The Queen’s University of Belfast (2003)
After World War I, Calvin Coolidge, as part of his acceptance speech for the Republican vice-presidential nomination on July 27th 1920, cautioned the country against abandoning its social contract with its warriors, ‘The nation which forgets its defenders will itself be forgotten.’ This oft-repeated quote may well have a resonance for the Byzantine emperor Maurice, an emperor who came to prominence through his military success in the late antique period, but in a different context.
In the Strategikon attributed to Maurice, in the section dealing with the Sassanian Persians, it intriguingly states that, ‘They are formidable when laying siege, but even more formidable when besieged’, and it is the underlying reasons for the latter part of this statement that this paper aims to explain. This definitive statement is in many respects a surprising admission for a Byzantine general to make. The Byzantines were well aware and respectful of the fighting qualities of other nations and military manuals, particularly Maurice’s, stressed the need to always be prepared for the unexpected and never to underestimate the enemy, though practice was often different from theory in this regard. While I had investigated this particular view of Maurice’s in my PhD I felt it needed further study and was worthy of at least a short note. The whole issue of siege warfare in late antiquity has been largely ignored and most articles that do touch on it refer either to the late Roman/early Byzantine forces or the incompetence of the western ‘barbarians’. Few have even mentioned the Persians and their poliorcetic abilities, although they were the single most important enemy the empire faced in the late antique period. I concluded in my PhD that the Sassanians were in fact the most competent military force of the period in terms of sieges. With regards to tonight’s topic I intend to briefly explain the background to Maurice’s Strategikon and to analyse the key factors which may have led Maurice to issue this stark warning to his subordinates attacking Persian strongholds.
According to George Dennis, in his background discussion of the Strategikon, the work was probably composed around or shortly AD592 and this is based on arguments clearly discussed by John Witta in his thesis. The surviving versions of the text, which was influential on military thought up to at least the 17th century, can be traced back to three copies which appear to have been produced at or shortly after the time of the original. Only one of the surviving manuscripts, admittedly the principal one, links it to anyone other than Maurice. It ascribes it to Urbikios a 5th century tactician. However, this could be a mistake made when copying the manuscript as Marikios in the Greek could accidentally have been transcribed as Urbikios. The other works all state that it was a product of the general Maurice who went on to become the emperor Maurice. While he had no prior military experience that we know of before his appointment as Magister Utriusque Militum per Orientum under Tiberius in 577, which gave him command of the eastern frontier, he was evidently felt to be competent enough to handle the renewed war with Persia. We know from Menander Protector that in early summer 578 he gave his army training in the art of fortifying camps, and diagrams concerning this appear in the Strategikon. He tried to reorganise the army and this book apparently reflects the systems in use during or after his reign. Menander tells us that Maurice enjoyed poetry and history and he may well have been inclined to compose this handbook. However, John Witta’s thesis lists the various candidates proposed in recent scholarship for the authorship of the Strategikon, and has put forward Maurice’s brother-in-law, Philippicus, an experienced general, as the most likely author. He had an interest in Hannibal and several anecdotal references to the Carthaginian appear in this taktika. The internal clues to the text’s dating also suggest a date between 585 and 628, including apparent references to the siege of Akbas in 583 and the Avar siege of Heracleia in 592 which fit with information supplied by Theophylact Simocatta. This information implies composition during or shortly after the reign of the emperor Maurice. The years that Philippicus spent in a monastery at Chyrsopolis after Maurice’s deposition by Phocas in 603 would have given him the time to write such a work and would explain the first hand, practical feel that comes through in the text. So, while Maurice’s authorship cannot be proven categorically, it is very likely that either he or one of his generals was indeed the author, or that the book was written at his instigation.
The original version of the Strategikon was organised in eleven books, the author clearly ending his work in the last paragraph of Book 11. Book 12, is a collection of five different topics, including a treatise on the infantry. Although it was a later addition, stylistically and in terms of content it is reasonable to assume Book 12 was composed by the same author. The books are as follows:
BOOK 1 Introduction
BOOK 2 The Cavalry battle Formation
BOOK 3 Formations of the Tagma
BOOK 4 Ambushes
BOOK 5 On Baggage Trains
BOOK 6 Various Tactics and Drills
BOOK 7 Strategy: The various points a general must consider
BOOK 8 Untitled – general instructions and maxims
BOOK 9 Surprise Attacks
BOOK 10 Sieges
BOOK 11 Characteristics and tactics of various people
BOOK 12 Untitled – mixed formations, infantry, camps and hunting
Unlike other taktika, the Strategikon of Maurice does not appear to copy previous authors or have antiquated references. Material is organised, clear, comprehensive and in some cases illustrated with diagrams. It is written in Greek, by then the common language of the empire even at official level, although terms are clearly based on the original Latin military vocabulary. It is a practical handbook for the officer in the field, based on recent events and examples, written by somebody who has experience in these matters. Maurice campaigned on the eastern frontier from 577-582 and once elevated to the throne, continued to direct operations against the Persians, Slavs, Avars, Lombards and others. Therefore, the chapter on the characteristics and tactics of various peoples is something he was in a good position to write himself and he had a pressing reason to have it written.
