In his landmark study of the rapid rise and de-cline of the Huns in the fifth century CE, E.A.

Thompson observed that “history is no longer sat-isfied to ascribe so striking a movement as the riseof the Hun empire to the genius of a single man …it is only in terms of the development of their soci-ety that we can explain … how they came to buildup so vast an empire of their own, and yet provedunable to hold it for more than a few years” (1996:46). By making this claim, Thompson did not in-tend to diminish the role that gifted Hun leadersplayed in guiding their society to internationalprominence—only to point out that monocausalexplanations cannot adequately capture historicalreality in all its completeness. Leadership is obvi-ously important, but even the most talented leaderis limited by his or her circumstances.

Even so, much nomadic scholarship has tended toprivilege charismatic leadership as one of the mostimportant factors, if not the single most importantfactor, that contributed to successful nomadic mil-itary organization (see, for example, Di Cosmo1999: 19-21; Drompp 2005: 108). This tendencycomes to a head in studies of the Mongol expan-sion, where scholars note that Chinggis Khan suc-cessfully set up a ruling system based on loyalty tothe “holy charisma” of the ruling house (Golden2000: 36), redirecting old tribal loyalties from (realor fictive) kinship-based structures to a new andexclusive focus on duty to the Mongol royal house(Morgan 1986: 90). While these points and thescholarship that supports them are certainly valid,there is sometimes a tendency toward too great afocus on the importance of leadership at the ex-pense of other important contributing factors tonomadic military successes.

Of course, leadership is not the only explanationoffered for nomadic military prowess. The mobilityof nomadic troops is also an oft-cited factor usedto explain their military successes (Morgan 1986:86; Thompson 1996: 55), as is the quality and num-

ber of mounts which made such mobility possible(Sinor 1972: 171). Other such factors often includenomadic battle tactics, such as the art of luring en-emies into vulnerable positions before attackingthem (May 2018: 1), along with specific political de-velopments, in both nomadic polities and those oftheir adversaries, that altered the nomadic balanceof power vis-à-vis their opponents. While worthyfoci of scholarly attention, all these factors offeronly a partial explanation of nomadic successes.Scholars should also look for additional factorsthat contributed to nomadic successes and canhelp explain historical realities that are only par-tially explained by appeals to leadership, mobility,politics, and tactics.

For example, it is significant that the Huns even attheir peak under Attila never won a victory againsta full-strength Roman field army, mostly chalkingup victories against disorganized opponents whenthe Roman legions were engaged elsewhere. Everytime the Huns did meet the Roman military foropen battle proper, they either lost miserably orwon Pyrrhic victories—Attila’s bloody victory overthe Byzantine army in 447 CE is a good example(Thompson 1996: 227). The Mongols, on the otherhand, routinely and ably trounced the best soldiersand armies the most powerful sedentary statescould throw at them. I contend that gifted leader-ship or better use of mobile armies in the “exposedzones” where many nomadic victories were wonand where nomadic political and cultural influencewas most heavily felt (Lieberman 2008: 693) arenot enough to explain these differential successes.

Likewise, overly simplistic explanations that attrib-ute nomadic victories to superior numbers of com-batants (Smith 1975: 272) or the incompetence ofthe nomad’s enemies (Smith 1984: 345) are the re-sult of putting too much trust in flawed and frag-mentary primary sources. Close analysis revealsthat many of the innate advantages we assume no-madic societies to have enjoyed over their seden-

One Bow (or Stirrup) Is Not Equal to Another:A Comparative Look at Hun and Mongol Military Technologies

Samuel RumschlagUniversity of Wisconsin-Madison

The Silk Road 16 (2018): 78–90 Copyright © 2018 Samuel RumschlagCopyright © 2018 The Silkroad Foundation78

tary foes are in fact illusory. May (2006) has notedthat although scholarship has tended to character-ize nomadic armies as mainly achieving victory by“overwhelm[ing] their opponents through sheerferocity or superior numbers” (517) or has simplybrushed off their prowess by claiming that nomadswere “natural warriors inured since birth to ridingand archery in the harsh climate of the steppe”(517), nomadic armies were in fact often quitesmall when compared to those of their opponents(623) and required every bit as much training tobecome battle-ready as the professional soldiersthey fought. Indeed, the martial lifestyle came nomore “naturally” to them than to anyone else. In-deed, to many observers in the ancient world itmust have seemed that the armies of stable seden-tary states enjoyed innumerable advantages overtheir mobile counterparts: funding, equipment,supplies, professional leadership—the list goes on.

Why then do we see powerful armies in the serviceof sedentary states so often trounced by nomadicfoes? Were nomadic victories really as “inevitable”as some incautious authors have claimed (Bartlett2010), or is some other overlooked factor at play?To help explain nomadic successes, I will highlightone aspect of nomadic society that is not fre-quently discussed. I argue that superior militarytechnology was as crucial to nomadic military vic-tories as were other factors such as gifted leader-ship and extreme mobility. Improvements made to

nomadic military technologies over time allowedsuccessive nomadic groups to be increasingly suc-cessful vis-a-vis their sedentary enemies until theeventual invention of firearms leveled the playingfield. Far from being a peripheral consideration,uniquely nomadic military technology operated si-multaneously with good leadership and high mo-bility in successful nomadic armies, and eachfactor complemented the advantages conferred bythe others. The loss of even one of these advan-tages would have seriously impoverished the abil-ity of a nomadic society to mount successfulcampaigns against well-equipped sedentary foes.

