire when the Crusaders arrived in the front of the walls of Constantinople, but was geographically distant, had a strong tradition of autonomy, and did not much involve itself in the wars and politics of West Anatolia during the period I am discussing. On the other hand, I include all the Ionian Islands, although some of them were already conquered by the Norman kings of Sicily in the late twelfth century. It would however be problematic to leave these out when the other Ionian Islands are included. Besides, their rulers played a part in the politics of mainland Greece. 1 9 After the death of Emperor Manuel I Komnenos in 1180 the Byzantine Empire had begun to weaken, and several territories had seceded from the control of the central government. These included the area of modern northern Bulgaria, which is left outside my main scope, where the Bulgarian realm was re-founded with help of the Cumans. These were steppe nomads from the northern side of the Black Sea. The Fourth Crusade set out from Venice in 1202, and in less than two years this led to the conquest of Constantinople. The Byzantine Empire collapsed, and was divided between the Crusaders and the Venetians. These Catholic West Europeans were commonly called Latins (sometimes Franks). Certain high-ranking Byzantines were nevertheless able to establish their own states and claim to be successors of the old Empire. The most important of these “successor states” were the Empire of Nicaea and the Despotate of Epiros.5 In addition, Turks and Bulgarians also carved out new territories for themselves. The Turks were originally steppe nomads who had migrated to plateaus of central and eastern Anatolia in the late eleventh century. They were Muslims, and ruled by the Selçuk-dynasty. Although the Empire of Nicaea conquered Constantinople in 1261 and re-established the Byzantine Empire, the overall situation did not fundamentally change. The Venetians and most of the vassal states of the old Latin Empire in Constantinople were able to defend themselves. The most important of these was the Principality of Achaia, which ruled the Peloponnese and had gained the overlordship of the remaining vassals of the Latin Empire. In addition, new Latin groups arrived, especially during the early fourteenth century. The most important of these were the Knights Hospitallers and the Catalan Company. The latter was a mercenary group that had been hired by the Byzantines, mutinied, and after numerous events established its own state. Moreover, Serbia conquered territories in Romania and Albanians spread into new regions. Already in the latter half of the thirteenth-century, the turmoil created by Mongols pushed groups of Turks into West Anatolia, where they founded numerous emirates.6 During the fourteenth century, the one ruled by the Ottoman dynasty subdued most of the area in question in this study. However, the Central Asian Timur Lenk (1370–1405) defeated the Ottomans at the battle of Ankara in 1402. This battle, and the ensuing civil wars, resulted in a temporary loss of hegemony for the Ottomans. It was not until the time of Murad II (1421–1451) that the Ottomans could reThe terms “Empire of Nicaea” and “Despotate of Epiros” are anachronistic. They are, however, widely used in research, and the alternatives are not unproblematic. Thus I have decided to employ them. “State” itself is also a difficult concept, particularly in a medieval context. I find the best definition of state in this era to be “a territorially coherent political formation dominated by a centralised court and associated administrative apparatus” (Haldon 2008: 157–159; see also Hohrath 2006: 250–253; Wickham 2005: 57; Paul 2003: 26–28; Tammisto, Wilenius 2021: 11–12). 6 The original Turkish term is beylik (plural beylikler). The term emirate is however widely used in the research literature. Besides, the Greek sources usually use the term emir (if not some more archaic title), which in my opinion indicates that it was in common use in Romania. 5 10 establish their dominant position in the region. During the fifteenth century they annexed practically the whole study area with the help of gunpowder weapons, and were expanding beyond its limits. To maintain the coherence of the study, I end the detailed examination at the time of the accession of Murad II, when the influence of gunpowder weapons begins to become marked. Romania during the period under discussion is a very suitable area to test scientific and culturalist understandings of military history. This geographical region had long belonged to a single state, so the original settings for the socio-economic structures were more or less held in common, although there were local variations. In the thirteenth century, two major groups of conquerors arrived in the area. The Latins were representatives of the “western” military traditions of the time, and the military traditions of the Turks originated from the steppes and from the east. Both aimed to settle permanently in the region and founded realms. Therefore, they had to adjust themselves to the local environment, and similarly rulers with a local background also had to adapt to the new situation. The backgrounds of the military traditions of the people who acted in the study area corresponded neatly with their religious divisions. The Byzantines and the other indigenous peoples were Orthodox Christians, the Latins Catholics, and the Turks Muslims. The nomads of the northern side of the