III/ 7 - Pre-Samurai Japanese: a suggested revision
Author: Duncan Head
Third and final version April 2006
Introduction:
The major changes that occurred between published versions 1 and 2 of this list reflect the new information that came to light in the intervening period, particularly in the books of Farris and Friday (both 1992). Since then there has been more time to consider this evidence, and more relevant material has been published (notably Friday 2004). I therefore suggest another revision, breaking up the subject for consideration into four lists.
Proposal:
Replace the existing list III/7 with the following four lists:
III 7/A Yayoi Japanese 500 BC – 275 AD
III 7/B Kofun Japanese, 275 AD – 646 AD
III 7/C Ritsuryoh Japanese, 646 AD – 811 AD
III 7/D Early Heian Japanese, 811 AD – 1020 AD
(Numbering as “III 7/A” and so on is purely for convenient reference; the first two proposed lists would actually fit into Book II, as presently constituted. List D could alternatively be treated as a sub-list of the Early Samurai list III/54.)
Two issues on troop classification that span more than one of the proposed lists are discussed towards the end of this document.
*******
III 7/A Yayoi Japanese 500 BC – 275 AD
{DBM:} Cold. WW, Rv, _H(S)_, Wd, O, E, RGo, BUA.
{DBMM:} Cool. S, Rv, _DH_, SH, Wd, O, E, SF, BF, BUA.
Ag 1.
Nominal list scale: One element = 250 men (normal scale)
Commander-in-chief – as archers Irr Bw (I), swordsmen Irr Bd (I), or spearmen Irr Sp (I) 1
Allied general with the same options 0-3
Replace allied general with sub-general 0-1
Priestess-queen in litter or manhandled wagon, with guards – Irr WWg (I) under DBM, Irr Bge (S) under DBMM 0-1
Archers – Irr Bw (I) 40-80
Upgrade archers to crossbowmen – Irr Bw (O) 0-2
Replace archers with slingers – Irr Ps (O) Any
Dagger-axe-men and/or swordsmen – Irr Bd (I) 15-30
Spearmen – Irr Sp (I) 15-30
Boats – Fishing-boats Irr Bts (I), or larger boats – Irr Bts (O) {Any DBM infantry/DBMM foot} 0-4
Rules considerations:
1. A priestess-queen must be in the C-in-c’s command.
Draft list notes:
This list represents the Yayoi culture of Japan from its origins until the first use of mounded tombs and iron armour ushered in the start of a new era. Continental immigrants mixing with the local people of the neolithic Jomon culture established a rice-farming, iron- and bronze-using culture that soon covered all the Japanese archipelago except Hokkaido and the Ryukyu islands. The Yayoi were not united and even their most powerful rulers controlled only alliances of semi-independent communities, hence the limited sub-general allowed. Chinese sources say that over 100 “Wa” (Japanese) states existed. They were centred on fortified settlements whose moats, palisades and towers indicate frequent internecine warfare. The most important was Yamatai, in either north Kyushu or western Honshu, ruled approx AD 183-248 by the priestess-queen Himiko. She lived in seclusion in a palisaded palace surrounded by armed guards, “occupying herself with magic and sorcery” and attended only by women, while her brother was the public face of the régime. Over 20 statelets were reckoned as her subjects. Despite Himiko’s seclusion we allow such a priestess-queen to be fielded for local colour, in the hope that her total isolation may have been exaggerated by Chinese sources or that other such female rulers (such as her successor Iyo, c.250 to sometime after 266) may not have followed her practice so strictly. Warriors wore simple clothing and tattoos, and fought with spears, dagger-axes, swords, wooden bows with the lower limb shorter than the upper, and slings. We assume archers were in the majority, as in later periods, though some regions favoured the sling over the bow. Spears were 2-3 metres long and shields could be very large, suggesting classification as Spearmen. Some wooden armour has been found. At first both bronze and iron weapons were used, iron being predominant before the end of the period.
Justifications for list A:
Dates –
300 BC used to be the conventional starting date for the Yayoi culture. More recent works suggest about 500 BC (for instance Imamura p.127). The most recent finds and calibrated radiocarbon dates would suggest a date as early as 800 BC for the beginning of Early Yayoi, and perhaps as early as 1000 BC for the more controversial “Earliest Yayoi” period (Keally 2004, 2005). However, since “Early Yayoi dates fall in a range that is effectively impossible to calibrate usefully” (Keally 2005) I have cautiously gone for 500 BC.
As for the end date of the list, 300 AD is sometimes cited as the transition between Yayoi and Kofun periods but 250 is now probably more popular. The date is somewhat arbitrary as the transition was not marked by any single event, but both the monumental tombs from which the Kofun period takes its name and the use of iron armour begin in the later 3rd century – Himiko may have been buried in one of the first mounded tombs. I have settled on 275 to include in this list the rule of Himiko’s female successor Iyo – probably the ruler who sent an embassy to China in 266 (Fredholm p.30).
Terrain –
(For all lists) This is just the terrain in the existing published list, except that I’ve deleted Marsh. Contrary opinions welcome, but I can’t think of a marsh being important in a Japanese battle, and we’d need one that couldn’t be either (a) boggy rough going, or (b) the riverside marsh that any army can have.
For this list only, I’ve dropped Roads, as it seems too early.
Aggression –
The Yayoi expanded throughout Japan at the expense of the previous Jomon-culture inhabitants, but these are not represented by a list since this expansion is unlikely to have involved any warfare on the DBM scale, given the relatively sparse Jomon population (see for instance Imamura p.155 on the “enormous increase in population” in early Yayoi Japan). Later sources (including the Korean Tongguk t’onggam, compiled in 1484 from earlier sources) record several Japanese attacks on Silla in southern Korea, including landings in AD 219, 233 and 249 (Aston mentions the 249 attack, Nihongi v.I p.235 n.1; for the rest see Kanzaki, p.3). Korean lists for this early period do not yet exist, but aggression for them would probably be 0 if they did. A score of 1 seems adequate to reflect isolation but give a reasonable chance of invading Korea.
Climate –
Unlike later lists, it is possible during this period that the main Japanese polity, Himiko’s Yamatai (the location of which is debated), was based in Kyushu, and her rival King Pimikuku’s state of Kunu would almost certainly be Kyushu-based. I did consider whether Kyushu would be far enough south to be classed as Warm, but Luke Ueda-Sarson suggests (from personal experience) that this is not appropriate, except perhaps in the extreme south.
Scale –
There are no hard figures on the strength of Yayoi armies. Chinese sources reckon Himiko’s Yamatai as containing 70,000 households, while the smaller statelets subject to Yamatai range from 1,000 to 50,000. Chinese sources often rate the military potential of foreign states at one warrior per household, which would give Yamatai with its 20-odd subject-states a potential military strength of over 150,000 men. However we don’t know what the warrior:household ratio was in this case, it could well have been much lower than 1:1; and nor do we know how the armies were mobilised or supplied, in other words what fraction of this potential might have been assembled on a battlefield. But the figures do suggest that raising armies of 15,000 or even more men is not beyond the bounds of possibility, and that the largest armies could well have been within the normal scale of the DBM/MM rules.
Generals –
Given that major Yayoi polities seem to have been alliances of communities rather than unitary states, most assistant generals should probably be allies. States as powerful as Himiko’s Yamatai probably merit some subs, but limiting them to one seems adequate.
Priestess-queen –
Obviously this mainly represents Himiko (or Himeko, or Pimiko) and her successor Iyo (or Ichiyo, or Toyo), rulers of Yamatai. The Chinese sources report a period of 70-80 years of continuous warfare among the Wa, ended by the rise of the priestess-queen Himiko of Yamatai (c.AD 183-248). She sent embassies to Wei China between 239 and 247 and over 20 statelets were reckoned as her subjects. After Himiko’s death, a king succeeded to the throne but was overthrown, ushering in a period of civil strife. This was only ended when a 13-year-old female relative of Himiko’s, Iyo, was placed on the throne; she may be the queen who sent an envoy to China in 266. The accounts of the Hou Han shu and Wei zhi are in Tsunoda & Goodrich.