An added benefit of incorporating technologicalimprovements into our explanatory frameworks isthe potential for such a perspective to explain notonly nomadic victories over powerful sedentaryfoes, but also differential successes between differ-ent nomadic groups over time. Using two compar-ative case studies, I will argue that the mediocresuccesses of the Huns in the 5th century and thedazzling successes of the Mongols in the 13th cen-tury are due to differences in archery andsaddle/stirrup technology in addition to other fac-tors such as quality of military leadership. Despitethe tendency of posterity to assume that onemounted archer is equal to another, from a techno-logical perspective, this is simply not the case.

Although the Huns and Mongols are hardly theonly two nomadic groups to practice mounted

archery successfully, sev-eral factors make themideal for comparison.First, their origins trace tothe same geographic areaof the world (Kim 2016: 6;May 2006: 630). Theywere likewise both inheri-tors of similar nomadicmilitary traditions derivedfrom their common ances-tral group, the Xiongnu(circa 300 BCE-200 CE)(Golden 2011: 33; Vaissere2005). The Xiongnu wereimportant technologicalinnovators, introducing tomounted archery severalimportant new develop-

Fig. 1. Mongol riders escorting prisoners, from an early 14th-century illustration of Rashidad-Din’s “Universal History” (Gami at-tawarih). The riders and mounts pictured on eitherside of the prisoners offer a glimpse of Mongol stirrups and quivers, while the mount on theleft is also equipped with a saddle. Bibliothèque national de France, Département des Manu-scrits, Division orientale, Supplement Persan 1113, folio 231v.

79

ments, including paired stirrups in the fifth cen-tury CE and stiffening bone plates on the limbends of their composite bows. The Huns of Europehad the stiffening bone plates that were first devel-oped by the Xiongnu but lacked the technologicalinnovations that the Xiongnu remaining in InnerEurasia developed in the fifth century and subse-quent periods, such as thepaired stirrup. But by the timeof the Mongols, these inven-tions had been widely adoptedand mastered in Inner Eurasia.An understanding of Mongoltechnology, such as their use ofpaired stirrups and an im-proved composite bow design,is important in explaining thetechnological supremacy and,by extension, the enhancedmilitary capabilities of theMongols. Like leadership, tac-tics, and politics, however, anyappeal to technology remainsonly one part of a larger com-posite picture.

Not all aspects of Hun andMongol military technology can

be attributed to the Xiongnu.Both groups were inheritors of along nomadic tradition ofmounted archery, and the arse-nal and practices of both groupsreflect the contributions ofmany others. Still, though, at afoundational level Hun andMongol military practices aremarked more by similarity thanby difference, and this makesthe subtle differences that doexist between them especiallyilluminating. The Huns andMongols are also comparativelywell-studied archaeologically,with enough surviving exam-ples of their bows and eques-trian accoutrement to permit athorough discussion that iswell-grounded in empiricaldata.

The Technology of Mounted Archery

The primary weapon of every nomadic mountedarcher was the composite bow, defined as a bowcomposed of at least three layers of varying mate-rials (Reisinger 2010: 44). Sometimes, these bowsare also termed “Scythian bows” after their sup-

Fig. 2. Diagram of a composite bow. (After: Hank Iken, in Grayson et al. 2007, Tradi-tional Archery from Six Continents)

Fig. 3. Diagram of a self bow. (After: Hank Iken, in Grayson et al. 2007, TraditionalArchery from Six Continents)

80

posed inventors (Mock 2013: 52). Composite bowsare distinct from self bows, which are made from asingle material such as a wooden stave, and lami-nated bows, which are made from several bondedlayers of the same material, usually wood(Bergman, et al. 1988: 660). The earliest archaeo-logical example of a nomadic composite bow datesto at least 1000 BCE., based on the 2010 discoveryof a Scythian-style bow in the Yanghai cemetery ofXinjiang province in the People’s Republic ofChina (Beck, et al. 2014: 225; Karpowicz and Selby2010: 94). All later nomadic bows were variants ofthis basic type. While far from common, thesebows are not as archaeologically rare as one mightthink (Hall 2005: 28).1

For most nomads, the three dissimilar materialsthat comprised the composite bow were wood,horn, and sinew (Paterson 1984: 38). Wood formsthe core and grip of the bow and is backed bysinew to add tensile strength. It is then frontedwith horn, which has a high coefficient of restitu-

tion—that is, its springiness lets it return quicklyto its original shape after being subjected to com-pression. The energy of the decompressing hornserves two purposes simultaneously: it lends powerto the bow’s release and helps the front of thewooden core—the belly—resist compression amidrepeated use (Bergman and McEwen 1997: 145;Reisinger 2010: 44). The sinew serves the samefunction: after being stretched, it returns quicklyto its resting position, again protecting the woodencore of the bow and storing additional potentialenergy to be transferred to the arrow upon the re-lease of the string. The impact on the capabilitiesof the finished bow are significant, as horn has 3.5times the compression resistance of wood, whilesinew can stretch five times as far as hardwoodwithout breaking. The end result is a bow thatboth stores energy and transfers it to the arrowmuch more efficiently than a self bow, and is alsomuch smaller (Bergman and McEwen 1997: 145).The application of these materials is not uniformand varies across space and time. Some bows, suchas examples uncovered from Miran, China arebacked with sinew nearly to the nocks (Hall andFarrell 2008: 90). Others, such as the Mongol-pe-riod Omnogov bow (discussed below) adopt amuch more minimalist design that increases therecovery speed of the bow’s limbs and therefore itsenergy efficiency (Atex and Menes 1995: 75).