Black Sea who raided the Balkans tended to be Muslims or pagans. Religion offered a simple way for people to group individuals as “us” and “them”, and thus helps to establish the classifications used in this study. Religion should also have had, at least in theory, some influence on what was considered the proper way of conduct. The religions could also be seen as cultural commonwealths. It is also important that we have written sources from all three major cultural groups, although not in such a balanced way as one would hope. The socio-political organizations of the belligerent political entities in Romania varied from warrior bands to small empires, even inside the same cultural commonwealths. This helps to avoid the pitfall of using a cultural explanation in a situation where referencing a socio-political organization would be more fitting (or the other way around). If geography, environment, and reciprocity were the main factors shaping warfare, as it is commonly assumed in the scientific understanding of military history, all of the belligerents in the study area should have ended up acting in similar ways when the resources and political situation allowed it; otherwise, the culturalist explanation must be taken under serious consideration. There were no wide gaps in the technological potential between the belligerents in the study area during the period under discussion, and before the advent of effective gunpowder weapons the technology of warfare was relatively stable. One might have had better possibilities to 11 pursue technological innovations and exploit them than another, but there were no insurmountable reasons why these could not have been copied, if there was time, resources, and the will to do so. A comparison of the belligerents not only illustrates the relative importance of different factors shaping warfare but, in addition, it helps to achieve a better understanding of the different dimensions of warfare associated with each belligerent. The comparison makes it possible to evaluate the advantages and deficiencies with better arguments than were previously used, and to estimate to which extent old views about warfare hold. Scholars dealing with the military history of the study area have usually concentrated on one power only (e.g. to the Byzantines or Ottomans), and thus traits which could be revealed from a closer look at the others might have been left unnoticed. Moreover, scholars often evaluate things that are relative, such as the performance of the armies. This can result in a yes–no debate. A good example is the disagreement over the question of whether the medieval western and Byzantine ways of warfare were battle-avoiding. Comparison is arguably the only way to arrive at sound answers for this. Warfare was also at the centre of the ascendance of the Ottoman Empire, a pivotal event in the history of Europe and the Near East, and my study helps in understanding why the Ottomans could rise over their opponents. A comparison of the belligerents in Romania and the factors shaping warfare and the military cultures there allow an investigation of a myriad different issues. To keep the study within acceptable limits, I concentrate on the two subjects that have the most relevance from the point of view of the “military culture” discussion and the history Romania, that create a coherent whole, that are reasonable to investigate considering the chronological and geographical limits and the available source material, and that the study of which provides new meaningful information and interpretations.7 The first main issue is the composition and reliability of the armed forces. Who fought for whom, and why? With “why” I mean the general and often obligating reasons for performing armed service, such as fulfilling the conditions for fief-holding, for payment, belonging to some social or ethnic group with its entailed obligations, or for a share of booty, anticipated spiritual benefits, etc.8 In other words, what was intended to be the “glue” between the armed “employed” I originally planned to deal with more themes, but the lack of time and the amount of work led to cutting them out one by one. For example, the treatment of prisoners and non-combatants would be an important subject, but I find that the current research situation on this topic is reasonably satisfactory. Fortification building, on the other hand, is an issue where a lot of investigations have been done, but way more should be done, particularly in form of excavations, as the construction history of several sites is uncertain and old assumptions are been challenged (see chapter 1.2.2. and 1.4.). 8 Land and naval forces are too often separated in military history, which has resulted in neglect of the later (Black 2004: 22–23; Moyar 2007: 239). However, in this study the military function of le gens de mer is treated in the same way as those who fought on land (about the term gens de mer Rüdiger 2017: 36–37; Stöckly 2017: 168–169). 