The quotation about the Persians being so formidable on defence is in the opening section of Book 11 where the author details his view of the general Persian military characteristics. The Sassanians are the first nation to be mentioned, but possibly surprisingly, of the four groups of people dealt with, only the section entitled, ‘Dealing with the Light-Haired Peoples, Such as the Franks, Lombards, and Others Like Them’, is shorter:
The Persian nation is wicked, dissembling, and servile, but at the same time patriotic and obedient. The Persians obey their rulers out of fear, and the result is that they are steadfast in enduring hard work and warfare on behalf of their fatherland. For the most part they prefer to achieve their results by planning and generalship; they stress an orderly approach rather than a brave and impulsive one. Since they have been brought up in a hot climate, they easily bear the hardships of heat, thirst and lack of food. They are formidable when laying siege, but even more formidable when besieged. They are extremely skilful in concealing their injuries and coping bravely with adverse circumstances, even turning them to their own advantage. They are intractable in negotiations. They will not initiate any proposal, even one they regard as vitally important for themselves, but will wait until the proposal is made by their opponents.
So just why were Persians defending fortified objectives such a ferocious opponent for the Byzantines? No other late antique source states the issue quite so bluntly. Had the balance of power shifted so radically in the Persian’s favour by Maurice’s time compared to earlier periods? A quick review of military relations between the two powers is called for to set the context of Maurice’s Strategikon more clearly. The Sassanian Persians and Byzantines had been both uneasy neighbours and relative friends over the three centuries preceding Maurice. There had been on-off warfare from the accession of the Sassanians in the 230s up to Julian’s ill-fated expedition to the heart of Mesopotamia in 363. Control of the city of Nisibis had been central to the fighting and even though the Byzantines had never lost it in combat its transfer to the Sassanians was the price of safe passage from Persian territory in 363. The partition of Armenia in 387 coupled with the fact that both sides had their own internal and external pressures to deal with led to peace over the next century and even a good deal of co-operation and mutual trust. The later 5th century saw repeated attempts by the Sassanians, partly due to internal politics, to demonstrate an upper hand in any negotiation. This was particularly evident with regard to demanding Byzantine payments to Persia for the defence of the Caucasian passes. Hostilities broke out again in the reign of Anastasius. This period of conflict from 502-506 resulted in the Sassanians capture of Amida but it had to be returned as part of the ceasefire negotiations. What could not be restored was the trust and co-operation of a century earlier and Justinian’s campaigns from 527-532 have been seen as a war of revenge concluded by the Endless Peace once the Persians had the upper hand. By 540 the war was being renewed and another full scale Persian invasion again caught the Byzantines unprepared. Despite several truces, the conflict carried on continuously fought either by the two main combatants, or through their Arab and more commonly, Armenian and Laz allies, right up to 561. As the northern frontier reaching to the Black Sea became the main theatre of operations, Justin II allied with the Turks which sparked further fighting which lasted almost continually until 630.
If Maurice was indeed the author of the Strategikon attributed to him it may be useful to note that he was born in Cappadocia in 539. Cappadocia’s main town Caesarea was a key strategic centre of operations in these northern conflicts. It is against this background of renewed and intensified fighting, much of it revolving around sieges both of Byzantine and allied towns on the one side and Persian forts and recently urbanised sites on the other, that we must focus in assessing the reasons for Maurice’s judgment on Persian poliorcetic ability.
The Sassanians were well able to build, defend and attack fortifications. Siege warfare had first arisen in the regions their territory covered, so it could be argued that they had a greater tradition of poliorceticism than the heirs of the Roman empire. They were a match for the Byzantine forces in military operations, as the former’s generals recognised, and this was particularly true in sieges. It is also important to observe that this form of warfare accounted for over half of the military engagements between the two powers, since the areas they were fighting over contained many towns and additional fortifications to protect them. Not only were the Sassanians a worthy opponent equally skilled in not just in all forms of combat, but they had an equivalent administrative and logistic capacity. They were the only power of comparable size to the early Byzantine army, and they had an effective military system. ‘Indeed one might be led with good reason to marvel at the assiduity and resourcefulness of the Persians in the prosecution of their wars;…,’ remarked Procopius. Any references to the Persians as ‘barbarians’ should not be taken as a slur on their martial abilities.