Composite bows are also often recurved, reflexed,or both. In a strung recurved bow, the limbs bendforward, away from the archer. In an unstrung re-flexed bow, the entire limbs of the bow reversethemselves away from the direction of the draw.This innovation invests composite reflexed and re-curved bows with greater efficiency than non-re-curved or reflexed bows. By preloading tension oneven the undrawn strung bow, the reflexing and re-curving limbs allow more potential energy to bestored in the limbs at full draw with a lighter drawweight (due to the leverage conferred by the re-curved limbs). This lends greater force and velocityto the arrow upon release and allows the bow unitto be physically much shorter without reducingthe draw length, an important consideration forarchers aspiring to shoot from horseback(Bergman, et al. 1988: 660; Reisinger 2010: 45). Selfbows, on the other hand, cannot be shortenedwithout significantly shortening the draw length,

Fig. 4. Early 14th-century depiction of Mongol archers shoot-ing with composite bows, from an illustration of Rashid ad-Din’s “Universal History” (Gami at-tawarih). Watercolor onpaper. Bibliothèque national de France, Département desManuscrits, Division orientale, Supplement Persan 1113, folio231v.

81

since wood unsupported by other materials canonly bend so far before breaking. As an added ben-efit, the shortness of the composite recurve’s limbsmake them lighter, so less potential energy iswasted moving the limbs back to their original po-sition. This energy is instead transferred to thearrow, and ultimately, the target it impacts.2

When strung, many composite recurves are lessthan three feet from end to end. Most are in thevicinity of sixty centimeters (Drews 2004: 101). Wemight compare this figure to the English longbow,a self bow made from a single stave of yew or elm.These bows were usually six or more feet in length(~183 cm+) and required much more effort to drawthan a similarly powerful recurve bow. Without theleverage of recurved limbs, and the additional po-tential energy stored in the sinew and horn of acomposite bow, all the energy to be transferred tothe arrow had to come from one source only: themuscle power of the archer, who bent the bow’swood. Composite bows clearly were superior fromthis perspective, as they provided as much or morepower with much less energy required for eachdraw (Emeneau 1953: 78). In addition to allowingmore effective archery from horseback (longbowscan be used from horseback with difficulty), short,efficient nomad bows allowed people who wouldnever have been strong enough to draw an English-style longbow to be full participants in the no-madic mounted archer army. Composite recurvebows are also sometimes asymmetric, with thelower limb being shorter than the upper limb—animportant design choice that allowed mountedarchers to rotate easily to aim at targets on eitherside of their mount, provided they had the appro-priate saddle technology to enable this. Despitethe general features discussed above that werecommon in all composite bows, there were definitedifferences between Hunnic and Mongol bows thatrendered Mongol bows superior in a variety ofways, differences that I propose were at least par-tially responsible for their differential successes onthe battlefield.

Maintenance was a constant issue. Extreme tem-perature changes or exposure to dampness couldwarp the limbs, and twists in the limbs could ren-der such bows inaccurate at best and useless atworst. Taybughā l-Ashraf ī l-Baklamishī l-Yūnanī, aMameluke author who penned an archer’s manual

for beginners in the fourteenth century, advisedarchers in cold weather to “put the bow inside hisclothes and warm it with his body. When going tobed at night, he should also keep the bow insidehis clothes to protect it from the damp” (Lathamand Paterson 1970: 94). Such discomfort wasworthwhile given the difficulty of repairing warpedlimbs. In order do so, archers would have had towarm their bows over a fire and apply the appro-priate corrective pressures. Even after careful andskilled repair, however, the bow would never bequite the same, especially if an overly zealousowner overcorrected for the original warping(Loades 2016: 27). Though the Huns used a bowthat featured a design change that made it moredurable in the long term, this came at the price ofreduced energy transfer efficiency to the arrow.

Hun and Mongol Bows

In the seasonally variable and damp climate ofeastern Europe, where most Hun sites are found,organic material such as horn, wood, and sinew donot preserve well. If these were the only compo-nents of Hunnic bows, archaeologists would belimited to the few fragmentary and questionableprimary sources passed down to us by Roman his-torians with an interest in Hun culture. Fortu-nately, by the fourth century CE, a new technologyhad been applied to the traditional composite re-curve design: stiffening bone plates attached to thegrip and limb ends of bows, which minimized thewarping effect that humidity and fluctuating tem-peratures could cause (Boie and Bader 1995: 29).Man (2005) compares the bone plates to finger-nails on the end of a human digit (99). This is anapt analogy—the bone plates provide a rigidity tothe limb ends that wood alone cannot, thus help-ing to prevent twist and warp. Although bows withthis feature are frequently called “Hun bows,” theXiongnu of Inner Asia from whom the Huns de-rived were actually the first to add such plates tothe nomadic composite bow design, and suchmodifications appear across Eurasia after the ini-tial Xiongnu heyday. Strictly speaking, it is there-fore a pan-Eurasian design rather than a uniquelyHunnic design.3