7 12 and “employers”, 9 or if it is even meaningful to speak about such relationships in the different forces.10 In this part I also focus on the question of the reliability of the troops. Did the “glue” hold? In my opinion, the main determinant of the effectiveness of armed forces is their ability to carry out the chosen strategy.11 Nevertheless, every strategy could lose its value, and even the strongest armies could be rendered useless, or even be mortally dangerous for those for whom they were supposedly fighting, if the troops refused to obey orders or turned their weapons against their employers. Besides, there have been contrary opinions among scholars about the significance of the reliability of certain troop categories for the fate of the policies of Romania.12 I give special attention to an idea put forward by Stephen Morillo. According to him, due to its “militarized legal framework” the military potential of the whole society was put to more effective use in Latin Europe than, for example, in the Muslim world, where was a wider gulf between the structures of the state and the rest of society, and the legitimacy of the former could more readily be put in question. Morillo uses as an example the Iberian Peninsula during the Reconquista, where the Muslim rulers were as result inclined to rely on “outsiders”, i.e. on groups of soldiers not embedded in the local society and often hailing from distant regions. The Christian rulers could, according to him, better benefit from different institutions with military potential inside their own realms, such as autonomous towns with militias.13 But how do these ideas work in the case of Romania? The second main issue concerns whether pitched battles were sought or avoided the campaigns of the belligerents in the study area, and a discussion of the plausible reasons and results of their chosen methods of operation. Whether pitched, decisive battles were the focus of warfare has arguably been the most important topic in the “military culture”-debate, and an issue that has I mean employers and employed in the widest sense, and not only those whose relationship is based on paying wages. The division of the participants in these conflicts falls into those on whose behalf the wars are waged (i.e. above all the rulers and political elites), and those to whom the actual warfare was delegated (see e.g. Hothrath 2006: 250; Brauer, Tuyll 2008: 33–39; 83–85). 10 I leave outside my systematic discussion the moral boosting of already recruited troops and possible personal motives for individuals to join the armed forces (i.e., did they have, for example, money or adventure in mind). Neither do I deal much with the psychology and motivations of combatants immediately before and during the actual engagement. These would require dealing with the sub-conscious feelings and motivations of individuals and would be more or less impossible to detect from the available source material without heavy reliance on hypotheses and anachronism (cf. Ntafoulis 2016). 11 I return to this ability to carry out the chosen strategy in Part 3 and in the Conclusions. In Part 3 I also deal with the cases in which towns and fortifications, apparently defended mainly by the local inhabitants have surrendered. These have more relevance from the point of view of performing military operations than the composition of armed forces. 12 One can cite, for example, the discussion concerning the role of mercenaries and holders of pronoia-fiefs for the decline of Byzantium (see e.g. Vryonis 1971: 468–469; Kyriakidis 2011: 101–104; 222–223; Bartusis 2012: 430–431; Vryonis 1975: 128; Kanellopoulos 2011: 338–339; Korobeinikov 2014: 270–271). 13 Morillo 2006c: 15–17. See also Morillo 2008: 250–253. 9 13 been considered to differentiate “western” and “eastern” ways of war. I employ Andrew Villalon’s concept of “battle-willing”, which he defines to: encompass medieval commanders whose strategic plans, while not irrevocably committed to battle, clearly envisaged that possibility. If a battle offered itself under circumstances they could accept, they would engage the enemy. Realizing (perhaps even hoping) that their goals could be achieved short of the ultima ratio, they would nevertheless be ready to accept the wager. While commanders with such a strategic outlook cannot be said to have been “battle-seeking” in the strictest sense of the term, they were most certainly not “battle-avoiding”. This helps to circumvent the question of whether the intention of the commander was to look for an encounter (for example, if there was a battle was it something one or both sides had sought). Due to source problems, the true intentions of the commanders are almost always impossible to detect.14 I consider an army15 as battle-willing if it, for example, invaded enemy-controlled areas when it was possible to encounter an enemy army, and if the enemy army was met, a “battlewilling” army should have also been ready to engage, and not only to negotiate, retreat, etc. In defence, if an enemy army was threatening some area, perhaps taking fortifications, a “battlewilling” army should have been willing to prevent that by meeting the invader in battle. Other possibilities could have relied on indirect methods, for example by shadowing the invading army, harassing it, and denying supply, or attacking enemy lands to divert attention, etc. I also pay attention to whether the battles were pitched ones, or could possibly be better characterized as surprise attacks or ambushes.16 The article where Villalon introduced the battle-willing concept was partly an answer to John Gillingham’s challenge in the so-called Vegetian warfare debate concerning West European medieval warfare. Gillingham claimed that the only way to provide an answer about the preference of medieval commanders to seek or avoid battles is either to investigate the careers of individual generals or to study what events the individual campaigns contained: battles, sieges, pillaging, and Villalon 2010: 149–152. The citation is from page 150. In this issue, it is above all a question about the preferences of the commander(s). The attitude of troops in this respect mainly mattered if they displayed disobedience. 