The Sassanians had fine troops of all types ranging from heavily armoured cavalry to light armed mountain troops. They were able to call on a whole host of subordinate peoples, some of whom probably possessed specialist knowledge in certain types of warfare, just as the Byzantines employed federates with particular skills as the need arose. Although evidence is sparse, it is generally argued, most recently by Zev Rubin, that it was not until Chosroes II’s reign, 531-579, that the Sassanians attained a standing, professional army. The adoption of such a force at this time, coupled with increasing urbanisation and a period of prolonged warfare, surely galvanised the Sassanians’ military capability to an even higher level. What is particularly significant for this paper is that this is precisely the period that our author would have been most familiar with.
As has been stated earlier, the fourth century had witnessed many sieges as each side struggled to deal with the aftermath of the treaty of 299. By contrast, the fifth was relatively quiet as both tended to observe the treaties of 363 and 387. The sixth century started violently and the trend continued until the mid seventh when the Arabs conquered the Sassanians and forced the Byzantines onto their back foot. It should be remembered that the Persians frequently had to fight on their northern frontier, as well as their western one. The nomadic tribes of the Eurasian Steppe often tried to break through the Caucasus, and there were many Persian requests and demands for the Byzantines to share this burden. However, it is Persia’s western frontier that concerns us most, for here they faced the Byzantines for 300 years. In the turbulent periods it was the Persians who took the initiative on most occasions as far as siege warfare was concerned. A database compiled from extant sources shows that there are only 35 recorded examples of them defending fortifications from Byzantine assault, compared to 107 where they tried to capture fortified sites from the Byzantines. There are only three other non-Byzantine examples: Nisibis in 499 where they successfully held off the Kadishaye rebels, Nisibis in 602, where rebel Sassanians appear to have resisted loyalist forces resulting in a sack of the city, and Tiflis, where Turkish nomads unsuccessfully blockaded the city for a while in late 626. Unfortunately, however, we have few very detailed accounts that relate how the Persians conducted their defence, and there are always difficulties in how far statistics can give a true reflection of what happened. Nonetheless, on the surface these figures do tend to be in accordance with the Strategikon’s reckoning. The Sassanians were apparently capable of avoiding sieges better than the Byzantines. This of course opens up all sorts of arguments over strategic planning and whether or not the Sassanians preferred to have open battle rather than retreat to fortifications. More probably, it is reflective of the fact that the Persians took the offensive more often than the Byzantines, usually in the form of raids rather than invasions with the aim of permanent capture, and were therefore moving in to more heavily urbanised zones than if the Byzantines were the invaders. What is perhaps more interesting is that in these 35 sieges, the Persians successfully held out in only 14 (40%). This compares with a figure of 57% for the Byzantines repelling Persian attacks. Even given that the Strategikon is a late sixth-century work, the figures for that period alone are 9 successful holds out of 18, still only 50%. It must be admitted that this is a small sample number to be drawing overall conclusions from, but it is all that we have to go on, and Sassanian defensive tactics and strategies therefore require closer examination with regard to Maurice’s statement.
How then did the Persians go about protecting their cities? Very little work has yet been done on Persian fortification, certainly by comparison with late Roman/Byzantine archaeological studies. It is clear that garrison commanders were supposed to have laid up supplies in advance of an expected siege, or in areas on the eastern frontier particularly afflicted by warfare, cities and forts should have had a permanent stockpile. For example. the Byzantines besieging Melitene in 504 intercepted a Persian supply train heading for the city. The Persians at Sisaurnon in 541 did not put away the annual food supply in public store-houses, ‘as they did at Daras and Nisibis’, and soon ran out of provisions, records Procopius. This was exacerbated by the fact that more people were now inside the city because of the attack. When the Byzantines found this out, they switched from assaulting the city to blockading it, and it capitulated as a result. Equipment reserves were also necessary. Petra in Lazica had an elite garrison and was well stockpiled with arms and armour as well as food, five years worth in fact. Thus the Persians, like the Byzantines, believed in being prepared. If a siege was imminent then additional measures could be taken. The Persian marzban in Nisibis gathered supplies, felled trees outside the walls, expelled the Christian population and made other preparations.
When actually faced with an assault by imperial forces the Persians knew the full range of counter measures available to a besieged garrison and used them all when appropriate. For example, when Constantius besieged Bezabde in 360, the Persians demonstrated their knowledge of close defence tactics. They dropped heavy objects including millstones, large jars, and bits of broken columns on men and machines near the base of the wall. They set up hairmats on the battlements to provide protection from missiles, even though this blocked their view of proceedings. When attempts were made to scale the wall at these points, the defenders thrust their weapons blindly through the mats to attack the assailants. When the battering rams approached the walls, lassoes were dropped over the sides to entangle them and haul them up. Sallies were made to try and drive the attackers away from the wall, or kill the engine crews and burn the machines. Initially these attempts were resisted, but a later sally succeeded in burning some engines. Another assault on the walls was repulsed by a sally and siege mounds that had been erected were fired. Thus a determined, almost textbook defence was conducted in this instance, but one factor was conspicuous by its absence, as it is in many other accounts, namely mural artillery.