Hun tradition dictated that warriors be buriedwith their bows across their chest. A number ofHunnic graves across Europe and West Asia have

82

yielded stiffening bone pieces that were recoveredboth intact and in situ (Loades 2016: 17). Carefulmeasurements have allowed for the reconstructionof the size and shape of the original bows, thoughunfortunately without the other original materialsof construction. For archaeologists, the inclusionof these stiffening plates is fortunate, for the bonethey are made from preserves quite well in poorconditions. They therefore allow us to study theconstruction of Hun bows that have otherwiselong since decomposed.

The durability these plates added to Hun bowscame at the price of efficiency. Bone is heavier thanthe other materials that make up composite bows,and therefore, it takes more of the drawn bow’s po-tential energy to accelerate these heavy bone addi-tions and move them back to the strung, undrawnposition. While this is also true of horn and sinew,the crucial difference is that both horn and sinewstore additional potential energy in a drawn com-posite bow, more than compensating for their useof some additional potential energy during release.The stiffening bone plates, on the other hand, con-sumed energy without contributing any. The en-ergy used to move them, which would otherwisehave been transferred to the arrow, was insteadlost, with a result of decreased arrow velocity andpenetration at the point of impact. It is not cur-rently possible to determine the exact amount oflost energy per shot, since without the specifica-tions of the other materials in the bow it is impos-sible to do so accurately. Given the weight of thebone, however, the amount of lost energy entailedby such an addition must have been significant(Atex and Menes 1995: 75).

As far as the limited archaeological evidence candemonstrate, conquest-period Mongol bows werevirtually identical to Hunnic bows, minus the stiff-ening bone plates. Modern Mongolian bows are ofno comparative use here; by the 17th century, theMongols had abandoned the use of the bow in warand it was only in the mid-eighteenth century thatthey re-adopted a variation of the ChineseManchu/Qing bow into their arsenal. Qing bows,designed to compete with European muskets bydelivering extremely heavy arrows at high veloci-ties, are much larger and heavier than the Mongolconquest-period bows. Michael Loades describesthem aptly as “the longest and most massive of allcomposite bow types … it was a bow for the powershot, rather than the rapid shot” (2016: 20-21). Assuch, they were very different from the light butstill powerful bows of the Chingissid Age.

Only two complete conquest-period Mongol bowshave ever been found, the most recent one in 2010in a cave at Tsagaan Khad (White Mountain) inmodern Mongolia’s Ovorkhangaj Aimag. The drycave environment in which it was deposited al-lowed for an extraordinary level of preservation:even traces of the original red, black, and yellowpigments survive, along with inlaid gold leaf. Dat-ing to the 14th or 15th century CE, even the red silkstring survived intact—upon recovery, the bow wasstill strung (Loades 2016: 19). The stress on thelimbs resulting from being constantly tensed bythe string over many centuries resulted in signifi-cant warping, but not so much that the originalshape, specifications, and composition of the bowcould not be analyzed (details in Ahrens, et al.2015: 685; Biro 2013: 17).

Fig. 5. The Tsagaan Khad (White Rocks) bow. National Museum of Mongolia. All photos courtesy of author.

83

The other conquest period bow is even better pre-served, and similarly designed. The Omnogov Bow,as it is known, was discovered in 1984, also in acave burial, at Ikh Bayany Agui in Mongolia’sOmnogov Aimag. Though some scholars have sug-gested that the bow dates to as late as 1720 CE,most analysts agree that the bow is in fact muchearlier, dating to the 12th or 13th century CE(Ahrens, et al. 2015: 686). The virtually identicaldesign of the convincingly dated Tsagaan Khadbow lends support to the earlier date. The Omno-gov Bow, like all known Mongol-period bows,lacked the stiffening bone plates of the older Hun-nic bows. The elimination of the performance-re-ducing bone stiffeners to the limb ends of the bowis the primary design difference between Hunnicand Mongol bows, one which rendered Mongolbows superior. According to Atex and Menes,“doing away with the mass and weight of the bonetips would have added a considerable amount ofspeed to the bow.” Bone, and the adhesives neededto bond it to the wooden core, they note, is roughlytwice the weight of an equivalent amount of hard-wood. Thus its elimination “would allow a muchhigher recovery speed of the tips, greatly increas-ing arrow speed.” The Mongol bow, then, “was a lit-tle shorter than that of the Hun and with the lighttips would have been far superior” in terms ofarrow speed (1995: 75).

While the stiffening bone plates of the Hunnicbows made them more durable, they would haveadded weight to the bow that resulted in wastedenergy from every shot, which translates into lowerarrow velocities, penetrating power, and shorterranges. The significance of even a slight edge interms of arrow velocity, range, and penetratingpower should not be underestimated. Althoughsuch an observation alone is clearly not enough toexplain Hunnic successes versus Mongol successes,we should bear in mind that different weapon ca-pabilities doubtlessly played at least some role inthe differential military successes of the twogroups. In the future, perhaps detailed reconstruc-tions of the Tsagaan Khad and Omnogov bows willpermit more detailed assessments of their capabil-ity. If a complete Hunnic bow is ever found and re-constructed, a much more rigorous comparison oftheir differential capabilities could be undertaken.