The Chinese accounts are clear on Himiko’s total seclusion, so strictly it is hard to justify allowing her on the battlefield at all. However it would be a shame to leave her out, and it is possible that the Chinese envoys – who were only in Japan relatively briefly, after all – exaggerated her isolation, or that other priestess-queens did not follow her example so strictly. Michael Fredholm suggested that a Chinese ceremonial chariot could be used as a conveyance for Himiko; according to both Hou Han shu and Wei zhi (Tsunoda & Goodrich pp.6, 12), Japan at this period had neither horses nor cattle – which, though exaggerated, suggests any conveyance was probably man-powered.
Weapons, general –
“Weapons are spears, shields, swords and wooden bows” (Hou Han shu); “Their weapons are dagger-axes (ge), shields, and wooden bows made with a short lower half and long upper half. Their bamboo arrows are tipped with iron or bone” (Wei zhi). The translations of both sources in Tsunoda & Goodrich list “spears” in first place; but Michael Fredholm says that the Wei zhi reads ge, dagger-axe (“ko” in Wade-Giles transcription). This character is quite frequently mistranslated “spear”. It is therefore quite possible that the Hou Han shu actually says “ge” as well. But archaeology confirms that both dagger-axes and spears were used, anyway.
Archers –
Stone, bronze and iron arrowheads are common, but I don’t have any way to estimate the relative numbers of archers to other troops. Purely on the basis of later practice, it seems likely that archers were the most numerous troop-type. Farris (1992 p.14) mentions a find of a wooden shield riddled with arrows (perhaps the Kitoragawa shield illustrated in Fredholm 1998?).
Crossbowmen –
What appears to be a Yayoi wooden crossbow-stock was excavated in Izumo-shi, Shimane prefecture, in May 1999 (Friday 2004, p.182 n.42). A bronze crossbow-trigger of Chinese style found a month later in Tsukidate-choh, Miyagi, may be of later date as Friday uses it in discussion of the later oh-yumi (p.75).
Slingers –
“Slings, used to hurl fist-sized rocks or spheres of clay shaped roughly like miniature rugby balls, also appeared during the Yayoi age, distributed in a geographical pattern that suggests mutually exclusive regional preferences for the sling or the bow” (Friday 2004 p.68). This might suggest an army should contain either large numbers of archers with no more than a handful of slingers, or vice versa. But I don’t know whether the “regions” in question would represent a state, several states, or part of a state, so an army (or an allied command) might, for all I know, contain a mix of archer-preference and slinger-preference communities.
The sling virtually disappears after this period, so no slingers in later lists.
Dagger-axe or swordsmen –
Bladed weapons were used in large numbers, but I’m not completely sure whether we need some separate Blade elements, or whether we can treat these mostly as secondary weapons and assume that any individuals who were armed primarily with these are mixed in with the majority spearmen. However, since one cache of buried bronze weapons contained 16 spearheads and as many as 358 swords (Imamura p.175), having some Blades seems the best option.
Spearmen –
Unfortunately we are told nothing of fighting style or tactics, so armament is all we have to go on in classifying these troops. Michael Fredholm says that “we know little of the length of the Yamatai spears, but they do not appear to have been very long” (1998, p.31) and suggests treating them as “light spears/javelins” (p.33). However Friday (2004 p.85) says that Yayoi spears were 2-3 metres long, which suggests thrusting spears significantly taller than their wielders (Yayoi males being up to 162-3cm tall {or about 5’ 4”}, according to Imamura p.147). This, plus the great size of the Kitoragawa shield (illustrated in Fredholm 1998) suggest to me that Spears is a better DBM/MM classification than Michael Fredholm’s suggested Warband (Ordinary). (However, there are alternative possibilities: what if the Kitoragawa shield is in fact an archer’s pavise? Fredholm’s reconstruction shows a substantial handgrip, suggesting a hand-held shield rather than a propped-up pavise, but is that present in the incomplete original, or is it part of the reconstruction? Of course even if the shield is hand-held, we can’t rule out the possibility that it was used to shield an archer, swordsman or dagger-axe-man, but the probability would then be that it was a spearman’s shield.)
A smaller shield is shown in Bryant plate A2. While the spiral shield-plaque shown there is well-attested archaeologically, I do not know what evidence has been used for the size and shape of the shield – possibly later artistic representations.
Boats –
“Naval technology consisted of simple boats constructed from dug-out tree trunks” (Farris 1992 p.14). Kidder (1959 p.102) shows crude illustrations of rowed or paddled boats, possibly fishing-vessels, with up to eight oars a side. Michael Fredholm (1998 p.32) also mentions that larger vessels with sides, prows and sterns built up from planks have been excavated, presumably the type of ship used for voyages to Korea. I assume these are the same as the boats discussed under the next list and therefore best classed as Boats (Ordinary). The fishing-boats would be Bts (I), if they were in fact used for military purposes.
Fortifications –
None. Despite the importance of ditch-and-palisade fortifications for Yayoi settlement sites, I have not come across any evidence for camp or field fortifications.
Non-Yayoi allies –
None. Remnants of the pre-Yayoi Jomon people would probably be Bw (I) and/or Ps(O) anyway so seem hardly worth separate inclusion. Alliances with “predynastic” Korean communities would be purely hypothetical.
*******
III 7/B Kofun Japanese, 275 AD – 646 AD
{DBM:} Cold. WW, Rv, _H(S)_, Wd, O, E, RGo, Rd, BUA.
{DBMM:} Cool. S, Rv, _DH_, SH, Wd, O, E, SF, BF, Rd, BUA.
Ag 2.
Nominal list scale: One element = 250 foot or 200 horse (normal scale)
Commander-in-chief – as uji archers Irr Bw (O) or uji spearmen Irr Sp (I) 1
Sub-general with the same options 0-2
Replace sub-general with allied general 0-2
Uji nobles, toneri and yatsuko retainers as armoured archers with standing shields – Irr Bw (O) {or Irr Bw (S)?} 24-72
Uji nobles and retainers as spearmen – Irr Sp (I) 8-30
Regrade general as swordsman with shield-bearer – Irr Bd (O) 0-1
Regrade uji Bw or Sp as swordsmen – Irr Bd (O) 0-2
Be levied archers – Irr Bw (I) 0-16
Other levies – Irr Hd (O) 0-8
Pacified Emishi archers – Irr Bw (I) or Irr Ps (O) 0-4
Boats – Up to 1/3 horse-transports as Irr Shp (I) {Cv or Kn}, the rest Irr Bts (O) {Any DBM infantry/DBMM foot} 0-6
Camp fortifications – TF @ 1 AP 0-12
Only from 366 AD to 562 AD:
Kaya allies – List: Paekche and Kaya Korean (Book 2)
Only from 366 AD:
Paekche allies – List: Paekche and Kaya Korean (Book 2)
Only from 408 AD to 500 AD:
Replace generals and/or Uji Bw with armoured horse-archers – Irr Cv (O) 0-3
Upgrade general or other Cv (O) to cavalry on armoured horses – Irr Cv (S) (under DBM only?) or Irr Kn (I) 0-1
Only from 475 AD:
Replace C-in-c with yugei bodyguard – Reg Bw (S) 0-1
Yugei guard archers – Reg Bw (S) 0-1
Only from 501 AD:
Replace generals other than yugei and Bd with armoured horse-archers – Irr Cv (O) All
Replace Uji Bw or Sp with armoured horse-archers – Irr Cv (O) 3-8
Only from 618 AD:
Artillery – stone-throwers Reg Art (I) or oh-yumi crossbow-artillery – Reg Art (O) 0-1
Only from 626 to 645 AD:
Replace C-in-c with Soga minister with Eastern Company – Irr Bw (S) 0-1
Rules considerations:
1. A general whose command contains Cv or Kn must be either Cv, Kn or yugei.
2. Yugei must be in the command of a yugei C-in-c.
3. Bd (O) must be in the command of a Bd (O) general.
4. Allied contingents drawn from this list for other armies may include up to 6 naval elements.
Draft list notes:
This list covers the Kofun or “mounded tombs” period, until the Taika "Great Reform" edict began the creation of a Chinese-style conscript army. This period saw great increases in the use of iron armour, the introduction of cavalry, the gradual consolidation of most of Japan under an Imperial dynasty based in the Yamato basin and claiming descent from the Sun goddess, and military interventions in Korea. Armies of 10-25,000 men were common. The members and retainers of aristocratic clans called uji provided most of the troops, initially fighting as archers with iron armour and large leather shields; the yugei “quiver-bearers” were a guards unit armed in the same style. The first iron armours were of laced plates in Chinese style and cuirasses of vertical strips in Korean style, from which the tankoh plate cuirass was developed. Others fought as spearmen. The lower classes were either directly dependent on the uji or organised into guilds called be; the be certainly supplied troops, and others may have done as well. The introduction of cavalry, and the lamellar keikoh armour they wore, was probably inspired by defeats of Japanese infantry against Koguryo cavalry around 400. In 408 a large-scale immigration from the former Chinese commanderies in Korea may have contributed to the introduction of mounted archery; cavalry are mentioned occasionally in the 5th century when there is even some evidence for Korean-style horse-armour, and became commoner in the 6th. In 474 one sub-general, armoured and with a shield-bearer, led a charge against enemy archers; the option for Bd assumes he was accompanied by a retinue. Artillery is first mentioned in 618 when Koguryo presented some oh-yumi bolt-shooters and stone-throwers captured from the Sui Chinese.
Justifications for list B:
Dates –
See List A for the choice of 275 for the start date; probably any date 250-300 AD would do, but both the monumental tombs from which the Kofun period takes its name and possibly the use of iron armour begin in the later 3rd century. The Tsubai Ohtsukayama tomb, which contained an iron cuirass with over 20 iron swords and 200 iron arrowheads, has been dated to the 3rd century (Farris 1998 p.71), but also to the 4th (Barnes 2000 p.82).
An alternative arrangement would be to extend the Yayoi list to include the early Kofun period, up to the adoption of cavalry, and incorporate the latter part of this list into the next one. However the arrangement suggested here at least corresponds to the conventional periodisation of Japanese history.
Aggression –
In this period the growing Yamato state expanded against its rivals in Japan, including both regional statelets organised comparably to Yamato and also covered by this list, and looser organised groups some of whom can now be identified as the Ag 0 Emishi. Japanese forces in Korea operated mostly as allies of Korean states, chiefly Paekche and one of the Kaya states known to the Japanese as Mimana; but there were probably some independent attacks on Korea, if not the great 4th-century invasion still supported by some traditionalist scholars. Some 5th-century emperors certainly claimed, in their embassies to China, to be military overlords of all southern Korea, though how far their power really extended is debatable. An aggression of 2 will be higher than that of historical enemies without completely ruling out a 4th-century invasion of Japan from Korea, which some revisionist or Korean nationalist scholars have proposed.
Scale –
While most internal Japanese struggles seem to have been smaller in scale, Imperial armies for expeditions to Korea could be 10-25,000 men strong; indeed armies of 60,000 are mentioned. In 602, one general “was allotted Shinto priests, local chiefs (kuni no miyatsuko), bureaucrats (tomo no miyatsuko), and 25,000 troops” (Farris 1992, p.23; Asakawa p.80; Aston Nihongi v.2 pp.125-6). The normal scale seems suitable.
Generals –
By the end of this period at least, the Yamato Japanese state was well enough organised to be allowed a “conventional” army with two sub-generals. Allies need to remain an option however for less organised non-Yamato regional polities (of which very little is known), for rebels against Yamato Imperial rule, and even for occasions of fierce rivalry between Yamato commanders – in Korea in the 5th century, two rival Japanese generals actually shot arrows at each other (Aston, Nihongi v.1 p.355, dated to 465 by Nihongi but possibly connected with one of the raids recorded under 459, 463 or 476 in Korean sources).
Leaders are frequently described fighting as archers, but spears also seem to have been acceptable high-status weapons in the early part of the period.
Uji archers –
Society and military power was based on aristocratic clans called uji. The chieftains’ relatives and toneri retainers served as warriors; there were also yatsuko, retainers of uncertain origin (sometimes translated “slaves” but apparently including convicts, prisoners of war, and youths of good family sent for training; see Asakawa). Two clans in particular, the Mononobe and the Ohtomo, whose chief the Ohtomo no muraji commanded the Imperial bodyguard, were military specialists and are frequently mentioned in the Imperial armies, but probably all clans had some military power.
The decoration of aristocratic tombs includes many portrayals of armour, quivers and large shields, and the haniwa tomb-guardian sculptures show the same equipment; swords and spears are much less prominent in art. Arrowheads are also plentiful in some tomb finds, such as the Ariyama kofun with 85 swords, 8 spearheads, and 1,542 arrowheads (Farris 1998, p.72). Shields were made of layers of leather, lacquered in bright patterns, and were about 1-1.5 metres tall by 60 cm wide (Farris 1992 p.19; Friday 2004 p.89); perhaps the larger examples are archers’ standing shields, the smaller hand-held by swordsmen or spearmen. The association of archery equipment with shields is also seen in the written sources, for instance Nihon Shoki for AD 587: "Hirafu no Muraji, therefore, took in his hand his bow and arrows and his leather shield" when serving as bodyguard to a minister; in 664 (admittedly early in the new system of the next list) grants of shields and bows and arrows were made to local chiefs (Aston, Nihongi v.2 pp.111 and 281-2). These archers seem to have been the main troop type, since one 5th-century Emperor referred in a letter to China to an army of “a million archers” (an embassy of AD 478 reported in Song shu 97; in Tsunoda and Goodrich p.26) and effective Japanese archery covered the retreat of their Paekche allies in 555 (Aston, Nihongi v.2 p.75). All this suggests that armoured archers shooting from behind a wall of shields, whether carried by shield-bearers or by the archers themselves, were the principal troop-type of the period. As time went on, some of these men adopted cavalry warfare; but some of the élite, and certainly their warrior-retainers, may have remained infantry archers until the end of the period. The published list makes all archers Irr Bw (I) (after AD 500, at least), but this is obviously inappropriate for well-equipped aristocrats and their warrior retainers.
See Shield walls later on for whether these archers should be Irr Bw (O), or upgraded to Bw (S) because of their shields.
Uji spearmen –
Spears modelled on Korean types are quite common in 4th and 5th century sites, becoming rarer thereafter presumably because those warriors rich enough to build kofun mounded tombs and leave weapons in them had by now become horsemen, and hence mounted archers (Farris 1998, p.73); so this need not necessarily mean a decline in numbers of rank-and-file spearmen. Farris suggests that spears at this period were 4 metres long (1992 p.19, no specific source suggested). In the light of later evidence this may be a maximum rather than a typical length, especially if they were carried with shields. “Long spears” are occasionally mentioned in Nihongi, but not in contexts that tell us anything about their battlefield employment (Aston, Nihongi v.2. p.191). Modern reconstructions of Japanese spearmen of this period in armour include a picture at http://horse.shrine.net/samurai/jinmu_e.html and a statue in Gifu Historical Museum (Turnbull 1987, p.16). Both show the spearmen with shields and tankoh plate cuirasses.
Despite the likelihood that many spearmen wore armour and (at least in the early part of the period) may have included high-status warriors, since there is no evidence for phalanx-like formations or any sort of mass formation training, they can’t really be classed any better than (I).
Swordsmen –
At Awohaka in 474, an Imperial Japanese army under Ushiro no Sukune attacked Asahi no Iratsuko, the local ruler of Ise to the east of Yamato. The Yamato troops were wary of the Ise leader’s reputation for archery and for two days the armies confronted each other with neither side attacking. Since Ushiro would not advance, his subordinate Me no Muraji attacked on his own initiative, protected by an armoured shield-bearer. The shield-bearer was wounded by an arrow, but continued to shield Me no Muraji who closed with Asahi and killed him with his sword (Aston Nihongi v.I pp.365-366). I assume that Me no Muraji should be treated as a sub-general, and assume that, though the account is couched in the language of heroic single combat, it is likely – or at least possible – that the named commanders were supported at least by retinues of personal followers, if not by their entire armies.