Despite their undoubted knowledge of such weapons, there are very few examples of Persian defensive artillery being mentioned by our sources. Ballistae are recorded at Pirisabora and Maozomalcha in 363, at Amida in 504 and Beiudaes in 587 and that is the sum total of references to Persian mechanised defensive firepower. This is a paltry number compared to the Byzantine figures, which themselves are hardly extensive. The use of impact reducing devices such as sacks filled with chaff, or screens are well attested, but these are passive devices. For active defence the Persians appear to have relied on archers and slingers. Assuming that the sources accurately reflect a Persian ambivalence to artillery in a defensive capacity, we must ask why that should be? They clearly had the technology to know how to build, maintain and use these weapons, and were happy enough to employ them on offence. So why did they not use them on defence?
The supply of raw materials may have limited the manufacture of these machines, and the consequent effect on spare parts may have been the reason. With these restrictions it is not unreasonable to speculate that the Persians were unable to maintain or stockpile the weapons. The available artillery may have been kept in a few key fortresses, rather than being split up around the country, in order to keep it together for campaigns and minimise the risk of damage from additional travelling. This in turn is bound to have had some effect on Persian military doctrine. However, any trade embargoes imposed by the Byzantines are not likely to have been very effective since it was relatively easy to cross the frontier most of the time, and while iron may have been a scarce commodity in Persia, there was relatively little needed in each weapon. Is the absence of artillery just a problem in our sources or was there another reason?
The Persians’ skill as archers is undisputed and a high proportion of their troops were so armed. Consequently large numbers of these men and ammunition were available in time of siege. Archery allows one combatant to hit another at a distance, which is the same principle as artillery. The only differences between the two are range, ammunition and the means of propelling the missile. Artillery could shoot heavier loads, either large arrows or stones, and it could fire them further than a man could. The advantage an archer possesses is mainly one of speed and manoeuvrability. Artillery cannot be moved as quickly and has a slower rate of fire than a trained archer. On the other hand it can achieve better results at a distance and with greater penetrating power. Therefore the artillery should be used to deal with the enemy when he is at a distance, while archers should be used at close range. This is especially true for the besieged. The enemy normally encamped out of missile range so defensive artillery was only of use in ensuring they stayed that far back. If they came any closer than that they would be quite vulnerable to sallies as well. While the besiegers were beyond the limit of missile fire the machines would be idle in any event. Should an assault be mounted, the attackers would be trying to close the distance as soon as possible anyway, in order to reduce the amount of time they were in the open. In this case artillery would only be able to fire a few shots before the attackers reached the walls. Assuming the weapons could be depressed sufficiently to fire along the curtain, they would still be useful, but at that distance archers would be a better defence because of their higher rate of individual fire. More archers could fit into the space in a tower than artillery pieces, so the advantage of a higher rate of fire would be further increased. At such close range, the penetrating power of an arrow would be quite high and there would therefore be no point wasting artillery ammunition if there was sufficient arrow fire available. The main use for artillery should have been against siege machines. These were stronger and better protected than troops and damaging them beyond repair required greater penetrating power than was possible from normal bows. However, once they reached the walls, sallies and dropping objects from the walls seem to have been more common ways of dealing with these engines. Owing to their trajectory, stone throwers like the onager would have had trouble hitting engines against a wall anyhow. The ballistae would still be useful, but these weapons would only penetrate the hides or plates covering the engine or possibly set it on fire in places with fire darts. The crew could try to extinguish these flames and could easily be replaced if wounded. Destruction of the machines was the only effective way to counter them and that really required great weights being dropped on them or large quantities of inflammable substances being poured on and around them. This was achieved most effectively from the battlements or by sallies, rather than by artillery. Therefore, on a cost/benefit analysis, for defensive purposes, archers on the whole would be more valuable than artillery. The Sassanians appear to have realised this, as did the Byzantines, but unlike their Mediterranean counterparts, they seem to have virtually abandoned their defensive artillery and were prepared to deal with the engines at close range, since they seem to have believed that the engines would probably reach the wall anyway. Unfortunately, we have no surviving Sassanian manuals to confirm this, but the lack of advice for defending against mural artillery during an assault in Byzantine taktika and narrative sources would imply that this was indeed Persian policy. Whether it was by choice or default owing to a lack or raw materials is another matter. Apart from the Persian lack of defensive artillery, there was little to choose between the two main settled peoples of the period. Both could be tenacious in defence and use every known tactic when on the attack.