But the recurved composite bow was not the onlycrucial piece of military technology deployed by

Fig. 6. 2nd-3rd century wooden saddle.

Fig. 7. 7th-8th century wooden saddle.

Fig. 8. 13th-14th century replica of a Mongol-era saddle.

84

the nomads. Saddles, and especially the later addi-tion of stirrups, provided the platforms fromwhich mounted archers traveled and fought.Therefore, in order to fully understand the battle-field dynamics of nomadic armies, an examinationof saddle and stirrup technological innovations isessential. Again, the Mongols enjoyed a subtle butsignificant technological advantage over the Huns,one that made them much more deadly asmounted archers. The Huns of the fourth and fifthcenturies CE used wooden saddles without pairedstirrups. This is confirmed not only by archaeologi-cal finds but also by primary source texts. The tex-tual basis for this claim comes from a reference inJordanes’ Getica. Jordanes, a Gothic historian,wrote from Constantinople in 551 CE, a centuryafter the Huns and Romans clashed at the Battle ofthe Catalaunian Plains. Jordanes records thatAtilla, sensing defeat, ordered a great funeral pyreof saddles to be erected, on which he would throwhimself into the flames so as to deprive the Ro-mans the satisfaction of killing or capturing him(Jordanes 2014: 43). The fact that saddles were suit-able for a pyre is ample evidence that they werewooden. Archaeological evidence lends furthersupport to this. In fact, during this period, woodensaddles among nomadic groups were the normrather than the exception; in fact, they are well-at-tested pieces of nomadic equestrian accoutrementin many places and times (Tkačenko 2010). But in

the case of the Huns, thereis no evidence of stirrups,either in textual or archae-ological sources. (There isdisagreement among schol-ars whether the Huns usedcloth or leather toe loopsstrictly as mounting aids,along with how widespreadthe toe loops may havebeen if they existed at all.)

The Roman cavalry, whichwas contemporaneous tothe Huns, used woodensaddle technology withoutstirrups that was borrowedfrom the Parthians. Pre-sumably, then, the pres-ence and utilization of

stirrups would have been worth observing, record-ing, and discussing for Roman authors. Though ex-cavations at Hunnic sites have turned up bits,fragments of wooden saddles, and bridle orna-ments, not a single stirrup or anything that couldbe interpreted as one has been found (Istvanovitsand Kulcsar 2014: 269; Maenchen-Helfen 1979:209; Man 2005: 56); it is generally agreed that stir-rups did not reach Europe until the arrival of theAvars in the late 6th or early 7th century CE (May2018: 5).

Fig. 9. 7th-8th century iron stirrups.

Fig. 10. Iron stirrups from the Mongol era.

85

Although the lack of stirrups was a challenge to thepractice of horse archery for groups like the Huns,it was not an insurmountable one. Loades, basedon experiments conducted using a Parthianwooden four-horn saddle without stirrups (a no-madic design temporally close to the time of theHuns), has noted that leaning into the front hornslifted the rider’s seat almost as effectively as stand-ing in paired stirrups, and enabled their hips to ab-sorb shocks and minimize the jostling that candisrupt aim at the moment of shooting. Stirrups,he concludes, were not an essential prerequisite forhorse archery. Archaeological evidence of Loades’technique for practicing mounted archery withParthian four-horn wooden saddles can be foundin a stone carving of a Parthian mounted archerheld in Berlin’s Museum für Islamisches Kunst(Driel-Murray, et al. 2002: 17; Loades 2016: 55).

The lack of stirrups constituted a significant hand-icap for mounted archers, even if it was not insur-mountable. Stirrups enable a more stable platformfor shooting by allowing archers to rise partially inthe saddle and use their knees as shock absorbers,and this in turn allowed the archer to recruit theirleg and core muscles in order to draw heavier bowswhile riding. From a seated position, only the mus-cles of the arm and chest can be recruited into thedraw. Without the muscles of the leg and core toaid in drawing, the Hunnic bows would most likelyhave been lighter in draw weight than later Mongolbows. The lack of preserved organic material froma Hunnic bow precludes the calculation of drawweights, but we can combine this observation withevidence noted above: namely, that the stiffeningbone plates of Hunnic bows would have reducedtheir efficiency. Therefore, even at the peak ofHunnic prowess, Hunnic bows would have beeninferior to the Mongol bow in three ways: drawweight, energy transfer efficiency, and in the ab-sence of sturdy paired stirrups, ease of handlingfrom horseback.

A skilled rider equipped with stirrups can control ahorse with his or her knees even without placinghis hands on the reigns. For a Hunnic rider, theonly solid point of contact with the horse whileshooting would have been the hips/pelvis. With-out stirrups, hands-free control of the horse wouldhave been impossible. In order to control theirhorses, Hunnic soldiers would have had to cease

firing and grasp the reigns. Though Loades’ experi-ments demonstrate that mounted archery can bepracticed with only a good wooden saddle, it alsoprovides evidence that mounted archery is muchmore effective if the rider can control the horseand fire their bow at the same time.