Be and other levies –
The lower classes were either directly dependent on the uji or organised into be (or tomo), communities of craftsmen, fishermen or farmers under chiefs (tomo no miyatsuko) directly dependent on the Court (Batten). The be were certainly called upon to provide troops directly to Imperial armies (Asakawa pp.69, 83). Whether commoners dependent on the uji were required to fight as infantry is less certain, but they may very well have done so: “One specialist has argued that a strongman’s troops comprised two groups: a peasant militia serving as footsoldiers, and elite mounted warriors belonging to the strongman’s family. It is reasonable to believe, as most Japanese scholars do, that royally designated regional strongmen fought as horsemen and commanded units of peasant foot soldiers...” (Farris 1992 p.25). Classifications are largely guesswork.
Emishi –
As early as 479, 500 “pacified” Emishi infantry formed part of an Imperial army intended for Korea, though they rebelled when Emperor Yuhriaku died and the invasion was called off (Aston, Nihongi v.2 p.371).
Boats –
Clearly enough sea-going shipping was available to transport substantial armies to Korea – note the "horse, foot and ships" sent in 465 (Aston, Nihongi v.2 p.355). Therefore, though not compulsory in this list, naval vessels should be available to Japanese allied contingents in Korean armies. Various boats shown in tomb-paintings or haniwa models (for instance Kidder 1959 pp.165-166, plate 102; Kidder 1964, p.129; Kidder 1977 p.45) have high stems and sterns, long steering-oars, rowlocks visible on the tops of the gunwales quite unlike the way the sweep-oars of later Japanese warships were used, and sails (one shows two masts). A reconstruction of a haniwa boat launched in 1975 “was 16.5 meters long and able to accommodate 30 people, with seven oars to a side” (Kidder 1977 p.45).
“Some scholars believe that Japanese boats were nothing more than scooped-out logs spliced together to make rafts (in some cases nails may have been used). Others point to clay models of boats and argue that these vessels were built entirely of planks by the year 600. The most widely accepted view, however, is that boats were made of several hollow logs tied together, with planks attached to the sides and top of the logs to enhance seaworthiness. A typical vessel measured 10 to 20 meters long, 1 to 2 meters wide, and 0.5 to 1 meter deep. Builders favoured cedar and camphor. Japanese boats usually contained a compartment for the crew and helmsman, and were driven by both oars and sails, but they were not safe for a voyage to the continent and rarely plied the ocean at night ... a picture from a tomb in southern Kyushu shows a red horse riding in a boat. An entry in The Chronicles of Japan lists one boat as carrying on average 25 men and 2 horses” (Farris 1992 p.23). The last reference is to the following passage: “Accordingly there is being sent” {to Korea, to assist Paekche, in AD 554} an auxiliary force to the number of 1000 men, 100 horses, and 40 ships” (Aston, Nihongi v.2 p.72).
To transport cavalry, Ships (Inferior) are necessary under the rules, but for the rest of the vessels Boats, probably Ordinary, seems to fit the evidence best. I did consider Boats (Superior), but there is no indication of more than one man per oar, nor, perhaps surprisingly, of archery from boats being significant in battle (the fact that Bts (S) can shoot whereas (O) cannot being one of the distinctions in DBM); while the built-up sides seem to amount only to a low-sided hull, rather than high sides that might provide protection in battle. There seems to be no sign of large flagships or command vessels, such as suggested for the 12th century in Turnbull (2003, Fighting Ships...).
See http://www.kippo.or.jp/KansaiWindowHtml/News/2000-e/20000607_NEWS.HTML, the pictures at http://www.hyogo-c.ed.jp/~maibun-bo/03remain/KENSITEI/13kensitei/hakaza/hakaza.htm which seem to be the wood carved with pictures of boats referred to in the first link, and bits of an actual boat at http://www.opas.gr.jp/neyagawa/sankakokugo/e/e23/e236/e240/e2407.htm.
Camp fortifications –
In 587 one leader had defended himself in a “rice-fort”, inaki; another usage of this term in a legendary context is explained as a fortification of piled-up rice-stalks, but in fact it may just mean a provincial granary, so could be camp TF or just a BUA (Aston, Nihongi v.1 p.172, v.2 pp.113-114).
Korean allies –
Japanese military involvement in Korea in this period is certain, but its form and significance has been hotly disputed. Fortunately the debate need not affect army list considerations very much. The oldest Japanese history, the Nihon Shoki or Nihongi written in the 7th century, claims that Japan conquered a foothold in Korea (Imna or Mimana, equated with one of the Kaya states) in either the 3rd century (taking the literal chronology) or the 4th century (the normally-accepted correction), installed a Japanese governor, and used this base to make most of Korea tributary. Some 5th-century Japanese rulers did indeed request and receive recognition from the Chinese Liu Song court as military overlords of the southern Korean states (see sources in Tsunoda and Goodrich); and armours of Japanese style from 5th-century Kaya, if too few to support the idea of a Japanese military occupation, do strongly suggest some form of military presence (Barnes 1994). But Nihon Shoki was written at a later date, and to boost the prestige of the Imperial dynasty; Japanese-Kaya relations may be better interpreted as alliance rather than conquest, while there is little evidence that real 5th-century Japanese influence in Korea extended beyond Kaya at all, whatever the Yamato court wished to claim. I suspect that genuinely close 5th-century relations may have been exaggerated, and perhaps projected backwards to the 4th century, by later writers.
But whether as overlords, vassals or allies, Japanese forces certainly fought in alliance with both Paekche and the Kaya states. The practicalities of ocean transport suggest that in most cases this would involve Japanese auxiliary contingents assisting larger Korean armies – which may be why the published list allows Kaya but not Paekche allies – but allied contingents can be allowed “both ways round” to cover circumstances where a Japanese expedition might find itself accompanied by only a detachment of its local allies. Nor can I really rule out the possibility that a Japanese force might have found itself fighting alongside Kaya and Paekche contingents simultaneously.
As a start date for allies I have used AD 366, the date generally ascribed to the events which Nihon Shoki describes as the Empress Jingu’s invasion of Korea (Bryant p.7 and Plate B) and some modern historians have interpreted as a Korean invasion of Japan (Hong). Kaya allies stop in 562, the date of Silla’s conquest of Kaya.
Cavalry –
The Yayoi Japanese may have known the horse, but if so apparently used it for food and perhaps as a beast of burden, not as a riding animal (Farris (1992) p.16, citing Hayashida Shigeyoshi, "Nihon zairai uma no genryu" ("Origins of the Japanese ridden horse") in Mori Kohichi et al, “Nihon kodai bunka no tankyu 9 Uma”, 1974); though Imamura (p.144) suggests that chemical analysis of the “Yayoi” horse bones indicates they were all later intrusions into Yayoi deposits. Riding, and cavalry, were introduced at some point during the period covered by this list. Like the earlier adoption of iron armour, the idea came from Korea. When, and in what circumstances, is debated.