16 How to define battle is a complicated issue, particularly as we seldom have reliable figures of the men involved in encounters. Thus the size of the armies per se cannot be the criterion. I consider as battle an engagement where the majority of troops in the area are present (so it is not skirmish between smaller groups) and it takes place outside permanent fortifications (i.e. it is not a siege). 14 15 14 so on.17 Villalon concentrated more on the former, and I on the latter. I also pay special attention to methods of taking fortifications, not so much to the details but on a general level, namely were such operations carried out: 1) aggressively, using large amounts of siege technology and possibly making a storming attempt, 2) by blockading the defenders until they ran out of food, or 3) relying on the unwillingness of the defenders to fight against the attacker, or stealth attacks, stratagems, traitors, or similar means? The first method required resources, the second time, and the third skill, and above all the right circumstances and luck.18 How, and particularly how quickly, fortifications could be taken may have had an impact on how campaigns were waged and whether the belligerents were willing to engage in battle.19 I also pay attention to assassinations as an “unconventional method of war”. I divide the main chapter dealing with the composition and reliability of the armed forces according to the cultural background of the political entities who employed or backed them up. The second main part is chronologically ordered, as campaigns are always interactions of at least two players and take place in a distinctive political situation. I deal with the military campaigns and armies in Romania as well as the campaigns that the political entities whose core areas were in the region waged outside the area as autonomous belligerents.20 The fairly long period of over 200 years under discussion is necessary to produce a sufficient number of relatively solid cases, and it helps in eliminating the possibility that exceptional situations, misleading sources, etc. could influence the results too much. In addition, the invaders needed time to settle in, accustom themselves to their new conditions, and in general many associated developments could have taken much time. I find that a major flaw in many previous studies dealing with similar themes is that they have concentrated only on one war or battle. The chosen case might not be so typical for the warfare of the time, or its particular conditions might distort the overall image.21 Besides, especially with ancient and medieval history, the potential Gillingham 2004: 150–151; Villalon 2010: 134–143. About the “Vegetian warfare” debate, see chapter 1.3. I have developed this typology of the ways to take fortifications mainly on the basis of Bradbury 1994: 79–84; Bachrach 1994: 125; Marshall 1992: 210–255. Marshall has, for example, noted in his study about warfare in the Holy Land in the thirteenth century that if the attacker used siege engines for more than harassment they usually employed several types and methods, and in a few weeks the fortress was either taken or the attempts at a quick conquest were abandoned. The Mamlūks usually used aggressive methods, while the sieges of the Christians of the Holy Land tended to be long hunger sieges. 19 See also Rogers 2002: 10–15; Christensen 1999: 45–47; Chase 2003: 61–67; Fighting techniques: 122–123. 20 Thus I, for example, include into the study the wars between Bulgaria and Byzantium, but not the wars between Bulgaria and Hungary. From the wars of the Selçuks whose main centres were outside the study area, I include those which were fought in Romania, but I encompass all the wars of the Ottomans whose core area was in the study area. 21 An excellent example is Carnage and Culture by Hanson, which deals with the history of western warfare through case studies of suitable battles. On the other hand, the starting point of David Graff’s Eurasian way of war was the notion of similarities between Byzantine and Chinese tactics at the turn of the sixth and seventh centuries (Graff 2016: ix–x). 17 18 15 unreliability of the sources always causes problems. Within this time limit I can still base my study on the meticulous examination of primary sources, and thus avoid an overreliance on secondary literature, or the possibility that the subject would grow beyond control. The Fourth Crusade is a self-evident point to begin with, because it was at that time that Romania emerged as a geopolitical region with numerous independent or autonomous political entities of diverse cultural backgrounds. The decision to close the detailed study with the beginning of the reign of Ottoman Sultan Murad II, and particularly with the siege of Constantinople in 1422, is more complicated. Nevertheless, from this point onward dealing with the changes that the effective gunpowder weapons brought to warfare in the fifteenth century would expand the study beyond its practical limits.22 Moreover, as the Ottoman conquests progressed, Romania as a geopolitical region started to lose it meaning, and this study would become more and more a military history of the Ottomans and the