Another decisive liability that came with the Hun-nic lack of stirrups was the inability of riders to in-flict or sustain a shock while on horseback withoutbeing dismounted (Christian 1998: 281; Dien 1986:36; Goodrich 1984: 285; White 1964: 1-2). The in-ability to maneuver one’s position in the saddlewhile riding at high speeds—let alone while shoot-ing—without stirrups was also disadvantageous.The Xiongnu who threatened the northern frontierof Han China provide compelling evidence thatpre-stirrup strategies were largely limited to hit-and-run style raids rather than prolonged conflictwith enemy armies (Christian 1998: 191; Drews2004: 116). The Huns, who never won a battleagainst a full-strength Roman field army and in-flicted most of their damage in the absence of seri-ous organized resistance, exemplified this strategicapproach. While effective in certain settings, suchlimited capabilities were best paired with strate-gies that aimed to keep an enemy off balancerather than engage them in a prolonged war ofconquest. Not surprisingly, this is precisely the sortof behavior ancient authors record the Huns as ex-celling at.

Of course, hit and run strategies remained impor-tant to pastoralists even after the acquisition of thestirrup, but its acquisition allowed pastoralist tac-tics to evolve in several significant ways that madethem devastating to opponents, especially footarmies. The horned wooden saddles of the Hunnicperiod were a tremendous improvement over ear-lier Scythian period “soft saddles,” which consistedof two leather pads, sewn together, filled with hairor plant material, and crudely attached to thehorse with a simple girth strap (Olsen 1988: 186;Tkačenko 2010: 1). Such simple saddles are knownarchaeologically from the 5th to 3rd century BCEPazyryk burials. But still, the lack of stirrups was aserious military handicap for the Huns, one thatlimited the effectiveness of their engagements withRoman military forces. Nearly a millennium later,however, exploiting the advantages of better bowand saddle/stirrup technology would have been

86

second nature to the Mongols—and enabled themto engage even the best-trained infantry or heavycavalry army much more effectively than the Hunshad ever done.

It should be noted that Mongol saddles were alsowooden. Not only that, but Mongol bows weremorphologically similar to those of the Huns.However, Mongols saddles and bows were com-bined with sturdy paired stirrups. Again, this is at-tested both archaeologically and textually.Generally speaking, usually only women madeleather and cloth goods among the Mongols, asrecorded by William of Rubruck in the thirteenthcentury (Dawson 1966: 97). Men are very clearlylisted as the makers of wooden and metal goods:“the men make bows and arrows, manufacture stir-rups and bits and make saddles …” (Dawson 1966:103). The Mongols retained the nomadic traditionof crafting saddles from wood and paired themwith metal (probably iron) stirrups.

It is, however, not likely that stirrups were a recentinvention at the time of the Mongols. Tkačenko(2010: 2) claims that saddles and stirrups were firstpaired sometime in the early first millennium CEin the region of the Xiongnu confederation, fromwhom the Huns split off and the Mongols de-scended. Littauer (2002: 439) argues for an evenmore precise origin point in the 5th century CE,and May (2018: 5) supports the view that they werepresent among the Xianbei (a nomadic group wholived in what is today eastern Mongolia, InnerMongolia, and Northeast China) by the early 4thcentury CE, from whence they made their way intoChina proper. Given the Hunnic lack of stirrups, itseems clear that the technology developed too latefor the migrating Huns to carry with them on theirway west to Europe, but probably developedshortly thereafter. So though the Huns lacked stir-rups, by the time of the Mongols, Inner Eurasiannomads had possessed them and been masteringtheir use in combat for nearly a millennium.

Ultimately, technology, like leadership or mobility,is only one piece of a tangled web of intertwinedcauses and effects that tell the tale of nomadicwarfare. It would be misguided to attribute the riseof a stunningly successful nomadic group like theMongols to superior technology alone. There weremany nomadic groups who were chronologically

much closer in time to the Mongols than were theHuns that possessed similar riding technology butdid not come even close to achieving the same mil-itary successes. This alone should be taken as suffi-cient evidence that leadership, politics, and soforth retain a considerable degree of explanatoryutility. To appeal to technology alone would be tovastly overstate the case the evidence supports. Atthe same time, however, it would be equally foolishto ignore the role that technology played in en-abling some nomads to best their sedentary peerswhere their predecessors had been only annoy-ances. In the future, perhaps detailed reconstruc-tions of nomadic bows will permit detailedcalculations of draw weights and, in turn, arrowspeed and penetrating power. If a complete Hun-nic bow is ever discovered, similar reconstructionswould allow the precise effect of the performance-limiting bone plates to be assessed. But even ifsuch discoveries never come to light, the evidencethat does exist more than supports the thesis thatconsiderations of technology deserve to be inte-grated into analyses of nomadic societies muchmore thoroughly than they previously have been.

ABOUT THE AUTHOR

Samuel Rumschlag is a Ph.D. student in the De-partment of Anthropology at the University ofWisconsin-Madison. His research focuses on con-flict between nomadic steppe peoples and earlyChinese dynasties in the frontier region of what isnow southern Mongolia and northern China. E-mail: <rumschlag@wisc.edu>.