Some archaeologists attribute the great influx of Korean culture at this period to invaders from the continent – the “Horseriders”, sometimes identified as refugees from the defeated Manchurian Puyo kingdom or as an invading Paekche army (Hong 1988). But while most supporters of a Korean or “Horserider” invasion put it in the 4th century, evidence for horse-riding itself does not appear in Japan until the 5th, and others argue that the archaeological evidence for Korean influence does not actually support any armed invasion. Defeats of Japanese armies allied to Paekche against Koguryo in the Korean wars around AD 400 are attributed to the weakness of their infantry in the face of Koguryo cavalry – “Japanese scholars are unanimous in the opinion that Yamato footsoldiers fled headlong from Korean mounted troops about AD 400, when the Japanese engaged the northern kingdom of Koguryo in southern Korea” (Farris 1992 pp.14-15) – and in response, it seems that cavalry begin to appear in the 5th century. Hesselink (1991) suggests that mounted archery was introduced by men like Achi no Omi, a man of Chinese descent from the former Taifang commandery in Korea, who came to Japan with a gift of horses from the king of Paekche, and then (c.AD 408) sponsored a large-scale immigration from Taifang founding the Aya clan, who were later famous as mounted archers. Horse-furniture starts to appear in Japanese tombs about the second quarter of the 5th century and becomes common in the 6th; in 456 and 460 Nihon Shoki mentions the Emperor hunting with a bow from horseback; and under 465 it gives what may be the first plausible reference to cavalry in the field, when a Japanese army assisting Paekche against Silla is described as "horse, foot and ships" and two of its rival commanders ride "with bridle-bits in line" and then shoot arrows (unfortunately, at each other) from horseback (Aston, Nihongi v.2 pp. 336, 341-2, 355). Chieftains and their retainers soon began to fight mostly as mounted archers, particularly in the Kanto provinces of eastern Honshu, where a strong tradition of horsemanship developed. It is difficult to put an exact date on the introduction of cavalry, but the Aya immigration in 408 is perhaps the earliest plausible date for the use of a few horsemen, with numbers increasing in the 6th century.
Cavalry seem to have been archers from the start. It is difficult to get an idea of what numbers or what proportion of a force they would form. The reference to “1000 men, 100 horses, and 40 ships” in 554 (Aston, Nihongi v.2 p.72) is the only hint I have found, suggesting 10% in a 6th-century shipborne expeditionary force. There were “more than 1,000 cavalry” in one division of Ohama’s army in the civil war of AD 672, after this list (Aston, Nihongi v.2 p.314), suggesting that absolute numbers available, and hence perhaps deployed at home, might be quite a bit higher.
Cavalry on armoured horses –
Korean-style horse-armour found in Japan includes not only a striking 5th-century iron chamfron (illustrated in Kidder 1964, 1977, and elsewhere) but also some pieces of lamellar bardings of similar date (“a large number of horse bardings” – Yang 2000 citing Onoyama 1959). These combine with the complete armour shown on some haniwa tomb-figurines – including lamellar leg-armour – to suggest that some early Japanese cavalry may have been equipped as Korean-style cataphracts; but the fashion may have been short-lived, perhaps because Japanese horses were too small to carry the weight of barding in action.
Armouring a horse-archer’s horse would normally make him Cavalry (Superior) in DBM. In DBMM the situation seems less clear. The definition requires Cv (S) to be “formed in solid bodies”; they may have lances as well as bows, or “they may often lack lances but be trained in defensive halted “shower shooting”, such as the Sassanids”; it seems highly unlikely that early Japanese cavalry qualified on either of these grounds. Knights (Inferior) seems an alternative valid under both rulesets, and perhaps the only one valid under DBMM, since they appear to be imitating the equipment of the Korean cataphracts classed as Knights (X), but (presumably) without the lances, and (probably) with underpowered horses. Just as Cv (O) with javelins become Kn (I) on armouring their horses, so should these Cv (O) with bows.
Yugei –
By 500 or so, the Imperial court could call upon two forces of guards. The yugei “quiver-bearers” were recruited from provincial clans in western Honshu. They served as archers under the command of the Ohtomo no muraji. They are generally considered to have been foot-archers, partly since they were raised from a region not as famous for its horsemanship as the eastern provinces of the Kanto, and partly because they are often identified with statues of haniwa warriors with cuirass, quiver and shield.
Farris (1992 p.27) says that the yugei were founded in the late 5th century, but notes that one Japanese scholar favours a 6th-century date and the Nihon Shoki puts it back to legendary times. As for numbers, the emon fu Gate Guards who were their successors numbered about 200, so two elements (including the C-in-c) is generous.
The other guards force, nominally also under the Ohtomo no muraji, was made up of the toneri personal retainers of various Imperial princes, who were mounted archers recruited from the Kanto. However there seems to be no reason to distinguish these Imperial toneri from other cavalry, who include the toneri retainers of other clans.
Oh-yumi –
Oh-yumi “crossbows” and stone-throwers are first mentioned in Japan in 618, when a delegation from the Korean state of Koguryo presented ten “crossbows” and catapults they had taken from the Chinese (Nihongi v.2 p.140). They do not seem to have been at all common during this period.
In a 1998 article, Stephen Turnbull suggested that there was a reference to oh-yumi in the Ruijuh sandai kyaku for AD 573 (1998, p.150); this appears to be an error, as the work in question doesn’t start until AD 701 and the passage relates to 753. (Prof. K F Friday, pers. comm.)
Eastern Company –
Soga no Emishi (or “Yemishi”), Prime Minister in 626-645, and his clan built fortified palaces with their own armouries and private forces of armed retainers, notably a group of 50 called the “Eastern Company” (Aston Nihongi v.2 p.189-190; Farris p.30). Suggestions that the Soga used Emishi guards refer to this unit – “Possibly the guard here spoken of consisted of Ainos” (Aston) – though whether they were in fact Emishi, or simply Japanese from the eastern provinces, is uncertain; the latter may be more likely. It may be the suggestion that the Emishi introduced the curved sword to the Japanese that prompts the classification of this bodyguard as Irr Bw (S) in the published list; unfortunately not only does Friday 2004 reject the idea of the Emishi introduction of the curved sword, even if correct it would apply to a much later date. We do not seem to have any description of the armament of these Easterners, though stockpiles of arrows are mentioned in the Soga palaces. (The Aya no Atahe, a pro-Soga clan leader, discarded sword and bow when they were overthrown, but he’s not necessarily connected with the Eastern Company: Aston v.2 pp.190, 193). They could perhaps be cavalry, since the eastern provinces had a reputation for horsemanship. However I see no great harm in allowing them to be Bw (S) – in any case, that may end up being the standard classification for an infantry C-in-c, depending how the grading of archers generally is resolved.
*******
III 7/C Ritsuryoh Japanese, 646 AD – 811 AD
{DBM:} Cold. WW, Rv, _H(S)_, Wd, O, E, RGo, Rd, BUA.
{DBMM:} Cool. S, Rv, _DH_, SH, Wd, O, E, SF, BF, Rd, BUA.
Ag 3.
Nominal list scale: One element = 250 foot or 200 horse (normal scale)
Commander-in-chief – Reg Cv (O) or with escort of emon fu guards – Reg Bw (S) 1
Sub-general – Reg Cv (O) 0-2
Emon fu guard archers – Reg Bw (S) 0-1
Mounted archers – Irr Cv (O) 4-8
Heishi militia archers – Reg Bw (I) {or Bw (O)?} 12-72
Heishi militia spearmen – Reg Ax (X) / Reg Pk (F) 1 per 4-8 heishi Bw
Replace heishi archers and spearmen with mixed bodies, all single-based Reg Bw (X) All/0
Heishi mounted archers – Reg Cv (I) 0-1 per 12 heishi Bw
Pacified Emishi archers – Irr Bw (I) or Irr Ps (O) 0-8
Upgrade Emishi to mounted archers – Irr LH (S) 0-2
Oh-yumi crossbow-artillery – Reg Art (O) 1 per 3-8 heishi foot
Replace oh-yumi with stone-throwers – Reg Art (I) 0-2
Boats – Up to 1/3 horse-transports Irr Shp (I) {Cv, Art}, the rest Irr Bts (S) {Any DBM infantry/DBMM foot} 0-6
Camp fortifications – TF @ 1 AP 0-12
Barricades to block roads – TF @ 2 AP 0-3
Only until 663:
Paekche allies – List: Paekche and Kaya Korean (Book 2)
Only from 646 to 795 AD:
Replace heishi with sakimori – Up to half Reg Cv (O), the rest Reg Bw (O) {or Bw (S)?} or Reg Bw (X) 0-12
Only until 701:
Downgrade Cv generals to irregular Any
Replace Heishi Bw with slaves and other retainers – Irr Bw (I) or Irr Bw (O) 0-12
Only from 702:
Downgrade Heishi Bw to Reg Bw (I)
(Only if they end up with a default grade of Reg Bw (O), obviously) Any
Only from 792 AD:
Kondei “strong fellows”, fushuh re-settled Emishi, and similar additional mounted archers – Irr Cv (O) 4-12
Rules considerations:
1. Emon fu must be in the command of an emon fu C-in-c.
2. Emon fu may not be used with sakimori or Korean allies.
3. Allied contingents drawn from this list for other armies may include up to 6 naval elements.
4. Sakimori may only be Bw (X) if heishi are.
Draft list notes:
This list represents the Chinese-style conscript armies of the ritsuryoh or “law-code” state from the Taika "Great Reform" edict until the end of the great Emishi wars. After the Soga were overthrown in a coup, the new régime established a centralised administration and national militia army on the Chinese model, though this was not complete until the Taiho code of 702. Armies were often over 20,000 and forces as large as 100,000 are reliably reported. A militia regiment or gundan was created in each province, and provided drafts for the jin battalions of expeditionary armies. Senior officers were local officials, junior officers commoners selected for “skill with the bow and horse”. The heishi militiamen mostly fought as infantry with bow, long sword and dagger, and were issued with armour on campaign. Each squad of five had one pavise, and these were formed into a shield-wall. Other heishi fought with 3-4 metre spears (apparently without shields) or as cavalry. The Bw (X) option allows for the possibility that archers and spearmen may have fought within the same squads, though there is no direct evidence for this. Two men from each 50-man platoon used oh-yumi stand-mounted crossbow-artillery, but whether they used one weapon between them or one each with assistance from others is unknown; so the proportion of oh-yumi to infantry is uncertain, but could have been very high. The oh-yumi was thought to be particularly effective against the Emishi. Heishi could also be selected for service as sakimori, stationed in the south-west to protect against Chinese and Korean invasion (they were archers with an unknown number being mounted); as chinpei in the north-east to fight the Emishi; or as guards in the capital. Chinpei and guards are treated the same as other heishi. In the 8th century abuse and corruption led to a decline in the standards of the militia and in 792 the regiments were abolished except in the frontier provinces; there they continued to fight the Emishi until the end of the great Emishi wars in 811 and beyond. Elsewhere new cavalry forces were raised to replace the militia, supplementing the cavalry who had always been recruited from the richer classes.
Justifications for list C:
Aggression –
This period saw large-scale and repeated attacks against the Ag 0 Emishi, now restricted to the northern region that became the two provinces of Mutsu (sometimes read Michinoku) and Dewa (or Ideha), and large-scale interventions in Korea, against Silla and Tang China. Nobody invaded Japan, though first the Tang and then Silla do seem to have been considered threats. If we consider actual wars rather than invasion scares, the aggression score should be at its highest.
Dates –
This list starts with the Taika "Great Reform" edict of 646, which began the creation of a Chinese-style centralised state and conscript army.
While the obvious end-date might seem to be 792, with the abolition of conscription in most provinces, or else 794 when the capital moved to Heian-kyo and the Heian era begins, militarily a more important date seems to be the end of the Emishi wars in the north-east in 811. The general abolition of conscription in 792 did not mean the end of the conscript regiments in field-armies since they were retained in precisely those provinces where they might have to do some actual fighting; and the large armies that are a diagnostic feature of this period continued to be fielded until the end of the Emishi wars.
Scale –
Some armies were quite large; 27,000 men were sent to Korea in 663, while the army raised in 788 and defeated at the Koromo river the next year consisted of 52,800 men, though this included porters (at least 12,400) as well as soldiers. Smaller forces were more typical, however: in 737, a force sent against the Emishi included 196 of the 1,000 eastern cavalry who had been sent north, 499 chinpei, 5,000 Mutsu heishi and 500 from Dewa, and 389 allied Emishi; in 740, Fujiwara no Hirotsugu’s Kyushu-based rebel army of 12-15,000 men was defeated by an Imperial expedition of 17,000. Even after the disbanding of the militia in most provinces, very large armies, including the surviving northern frontier militia regiments, were mobilised for the pacification of the Emishi until 811 – 100,000 men in 794 and 40,000 in 801.
Generals –
In the published list, generals are Irregular even at the peak of the regular organisation. This is probably wrong. Generals at this period are at the apex of a formal rank structure, commanding an organised, bureaucratically administered army and manoeuvring it to the signals of flags, drums and gongs in Chinese style. It is hard to see why they should not be regular.
Emon fu –
The Gate Guards (emon fu) consisted of 200 kadobe, who guarded the outer gates of the palace, and 30 mononobe whose role was to administer punishments. They were recruited from traditional military families of the capital and neighbouring provinces; though formally established in the 702 code they may have been the old yugei under a new name, so they may merit the same classification as in the previous list, and probably existed under one name or another from the start of the period. Whether they ever fought in a battle I don’t know, but they were at the very least theoretically available in the 672 civil war and other insurgencies; they would not be used much beyond the capital, hence the other restrictions. Their independent existence ended in 808, when they were merged with the Capital Guards, but I don’t think a line removing them for a mere three years is necessary.
The remaining guards units probably don’t need to be dealt with separately in a DBM/DBMM list. The 800 Capital Guards (eji fu) were provincial conscripts, who were to “practice mounted archery, sword-play, spear-fighting, and firing oh-yumi and catapults.” Revenues were set aside specially to reward those Capital Guards good at mounted archery. In other words, these Guards fought in the same style as their parent heishi units. The 800 Palace Guards (hyohe fu) guarded the inner gates of the palace, and the Emperor’s person. They were recruited from the official classes, sons and relatives of magistrates “strong and good at archery and horsemanship”, so they probably do not need to be distinguished from other cavalry.
Mounted archers –
Although some heishi militia were cavalry, most cavalry forces were raised outside the militia system from the richer classes, particularly from the eastern Kanto provinces. One division of Ohama’s army in 672 mustered “more than 1,000” cavalry. In 724 the court ordered 30,000 mounted archers to be trained in the east, though there is no indication this ambitious target was achieved. In 737 five eastern provinces provided 1,000 cavalry for the Emishi wars; one expeditionary force included 196 of those 1,000 cavalry, 499 chinpei, 5,000 Mutsu heishi and 500 from Dewa, and 389 allied Emishi – which may give an impression of the proportions of cavalry actually fielded. Again in 774 and 776 cavalry from the Kanto were raised to supplement the northern forces against Emishi rebellions. In 780, as the gundan system declined, “wealthy commoners skilled with bow and horse” were to be recruited instead of peasant conscripts.
Heishi –
A militia system was envisaged from the start of the Taika reform era, but details are scant until the Taiho codes of 702. By that law, all men from 21 to 60, except slaves, court nobles, and the unfit, were liable to military service. These conscript soldiers, heishi, were organised into one or more regiments, gundan, per province. The regulations laid down three sizes of regiment: small, of 500 men or fewer; medium, between 600 and 1,000; and large, of 1,000 men. They were only administrative, not tactical, units. Regiments were divided into ryo companies of 100 men and tai platoons of 50, mostly of infantry (hotai) but some cavalry (kitai). (Both Farris 1992 and Friday 1992 describe this organisation. Friday uses “company” for the tai, but I’ve used Farris’ translation of “platoon” to keep “company” for the next level up, the ryo.) These “platoons” seem to have been the basic tactical unit. They were in turn divided into ten-man ka “campfires”, who shared a tent on campaign, and five-man go squads. Senior officers were recruited from the official classes, junior officers were commoners recruited for skill with bow and horse, suggesting that all officers may have been mounted archers. There were also one or two clerks, shuchoh, per regiment. For expeditionary armies one to three gun “corps” would be raised, typically each around 3-12,000 men. Each gun was commanded by a shohgun general; if three gun were mobilised, overall command was given to a taishohgun generalissimo. The gun were divided into battalions, jin, which may have been made up of detachments, each of one or more tai platoons, from more than one provincial gundan (Friday 1992 p.27).