Once a siege had started the Persians usually fought tenaciously. Maurice’s Strategikon says that they were extremely skilful in hiding their injuries and coping bravely with adverse circumstances. The garrison at Petra in 549 did not throw their dead outside the walls, despite the stench, in case it gave away the fact that they had lost a large number of men. At Amida in 504, during one assault, only 9 Persians on the ramparts were seen to fall, compared to Byzantine casualties of 40 dead and 150 wounded, although defenders normally experienced far fewer losses. It is clear that the Persians seem to have been prepared to make a stand behind their walls more often than their imperial counterparts did. Apart from the example of Thilutha in 363, when the garrison pragmatically agreed to remain neutral until the outcome of the war was decided, the Sassanian troops frequently held out as long as they possibly could. We know how 30 of the 35 sieges ended and the Persians fought to the end in 19 of them (63%), whether this was until the city was stormed or until the Byzantines decided to abandon their efforts. Petra in 551 is the classic example. Only 730 of the 1800 Persians in city were left alive when the main area was taken. 500 more were killed in the acropolis, burned alive, despite being given a chance to surrender honourably and knowing they had no chance of escape. Why were the Persians so prepared to hold out? It could have been honour and pride, or patriotism, but more material reasons probably persuaded them as well. We have already seen that their garrisons should have been well supplied, which would naturally give a defending force more confidence and keep their morale up. It was when they were blockaded and had run out of supplies as happened, among other examples, at Amida in 504, an unknown fort in Persarmenia in 531, Sisaurnon in 541 and Akbas in 583, that they surrendered. There were a few examples of submission because of fear as at Pirisabora in 363, where the garrison surrendered just before the city was due to be stormed since they could see the end was in sight, or Edessa in 628 when they were overwhelmed by Byzantine firepower and surrendered, but even in these examples they had resisted for a while and had not just capitulated quietly. However it seems clear that the Persians appear to have been more determined to hold out in sieges, and in clashes between the two forces there are fewer negotiated Sassanian surrenders compared to their western neighbours. It is precisely this willingness to resist, rather than the number of successful defences, that is probably behind the Strategikon’s comment. The author is not thinking about Persian weaponry or numbers, though there is no doubt that they inflicted heavy losses on assailants, particularly with their arrow fire. Rather he is concerned with their stubborn resistance. The key question then is what led to this willingness to resist?
Like the Byzantines, the Persians appear to have studied war very seriously and written manuals on how best to prosecute it. While we have few detailed siege descriptions, certainly not enough to be dogmatic about the tactics of either side, our biggest loss, with respect to the question before us, may be the Persian military manuals which may have contained their ideal response to each situation that arose. The Fihrist of al-Nadim preserves the title of a treatise based on Sassanian military science called “Mobilisation for Wars, the Training of Horsemen, and How the Kings of Persia Administered the Four Frontiers of the East, West, South and North”. Unfortunately there are no extant Sassanian documents telling how they protected their fortified possessions. Their only surviving poliorcetic work is a fragment of a treatise contained in the Âyin-n meh which itself has survived through inclusion in the anthology ‘Uy n- l-akhb r by Ibn-Qutaiba. The Âyin-n meh or ‘Book of Rules’ in Arabic lists all sorts of rules for government but has a short section on military practice. This section contains only a paragraph dealing with sieges, and unfortunately makes no mention of how the Persians should defend their fortifications. It is concerned solely with using underhand and pyschological methods to weaken the resistance of defenders and then inducing them to surrender so that fighting could be avoided, a practice the Byzantines were also avid exponents of. However, battering rams and ballistae are included as part of the Sassanian inventory that should be mentioned by the agent sent to perturb the defenders. Nevertheless, despite the best efforts of these agents, not all objectives surrendered quietly and it was necessary for both armies to resort to direct action.