ACKNOWLEDGMENTS

The author would like to thank the Center for Rus-sia, Eastern Europe, and Central Asia at the Uni-versity of Wisconsin-Madison, who providedresearch funding for travel to Mongolia wheremuch of this research was conducted. Thanks arealso due to the Department of Anthropology at theUniversity of Wisconsin-Madison for their gener-ous support over the past two academic years, andto Dr. Nam C. Kim and Dr. J. M. Kenoyer, my aca-demic advisers for their generous gifts of guidance,insight, and time during my graduate career so far.

87

REFERENCES

Ahrens et al 2015Birte Ahrens, H. Piezonka, and G. Nomguunsuren.“Buried with His Bow and Arrows: The ExceptionalCave Burial of a 14th Century Warrior at TsagaanKhad Mountain, Mongolia.” In: T. La, ed., AncientCultures of the Northern Area of China, Mongoliaand Baikalian Siberia, Beijing: Huhhot, 2015: 683-91.

Atex and Menes 1995Wayne Atex and G. Menes. “The Omnogov Bow:The Results of a Search for the Bow of GenghisKhan.” Journal of the Society of Archer Antiquaries38 (1995): 71-75.

Baker 1992T. Baker. “Bow Design and Performance.” In: J.Hamm, ed., The Traditional Bowyer’s Bible, Azle:Bois d’Arc Pr., 1992: 43-116.

Balfour 1890Henry Balfour. “On the Structure and Affinities ofthe Composite Bow.” Journal of the AnthropologicalInstitute of Great Britain 19 (1890): 220-50.

Bartlett 2010William Bartlett. The Mongols. Chalford: Amber-ley Publishing, 2010.

Beck et al. 2014Ulrike Beck, Mayke Wagner, Xiao Li, DesmondDurkin-Meisterernst, and Pavel E. Tarasov. “TheInvention of Trousers and Its Likely Affiliation withHorseback Riding and Mobility: A Case Study ofLate 2nd millennium BC Finds from Turfan inEastern Central Asia.” Quaternary International348 (2014): 224-35.

Bergman and McEwen 1997Christopher Bergman and Edward McEwen.“Sinew-Reinforced and Composite Bows: Technol-ogy, Function and Social Implications.” In: H.Knecht, ed., Projectile Technology, New York,Plenum Press, 1997: 143-60.

Bergman et al. 1988Christopher A. Bergman, Edward McEwen, and R.Miller. “Experimental Archery: Projectile Velocitiesand Comparison of Bow Performances.” Antiquity62 (1988): 658-70.

Biro 2013Adam Biro. “Methodological Considerations on

the Archaeology of Rigid, Reflex, Composite Bowsof Eurasia in the Pre-Mongol Period.” Acta Mili-taria Mediaevalia 9 (2013): 7-38.

Boie and Bader 1995A. Boie and H. D. Bader. “Bows Used by the Huns.”Journal of the Society of Archer Antiquaries (1995):28-32.

Christian 1998David Christian. A History of Russia, Central Asiaand Mongolia. Malden: Blackwell, 1998.

Dawson 1996Christopher Dawson, ed. Mission to Asia: Narra-tives and Letters of the Franciscan Misisonaries inMongolia and China in the Thirteenth and Four-teenth Centuries. New York: Harper, 1996.

Di Cosmo 1999Nicola Di Cosmo. “State Formation and Periodiza-tion in Inner Asian History.” Journal of World His-tory 10, no. 1 (1999): 1-40.

Dien 1986Albert E. Dien. “The Stirrup and Its Effect on Chi-nese Military History.” Ars Orientalis 16 (1986): 33-56.

Drews 2004Robert Drews. Early Riders: The Beginnings ofMounted Warfare in Asia and Europe. New York:Routledge, 2004.

Driel-Murray et al. 2002Carol van Driel-Murray, Peter Connolly, and JohnDuckham. “Roman Saddles: Archaeology and Ex-periment 20 Years On.” In: L. Gilmour, ed., In theSaddle: An Exploration of the Saddle Through His-tory. London: Archetype, 2002: 1-19.

Drompp 2005Michael R. Drompp. “Imperial State Formation inInner Asia: The Early Turkic Empires (6th to 9thCenturies).” Acta Orientalia Academiae Scien-tiarum Hungaricae 58, no. 1 (2005): 101-11.

Emeneau 1953Murray B. Emeneau. “The Composite Bow inIndia.” Proceedings of the American PhilosophicalSociety 97, no. 1 (1953): 77-87.

Golden 2000Peter B. Golden. “‘I Will Give the People untoThee’: The Činggisid Conquests and Their After-math in the Turkic World.” Journal Of the Royal

88

Asiatic Society 10, no. 1 (2000): 21-41.

Golden 2011Peter B. Golden. Central Asia in World History.New York: Oxford University Press, 2011.

Goodrich 1984Chauncey S. Goodrich. “Riding Astride and TheSaddle in Ancient China.” Harvard Journal of Asi-atic Studies 44, no. 2 (1984): 279-306.

Grayson et al. 2007Charles E. Grayson, Mary French, and Michael J.O’Brien. Traditional Archery from Six Continents.Columbia, Missouri: University of Missouri Press,2007.