Conscript soldiers were not full-timers, but were called up for watch duty (ban), for expeditionary armies, or for specialised units – the Capital Guards and the sakimori and chinpei frontier garrisons. Ban was service in the soldiers’ home province: they drilled and trained, guarded and maintained government buildings including the provincial armouries, and performed police duties. Conscripts were trained to “swing swords, stab with spears, shoot oh-yumi (heavy crossbows) and catapult stones” and also in “the rules of line formation”; the drums, gongs and flags used in Chinese-style manoeuvres are prominent in, for example, descriptions of the fighting in 672.
Each man was to provide a bow with bowcase and bowstrings, fifty war-arrows and a quiver, a long sword and a short sword (or dagger?). Militiamen were not required to supply armour, but it was at times issued to them; armour was stored in provincial armouries and large quantities were issued for the north-eastern campaigns. They shot from behind large shields, one shield to each five-man go squad. Modern authors suggest these were the rectangular wooden standing shields shown in later samurai-period art, but I have not actually seen a contemporary source cited for the size, shape or material of these shields. The Taiho code prescribed that a 50-man tai “platoon” should deploy in two lines, each of five five-man go squads (Friday 1992 p.27). Each infantry squad fought around a single standing shield, perhaps in a five-deep file of four archers led by the shield-bearer or perhaps in a looser group. I don’t know how far apart the two lines were to deploy, so this formation could be represented either by two lines of Bows each one element deep, or by two-deep Bows – the latter would shoot more effectively under DBM at least (the DBMM archery rules seem still to be under discussion as I write).
The conscript system may have worked quite well at first, while the threat of invasion from Tang China or Silla Korea seemed real. But increasing abuses were reported in the 8th century. Officers and provincial officials used conscript soldiers for forced labour: “the conscripts all throw away their bows and arrows and instead take up spades and hoes”. The recruits who were coming forward by the late 8th century were neither physically impressive nor financially able to provide their equipment and supplies: “Their bodies are naked and their heads are bald. They know how to use only sickles and hatchets; how can their weak backs and tired shoulders bear to pull a bowstring? They come without provisions and, when we ask about this, they flee.” I therefore suggest an option to downgrade them in the 8th century – though if they stay at the current Bw (I), that won’t be applicable. Making them Hd would be too harsh, as massed archery remained their main tactic.
Because of these abuses, and because there was no military threat in most of the country, the provincial militia regiments were simply unnecessary, and in 792 they were abolished in order to reduce the burdens on the people, except in the “frontier” provinces of Mutsu, Dewa, and Sado island in the north of the country, and northern Kyushu (the jurisdiction of the dazaifu, the command post of an important regional Yamato official – see Batten 2004) in the south. Note that the regiments were retained in those provinces where they might be called upon to do some actual fighting, suggesting that despite any weaknesses they were still seen as militarily useful. The northern regiments no doubt contributed to the large size of the armies mobilised for the pacification of the Emishi until 811 – 100,000 men in 794 and 40,000 in 801. After 811 the size of the gundan was reduced, in Michinoku (Mutsu) to 2,000, though restored to 6,000 in 815, and they still provided troops against the Emishi rebellion of 878.
See Shield walls later on for whether these archers should be Reg Bw (I), as in the published list, or upgraded to Bw (O) because of their shields.
Heishi spearmen –
The militia trained with spears as well as bows and swords, but where the spears fit into the tactical system – perhaps carried by some men in each squad, perhaps by squads or companies entirely of spearmen – I do not know. Hoko spears preserved in the Shohsohin collection (established in 756 in Nara) “feature heads averaging around 36 cm, mounted onto 4-meter hafts” (Friday p.85). This (and the Heian spears mentioned later, almost as long) may be the “long spear” mentioned in one incident in the 672 civil war, when one brave warrior stormed an enemy-held bridge single-handed, discarding his long spear and attacking in "double armour" with drawn sword (Aston Nihongi v.2 p.313). Note that no shield is mentioned in this passage, and none seem to be mentioned among the militia equipment other than the archers’ standing shields.
Ax (X)/Pk (F) seems on the face of it the obvious classification for men with such long spears but apparently without shields. The main problem is that those troop-types can fight in great depth, whereas it is unlikely that these spearmen actually did so. It is also not absolutely certain that they were shieldless, since spearmen in both the preceding and following periods seem to have carried shields. If they turn out to have been Bw (X) – one spearman in a five-man squad with four archers, or three archers and a shield-bearer – or even Pk (X), spearmen holding their weapons in both hands with separate shield-bearers, I shan’t be very surprised. So far, however, I don’t know of any evidence for either of those classifications; but the Bw (X) possibility is allowed as an option here and in the next list.
Heishi mounted archers –
The government tried to raise as many cavalry as it could, and some militia platoons were cavalry, selected from those recruits “skilled with bow and horse”, provided with horses raised on government pastures but expected to bear the cost of looking after them.
Emishi –
“Pacified” Emishi were used on the northern frontier in campaigns against their unsubdued relatives; 389 Emishi are recorded in an army of about 6,600 in 737 (Friday 1997). Given their 5th-century use in an army intended for Korea, there seems no reason why they should not have been called up for service elsewhere.
Oh-yumi –
Two “strong men” in every tai of 50 were to be experts with the oh-yumi (heavy crossbow), and they were a vital part of the military system, being thought particularly effective against the Emishi: “even though there are bandits coming from 10,000 directions, they cannot resist the flying arrows of even one oh-yumi. This weapon is the best for overwhelming the barbarians”. Friday (1992 and 2004) has argued convincingly that this weapon was a stand-mounted “artillery” crossbow, not a hand-held weapon as Farris (1992) suggested. There seems to be no information on how the weapons were organised in practice: did the two oh-yumi-experts operate one weapon each, or one between them? And since a ballista-sized piece would be expected to require more than one crewman, did they have help – perhaps from the rest of the five-man go squad or even ka “campfire” they belonged to?
At one extreme, a tai might have contained 48 men fighting as infantry and one oh-yumi operated by these two “strong men”. Assuming an element is 250 men or 30 light bolt-shooters, this would require one artillery element to six elements of foot. At the other extreme, a tai might have fielded two oh-yumi served by a squad of five men each, leaving only 40 fighting as infantry. In that case, the 30 engines represented by one element would be accompanied by 600 infantry, or less than three elements. (The minimum ratio allowed here, one per eight infantry elements, assumes that whatever the number of oh-yumi required, numbers fielded may not always have matched it.)
Stone-throwing engines are occasionally mentioned, so should be an alternative to some of the arrow-shooters. Some may, like those presented by Koguryo in 618, have been Chinese-style pao rope-pull engines; but other sources refer to ishiyumi, “stone bows”, and some accounts imply that oh-yumi could throw stones (Turnbull 1998, citing references that mention ishiyumi being “bent” – suggesting a bow rather than a pao - and shooting stones).
Boats –
Naval forces could be considerable: 180 ships on an expedition against the Emishi in 658, 200 on another in 660 (Aston, Nihongi v.2 pp.252, 263) and perhaps as many as 400 ships in the expedition to relieve the Paekche city of Churyu in 663 (Farris 1992 p.39, Batten p.214).
“Nothing is known of the precise type or design of ships that took Japanese troops to Korea and fought at the battle of the Paekch’on River during the 7th century...” (Turnbull Fighting Ships..., p.22). However since an estimated 400 or more Japanese ships were defeated by 170 Tang Chinese warships (Aston, Nihongi v.2 p.280) they were probably a good deal smaller. Therefore I have stuck to the same classifications as in the previous list.
Fortifications –
Camp fortifications are mentioned in the civil war of 672, as is the blocking of roads with barricades made from the planks of torn-up bridges (both Aston, Nihongi v.2 p.313).
Paekche allies –
Included until the fall of Paekche to Tang and Silla forces.