Without guidance from military handbooks, we must return to the descriptions of various accounts to try and find underlying factors for this determination to hold out. In addition to the general preparedness of the army noted above one other moral boosting factor does recur regularly. The Persians seem to have sent a relieving force to almost every city when it was besieged. In Julian’s campaign they did not manage to force the Byzantines to abandon any sieges, but it is clear that they harassed them. It would even appear that they were able to co-ordinate their efforts with those inside the objective to maximise the effect. Whilst besieging a fort near Ctesiphon in 363, the Byzantines were subjected to one attack during the day which coincided with a sally from the defenders. Then at night another twin pronged assault on the besiegers was launched. While both attacks were repulsed it shows the Persian method for dealing with would be assailants. This response is essentially strategic, but it was the procedure commonly adopted by the Persians. Relief forces were sent to aid the besieged at Nisibis in 421 and 499, Amida in 504, and Petra in 549, among others. Two forces were sent to Martyropolis in 589 to force Philippicus, a potential author of our handbook, to withdraw and they succeeded in fighting their way into the beleaguered town. The continued use of this tactic by the Persians was good military practice. It showed a thorough understanding of the strategic concept of siege warfare and what fortifications were all about. More importantly, not only did they understand it, they were able to put it into practice. They may not have held on to too many forts once a determined siege was begun, but it could be argued, very plausibly, from the figures that the Byzantines were not able to risk sieges of Persian forts as often as they would have liked, because of the threat posed by a relieving force. This belief in the rest of the Persian army and its ability to rescue the situation may well underline Maurice’s statement. Further reinforcement of this view came when I reread sections of the Strategikon in preparation for this paper. Book 10’s opening section starts,
Our camp must be very strongly fortified, and a large number of our sharpest scouts should be stationed around, covering even the most unlikely places, in order to prevent the besieged or forces outside the walls from suddenly attacking us, either by day or night, and exposing the army to danger. This is what happened in Arzanene when some of our commanders were captured while besieging a fortress.
The events at Akbas and Aphumon in 583, the former alluded to in the Strategikon, could well be typical of how the system functioned. The Persians attacked the fort of Aphumon, but arranged that their garrison at nearby Akbas should signal them by fire if they were attacked by the Byzantines. When the latter did so, the Persians from Aphumon quickly responded to the signal and returned, trapping the besiegers and forcing them to flee down a mountainside with heavy casualties. The fact that this opens the section on siege warfare is surely not purely co-incidental. Therefore, paradoxically, in Maurice’s Strategikon, we may actually have preserved the very strategy that is missing from the Persian handbooks. Sassanian defence in sieges was based on a confidence that help would reach them in time if they were prepared, both physically and psychologically, to hold out. The fact that they believed this made it harder for the Byzantines to employ treachery or persuasive arguments successfully and thus avoid an assault or long blockade. Any attempt at beleaguering an objective also had to take into account the high probability of a counter-attack by a relief force. Thus the Persians were truly formidable in defence not just because of the casualties they could inflict and the fact that they would be willing to hold out for a long time, but also because of the difficulties such facts imposed on anyone planning to besiege them. Supplies for a long siege and extra troops for an assault and additional provisions for them were more likely to be required if a successful campaign was to be undertaken by the Byzantines. These additional logistic factors and the probability of higher casualties and more time spent in besieging forts rather than gaining ground made attacking Persia an option whose ramifications had to be more seriously thought through than might have been the case when the Persians were contemplating an offensive. The price of their stubbornness may have spared them from further invasions. Calvin Coolidge’s quote that, ‘the nation which forgets its defenders will itself be forgotten’, is a maxim by which the Persian’s seem to have defended their own empire during war rather than after the conflict had ended.
Maurice, Strategikon, XI, ed. G.T. Dennis and E. Gamillscheg (Vienna, 1981); tr. G.T. Dennis, Maurice’s Strategikon: Handbook of Military Strategy (Philadelphia, 1984).
C.S. Lightfoot, in his works ‘Facts and Fiction – the Third Siege of Nisibis (AD 350)’, Historia, 37 (1988), 105-25 and ‘Sapor before the Walls of Amida’, The Eastern Frontier of the Roman Empire, eds D.H. French and C.S. Lightfoot (BAR.Int.Series, S553, 1989), 285-294, is one of the few who have looked at Sassanian offensive tactics.
J.E. Witta, The Ethnika in Byzantine Military Treatises, (PhD Diss. University of Minnesota, 1977), 16-24.
 Menander Proctector. Fr. 58. ed. and tr. R.C. Blockley, The History of Menander the Guardsman (Liverpool, 1985).
 J.E. Witta, The Ethnika , 24- 41.
 George Dennis’ translation.
 For more detail see G. Greatrex, Rome and Persia at War, 502-532, (ARCA, 37, Leeds, 1998)
For more on Sassanian administration, defensive systems and urbanisation see J.Howard-Johnson, ‘The Great Powers in Late Antiquity: A Comparison’, States, Armies and Resources (The Byzantine and Early Islamic Near East, 3, Princeton, 1995).