Hall 2005Andrew Hall. “Some Well-Preserved CompositeBows.” Journal of the Society of Archer Antiquaries48 (2005): 28-36.

Hall and Farrell 2008Andrew Hall and Jack Farrell. “Bows and Arrowsfrom Miran, China.” Journal of the Society ofArcher Antiquaries 51 (2008): 89-98.

Istvanovits and Kulcsar 2014Eszter Istvanovits and Valeria Kulcsar. “New Findof Hun Age Saddle Plates From North-East Hun-gary.” In: S. Cocis, ed., Archäologische Beiträge:Gedenkschrift zum hundertsten Geburtstag vonKurt Horedt, Cluj-Napoca: Patrimonium Archaeo-logicum Transylvanicum, 2014, 269-78.

Jordanes 2014Jordanes. The Origin and Deeds of the Goths.Translated by C. C. Mierow. San Bernardino: HardPress, 2014.

Karpowicz and Selby 2010Adam Karpowicz and Stephen Selby. “ScythianBow from Xinjang.” Journal of the Society of ArcherAntiquaries 53 (2010): 94-102.

Kim 2016Hyun Jin Kim. The Huns. New York: Routledge,2016.

Kooi 1996B. W. Kooi. “Functioning of Ears and Set-back atGrip of Asiatic Bows.” Journal of the Society ofArcher Antiquaries 39 (1996): 73-77.

Latham and Paterson 1970J. D. Latham and W.F. Paterson. Saracen Archery.

London: Holland Press, 1970.

Lieberman 2008Victor Lieberman. “Protected Rimlands and Ex-posed Zones: Reconfiguring Premodern Eurasia.”Comparative Studies in Society and History 50, no.3 (2008): 692-723.

Littauer 2002Mary Aiken Littauer. “Early Stirrups.” In: P.Raulwing, ed., Selected Writings on Chariots andOther Early Vehicles, Riding and Harness, Leiden:Brill, 2002, 439-51.

Loades 2016Michael Loades. The Compound Bow. Oxford: Os-prey, 2016.

Maenchen-Helfen 1979Otto Maenchen-Helfen. The World of the Huns:Studies in Their History and Culture. Los Angeles:University of California Press, 1979.

Man 2005John Man. Attila: The Barbarian King who Chal-lenged Rome. New York: Thomas Dunne, 2005.

May 2006Timothy May. “The Training of an Inner AsianNomad Army in the Pre-Modern Period.” Journalof Military History 70, no. 3 (2006): 617-35.

May 2018Timothy May. “Nomadic Warfare Before Firearms.”In: Oxford Research Encyclopedia of Asian History.Available online: http://oxfordre.com/asianhis-tory/view/10.1093/acrefore/9780190277727.001.0001/acrefore-9780190277727-e-4

Mock 2013John Mock. “New Discoveries of Rock Art inAfghanistan’s Wakhan Corridor and Pamir: A Pre-liminary Study.” The Silk Road 11 (2013): 36-53.

Morgan 1986David Morgan. The Mongols. Oxford: Blackwell,1986.

Olsen 1988Stanley J. Olsen. “The Horse in Ancient China andIts Cultural Influence in Some Other Areas.” Pro-ceedings of the Academy of Natural Sciences ofPhiladelphia 140, no. 2 (1988): 151-89.

Paterson 1984W.F. Paterson. Encyclopaedia of Archery. London:

89

Robert Hale, 1984.

Reisinger 2010 Michaela Reisinger. “New Evidence about Compos-ite Bows and Their Arrows in Inner Asia.” The SilkRoad 8 (2010): 42-62.

Sinor 1972Denis Sinor. “Horse and Pasture in Inner Asian his-tory.” Oriens extremus 19, no. 1 (1972): 171-83.

Smith 1975John Masson Smith Jr. “Mongol Manpower andPersian Population.” Journal of the Economic andSocial History of the Orient 18, no. 3 (1975): 271-99.

Smith 1984John Masson Smith Jr. “Ayn Jālūt: Mamlūk Sucessor Mongol Failure?” Harvard Journal of AsiaticStudies 44, no. 2 (1984): 307-45.

Thompson 1996E. A. Thompson. The Huns. Cambridge: Blackwell,1996.

Tkačenko 2010Irina Dmitrievna Tkačenko. “Riding Horse Tackamong the Cattle-breeders of Central Asia andSouthern Siberia in the First and Second MillenniaCE.” Études mongoles et sibériennes, centrasia-tiques et tibétaines 41 (2010). Available online:https://journals.openedition.org/emscat/1552

De la Vaissere 2005Etienne de la Vaissere. “Huns et Xiongnu.” CentralAsiatic Journal 49, no. 1 (2005): 3-26.

White 1964Lynne White. Medieval Technology and SocialChange. Oxford: Clarendon, 1962.

NOTES1 Biro (2013) has provided a thorough discussion of the termi-nology surrounding the academic study of archaeologicalfinds relating to archery.2 For a more detailed discussion of bow mechanics, see Baker1992, Balfour 1890, Kooi 1996, Loades 2016, and McEwen etal. 1991.3 Though I have retained the designation “Hun bow” to de-scribe the technology employed by the Huns of Europe,readers should remain aware that the application of the boneplate technology was unique neither to the Huns or to Eu-rope.

90