people who tried to stand against them. In many ways, choosing the end point of the study after the first two decades of the fifteenth century is a compromise. There is no precise moment when gunpowder weapons become so important that they changed warfare; it was a long process. The first undeniable pieces of evidence about their use in the study area come from the last quarter of the fourteenth century. It is nevertheless more difficult to say when their presence began to make a significant difference. The references to gunpowder weapons and their importance constantly increased, but even at the conquest of Constantinople in 1453 the role of canons may have been smaller than is traditionally assumed.23 When it comes to the expansion of the Ottomans in the closing years of the fourteenth century, they had taken Bulgaria, most of Serbia, and most of Anatolia, but they never took all of the Ionian Islands. Following their defeat at the battle of Ankara in 1402, the Ottomans suffered It could be said that my timeframe falls between two disruptive technologies, i.e. innovations, which changed the whole field and could provide a decisive advantage (about disruptive technologies, see Malkki 2007: 46; Pierce 2004: 1; 202–203). Apart from closing with the breakthrough of gunpowder weaponry, it can be noted that the Fourth Crusade was arguably enabled by advantages in long-distance horse transportation by sea (Pryor 2020: 678–679; and chapter 1.5.2 in the present work). 23 I deal with these early instances of using fire-arms in Romania in chapters 3.2.3–4. It is possibly that the increasing use and effectiveness of gunpowder weapons c. 1400 onwards was particularly due to the introduction of crumbled gunpowder, which was more powerful and above all remained in good condition for a longer time, as well as during transportation. The earliest forms of gunpowder had serious problems with these factors. Another significant innovation was the method for the extraction of saltpetre, which was vital for the cheap and large-scale production of gunpowder in Europe. Before the last quarter of the fourteenth century, the size of guns may also have been too small to smash projectiles through defensive walls. Plausibly the first time that gun powder artillery decisively breached defensive walls was at Saint-Sauver-le-Vicomte in France in 1375 (McLachlan 2010: 10–12; 18–27; Purton 2010: 174; 214; 276– 278; 401; Keen 1999a: 274–278; May 2007: 103–104; Chase 2003: 60–61; Guilmartin 2007: 654–663; Andrade 2015: 252–255; 264; 274; Murphey 1999: 107–108). Unfortunately, we lack detailed information about the gunpowder variants used in Romania during the period under discussion. 22 16 from civil wars and lost some of these territories. A period of re-conquest began after Murad II had established his regime, and thus it is a convenient time to close the study. I want to emphasize that my results apply only to the studied area and scope of time in the subjects I am dealing with. Further studies are required to judge whether it is a typical case about what kind of issues shape warfare in general. Moreover, it is possible that, for example, the warfare of the Latins in Romania was dissimilar to warfare in western Europe. There are never certain answers in history, and it can be said that the true value of writing history lies in the ongoing discussion.24 I see my work more as a basis for further studies than as a definitive and final statement about the questions dealt with. The next study could, for example, compare the reasons for waging wars and their (grand) strategies. 1.2. Primary sources and methodology 1.2.1. Methodology and the use of historical narratives as sources The primary method of my study is to compare the belligerents in Romania from the Fourth Crusade to the early fifteenth century. In other words, I attempt through comparison to find out where the “fault lines” in conduct lay in the studied issues. Were they situated between some kind of military cultures, and if so, which cultures were they? Or, were the issues emphasized by the scientific understanding of military history were the only things that were ultimately significant? I assume that if the number of cases is large enough, patterns will emerge. Even if fault lines are not found, it is an acceptable and valuable result to show the problems in current explanatory models about the factors shaping warfare. Michael Howard claimed that there are three rules for an ideal study about military history: 1) it should be “wide”, having a wide chronological and geographical scope, in order to reveal what truly are the unchangeable principles of war; 2) it should be “deep”, using several different kinds of sources, not only the easiest and most obvious ones, and should scrutinize them; 3) it has to take the context into account, i.e. where war is fought, and the society and culture of the belligerents in general. In a way, the “wide” and “deep” ideals are incompatible. In practical terms, the “wide” ideal stands for comparison.25 As a method, historical comparison, i.e. a systematic discussion concerning two or more historical phenomena with respect to their similarities and differences in pursuit of certain intellectual aims, offers several benefits: it helps to identify questions and problems which otherwise would probably remain unnoticed; it serves to clarify the 24 25 Creveld 1983: 564. Howard 1984: 195–197. See also Travers 1992: 38; Creveld 1983: 563–564. 17