However, as Blockley points out, up to the early 6th century, the Persians did have a problem in organising large scale campaigns, and sometimes even maintaining a strong defensive posture, certainly until a large standing army was formed in the sixth century. ‘The effective military manpower (the cavalry) was both expensive to keep up and mostly under the control of the nobility, who supplied it to the king as part of their allegiance. Since the king himself had only a small standing army (his personal forces), for any significant campaign he needed both the nobles’ levies and whatever allied contingents were available. Not only did this make mustering a slow process, especially if the nobility were unenthusiastic, but it also inhibited the development of an infrastructure for supply and maintenance such as the Romans had to service their standing army. The deportation of Roman skilled manpower, which continued when the opportunity offered through the fourth and fifth centuries, and the use of it to people new settlements, often on favourable terms, appear to have been part of the king’s ongoing effort to amass more manpower and skills under his direct control. The repeated attempts to obtain financial resources from the Romans, by looting, extortion or agreement, had the same objective, of increasing the king’s direct control.’ R.C. Blockley, East Roman Foreign Policy, from Diocletian to Anastasius (Leeds, 1992), 105. E.g. The Sassanian led off to Persia those skilled in building after Jerusalem was captured. ‘Antiochus Strategos’ Account of the Sack of Jerusalem in AD.614.’, tr. C.F. Conybeare, EHR, 25 (1910), 507-8.
Procopius, Wars, VIII.13,5. ed. and tr. H.B. Dewing, (Cambridge, Mass., 1914-1928).
Agathias, Myrinaei Historiae Libri 5, II.21,4. ed. R. Keydell (CFHB, 2, 1967); tr. J.D. Frendo (CFHB, 2a, 1975).
Z. Rubin, ‘The Reforms of Khusro Anūshirwān’, in States, Armies and Resources (The Byzantine and Early Islamic Near East, 3, Princeton, 1995).
The Persians were constrained in their dealings with the Byzantines in the fourth and fifth centuries. The only significant military invasions of each others lands were Shapur II’s inroads to north Mesopotamia in 359-60, Julian’s campaign of 363, and the Byzantine initiated hostilities of 421-2. The other hostilities were basically confined to the border regions. The Persian reluctance to penetrate the empire and concentrate instead on cross-border raids and threats, may be explained by an increase in the imperial defences under Diocletian and Constantine and, from the mid fourth century, by the distracting activities of the Kidarites in the north-east. For more detail see Blockley, Foreign Policy, 102-106. See also Z. Rubin, ‘Diplomacy and War in the Relations Between Byzantium and the Sassanids in the Fifth Century AD’, The Defence of the Roman and Byzantine East, eds P. Freeman & D. Kennedy BAR.Int.Series, S297(2), Oxford, 1986), 677-695.
E.g. Priscus tells of a Persian embassy which asked the Byzantines to contribute to the cost of garrisoning the fortress of Iouroeipaach at the Caspian Gates, which was stopping the Huns getting through the pass. They also wanted financial assistance for their war against the Kidarite Huns because it was to the Byzantines advantage to have these people stopped before they got through the Persian empire. The Byzantines refused to contribute. Priscus, f.41,1 (Exc.de.leg.gent, 15), and f.47, ed. and tr. R.C. Blockley, The Fragmentary Classicising Historians of the Later Roman Empire, II (ARCA, 10, Liverpool, 1983), 222-400.
 Josh.Styl., XXII and XXIV, ed. and tr. W. Wright, (Cambridge, 1882). Theoph., Chron, AM.6117 ed. C. de Boor (Leipzig, 1883-1885).
 Chron.Seert, 74-5, ed. and tr. A. Scher, PO, 4;5;7 and 9 (Paris, 1908-1918),and Chron.Edessa, VI-VII, ed. and tr. E.W. Brooks (CSCO, Scr.Syr., III.4, 1903), 1-11
 Moses Dasxurançi., 85-86, A History of the Caucasian Albanians, tr. C.J.F. Dowsett (Oxford, 1961).
 Josh.Styl., LVI.
Procop., Wars, II.19,19-21.
Akbas surrendered in 583 when it ran out of supplies during a blockade. John Eph., Ecclesiastical History, Part 3, VI.36, ed. and tr. E.W. Brooks (CSCO, Scr.Syr., III.3, 1936).
Procop., Wars, VIII.12,17-20.
Chron.1234, 65, ed. and tr. J.B. Chabot (CSCO, Scr.Syr., III.14 (text 1920, tr.1937), 17-226. and Mich.Syr., X.8, (p.307), ed. and tr. J.B. Chabot, Michel le Syrien, Chronique, 3 vols (Paris, 1899-1910). The expulsion of the Christian population may have been to reduce the chances of treachery within the city and the number of mouths that had to be fed, though it could have been anti-Persian bias on the part of the authors.
Rather than repeat what was said for the Byzantines, suffice it to say that the tactics were effectively the same for dealing with machines, tunnels and mounds etc. The siege of Bezabde serves as an adequate example of Persian defensive tactics. Ammianus, XX.11,23.
 Theophylact Simocatta., II.18,7-25. ed. I Bekker (CSHB, 1834); tr. L.M. Whitby, The History of Theophylact Simocatta (Oxford, 1986).
Libanius, Oatio, LIX.66ff. ed. R. Foerster, 4 vols (Leipzig, 1908). Ed. and tr. A.F. Norman (Cambridge, Mass, 1969).
E.A. Thompson, Romans and Barbarians: The Decline of the Western Empire (Wisconsin, 1982), 10-15. However, A.D. Lee says trade across the frontier and the smuggling of illegal material would not have been that difficult, Information and Frontiers: Roman Foreign Relations in Late Antiquity (Cambridge, 1993), 54-66.
The Persians are noted for their high rate of fire, though it was not as powerful as Byzantine archery. Maurice, XI.1,6. Procopius takes this further when he says the Persians, almost all bowmen, shoot more rapidly than the Byzantines, but that their arrows broke upon hitting Byzantine armour and had no power to hurt them, because they were shot from weak bows. Procop., Wars, I.18,32-33.
This suggests that it was normal practice to dispose of corpses over the side in a siege.
Procop., Wars, II.30,16. In fact, 1000 of a garrison of 1500 are said to have died in Petra’s defence.
Joshua the Stylite, LIX-LXXXI. The Chronicle of Joshua the Stylite, ed. and tr. W. Wright, (Cambridge, 1882). It should be made clear that the context of this passage is to show the added protection the Persians had from small ‘houses’ on the wall, but it shows that the kill ratio definitely favoured the Persians, or at least was ‘seen’ to. However, this is hardly surprising as the attackers would have less cover than the defenders and would expect to have higher casualties. Napoleon estimated that a personnel ratio of 4:1 for attackers was needed when fortifications were to be attacked.
Ammianus Marcellinus, Res Gestae, XXIV.2,1, ed. and tr. J.C. Rolfe (Cambridge, Mass., 1935-1939).
Procop., Wars, VIII.11,63.
Malalas, 469. ed. L. Dindorf (CSHB, 1831); tr. E. Jeffreys, M. Jeffreys and R. Scott (Melbourne, 1986).
Procop, Wars, II.19,2-25.
John Eph., HE, VI.36. ed. and tr. E.W. Brooks (CSCO, Scr.Syr., III.3, 1936).
Ammianus, XXIV.2,9-22; Zosimus, Historia Nova, III.17,3-III.18,6, ed. L. Mendelssohn (1887); tr. R.T. Ridley, Zosimus (Australian Association for Byzantine Studies, Byzantina Australiensa, 2, Sydney, 1982), and Lib., Or, XVIII.227-8, ed. R. Foerster, 4 vols (Leipzig, 1908). Ed. and tr. A.F. Norman (Cambridge, Mass, 1969).
See J.E. Witta, The Ethnika, 65, n.2.
For more information see S.E. Inostrantzev, Cfcfyblcrît–’n.ls (St Petersburg, 1909); tr. M.L. Bogdanov, ‘The Sasanian Military Theory’, Journal of the K.R. Cama Oriental Institute, 7 (1926),16 and 33, and also 48-52, regarding ballistae, onagers and siege towers mentioned by J hiz in the Kit b ul-bay n wa-t-taby n.
Socrates, HE, VII.18, ed. R. Hussy (1853).
Procop., Wars, II.30,1-11.
Theoph.Sim., Historiae, III.14-16 and Evagrius, HE., VI.14,23-28, ed. J. Bidez and L. Parmentier (London, 1898).
A peasant army led by the Kadarigan in 586 did not fight the Byzantines, it merely outmanoeuvred them and forced them to abandon their siege of Chlomaron. Theoph.Sim., II.7,6-II.8,12.
‘The efficacy of a fortress is plainly composed of two different elements, the passive and active. By the first it shelters a place, and all that it contains: by the other it possesses a certain influence over the adjacent country, even beyond the range of its guns. This active element consists of the attacks which the garrison may undertake upon every enemy who approaches within a certain distance. The larger the garrison, so much the stronger numerically will be the detachments that may be employed on such expeditions, and the stronger such detachments the wider as a rule will be the range of their operations; from which it follows that the sphere of the active influence of a great fortress is not only greater in intensity but also more extensive than that of a small one. But the active element itself is again, to a certain extent, of two kinds, consisting namely of enterprises of the garrison proper, and of enterprises which other bodies of troops, great and small, not belonging to the garrison but in co-operation with it, may be able to carry out.’ K. von Klausewitz, On War, tr. Col. J.J.Graham (London, 1962), II, 196.
This did not only apply to Persian forts, since the Persians even helped potential allies by sending relief forces. They gave this support to the rebel Armenian Ata Khorkhoruni so the Byzantines backing the loyalist claimant were forced to abandon the siege because of their approach. Sebeos, 11. Similarly the rebel Narses seized control of Edessa and held it. Germanos was sent to retake the city but was forced to abandon his efforts by a Persian relief force. Theophanes, Chronographia, AM.6095-6. ed. C. de Boor (Leipzig, 1883-1885).