Irregular generals, slave archers –
As said above, we only have a full account of the militia organisation in 702, and Farris (1992, pp.33-47), argues that the reform was incomplete until then – “The Civil War of 672 reflects patterns of the old military technology and organization more than the new” (p.43). Command in 672 was exercised by aristocratic military families and local chieftains, not by a regular command structure, though the troops had been recruited by provincial governors holders of a post newly instituted under the ritsuryoh system. Another aspect is the participation in the fighting of infantry who do not seem to have been raised through the militia system, such as a group of five slaves of the Temple of Ohowidera who distinguished themselves serving as archers – “Tokumaro and his companions, forming an advanced guard, went forward and shot their arrows, so that Kujira’s army was unable to advance” (Aston, Nihongi v.2 p.317) – slaves were not liable to militia service at all. However, I suspect such non-standard troops were not numerous, and the embryonic militia provided most infantry combatants.
Sakimori –
The sakimori, “border guards”, garrisoned Kyushu and the smaller islands of Iki and Tsushima. One passage of Nihon Shoki says these guards were established by the Taika edict of 646 (Aston Nihongi v.2 p.206 has “Let barriers, outposts, guards, both special and ordinary, be provided…”; Farris 1992 p.37 renders this “Let… Border Guards (sakimori) … be provided”). Another passage in the same work claims that they were set up in 664, when there was a fear of Chinese invasion, and this date is more generally accepted (Nihongi v.2 p.283; this passage, unlike the former, does specifically mention the areas affected – Kyushu, Iki and Tsushima – and see Batten p.215 for the word sakimori in this passage, where Aston just has “guards”). Sakimori were recruited from the gundan regiments of the eastern provinces, the bandoh or Kanto, which had a tradition of mounted archery; poems written by sakimori mention horses and bows, though it is not clear how many of them were horsemen and how many were foot-soldiers. That they were used as farm-workers as well as soldiers suggests their status was not all that high, so perhaps – as with the ordinary provincial heishi regiments – only the officers were mounted (Farris (1992) pp.54-55: “most were archers, some mounted”; Friday (1992) p.24: “On the frontier, sakimori stood duty nine days on and one day off. When not actually fighting or training, they were put to work cultivating paddies or dry fields assigned to them. The tools and oxen necessary to work the lands were provided by the government, with the harvests regarded as rations issued rather than as the soldiers’ private property”). In 737, there were 2,300 of them. In the mid-8th century, local men began to replace eastern recruits: transfer of easterners was suspended by court decrees in 730 and 737, though they returned in 755-757 and perhaps again later, since abolition was necessary in 795.
It may be worth considering whether we need to include sakimori explicitly, or if they could just be treated the same as the heishi – particularly if these latter are to be Bw (O).
Chinpei –
These are included as Bw (O) in the published list, but I have not made separate provision for them here. The chinpei “pacification soldiers” (references to chimpei “fortress soldiers” seem to mean the same force) served as garrisons in the stockaded forts on the Emishi frontiers in the north-east, where they are first mentioned in 724 in Mutsu. At least some chinpei also came from the bandoh provinces, and again these were replaced by local recruits later on. In Mutsu there were from 500 to 3,800 chinpei at different dates, the numbers rising in times of war and rapidly reduced in peacetime to save the cost of supplying them (Suzuta). The published list classes them as Bw (O) in contrast to the heishi Bw (I), on the assumption that their longer continuous service would make them more experienced troops. However the chinpei were abolished in 815 because of their poor morale – they were “like women” when compared to the men of the local gundan – which argues against their receiving any special treatment (Suzuta). It seems reasonable, therefore, just to lump them in with the heishi.
Kondei and additional cavalry –
Numbers of cavalry increased in the latter part of this period. In 792, as general conscription was abolished, new troops called kondei, “strong lads” (Farris favours “Strong fellows”, Friday “Stalwart youth”, Suzuta “Sound boys”), were raised from the families of district magistrates to replace the conscripts as guards of provincial armouries and other government buildings (though peasants were conscripted if enough élite recruits were not found; Friday 1992 p.54). The 792 edict called for a total of 3,165 kondei, from 20 to 200 in each province – most provinces providing 30-60. The frontier provinces – Mutsu and Dewa, Kyushu – were excluded from the system. A later ordinance, in 905, lists 3,864 kondei and now Mutsu and Dewa were included. The kondei were intended for local defence against brigands rather than as an army, and in 866, in the face of a Korean invasion scare, were thought inadequate: “The kondei raised by the provinces have lacked ability; they are emptily called the ‘defence of teeth and claws’, but are no different from a guard of grasshoppers.” A not dissimilar force, the 1,740 senshi “select warriors”, were raised in Kyushu: again mounted archers, but recruited from huntsmen rather than the official classes. Transplanted Emishi prisoners-of-war (fushuh) were another source of cavalry: up to 10,000 were settled in the Kanto provinces, partly for internal policing duties, and they served as far away as Kyushu.
An earlier edict of 733 had also called for the raising of kondei; it was rescinded after a plague in 738 (Farris 1992 p.108). I have not made provision in the list for this earlier kondei force, since it was small and recruited at least partly from heishi, so can be treated as no increase on the existing cavalry (“300 heishi shall be made kondei” – the edict is quoted in Friday 1992 p.55). The interesting thing is that the kondei of 733 were allocated two grooms each, and Farris suggests “When battle lines were drawn, these grooms probably acted as foot soldiers. Since there are no examples of Strong Fellows in combat, it is difficult to determine whether they used mass or individual tactics, but the command structure in which each equestrian trooper led 2 foot soldiers into war is identical to that used by the twelfth-century samurai” (Farris 1992 p.109). However, I do not propose to allow any supporting infantry in this list: we don’t know that the 792 kondei had the same two grooms apiece, we don’t actually know that the 792 kondei fought much, if at all, in battle, and if the grooms did fight we do not, as Farris admits, know how they fought – quite possibly in the same way as other infantry.
*******
III 7/D Early Heian Japanese, 811 AD – 1020 AD
{DBM:} Cold. WW, Rv, _H(S)_, Wd, O, E, RGo, Rd, BUA.
{DBMM:} Cool. S, Rv, _DH_, SH, Wd, O, E, SF, BF, Rd, BUA.
Ag 1 to AD 900, 0 from AD 901.
Nominal list scale: One element = 125 foot or 100 horse (half scale)
Commander-in-chief – Irr Cv (O) 1
Allied general – Irr Cv (O) 0-2
Replace allied general with sub-general 0-1
Kondei and other tsuwamono armoured horse-archers – Irr Cv (O) 8-20
Retainer or other foot-archers with standing shields – Irr Bw (I) {or Bw (O)?} 24-80
Nimbei sei peasant conscripts or banrui “allied” infantry – Irr Hd (O) 2-8
Boats, Irr Bts (O) or (S) {Any DBM infantry/DBMM foot, including Cv dismounted as Bw (O)} 0-8
Camp fortifications – TF @ 1 AP 0-12
Until AD 900 only:
Replace Irr Bw (I) with northern provincial heishi – Reg Bw (I) {or Bw (O)?} *12-60
Northern provincial heishi spearmen – Reg Ax (X) / Reg Pk (F) 1 per 4-10 heishi Bw
Replace heishi archers and spearmen with mixed bodies, all single-based Reg Bw (X) All/0
Kenshi archers or crossbowmen – Reg Bw (O) *0-16
Pacified Emishi archers – Irr Bw (I) or Irr Ps (O) 0-8
Upgrade Emishi to mounted archers – Irr LH (S) 0-4
Artillery, up to half stone-throwers Reg Art (I), the rest oh-yumi crossbow-artillery – Reg Art (O) 1-6
From AD 901 only:
Artillery, up to half stone-throwers Reg Art (I), the rest oh-yumi crossbow-artillery – Reg Art (O) 0-2
Rules considerations:
1. Items marked * apply if any heishi are used.
Draft list notes:
This list covers Japanese armies from the end of the great Emishi wars until the last victory won by shield-bearing infantry, the defeat of a landing of Jurchen pirates in 1019. Armies were much smaller than in the preceding ritsuryoh period, in the low thousands or even mere hundreds. Although the provincial heishi militia had been abolished in most of the country in 792, they remained in existence in the northern provinces of Michinoku (Mutsu) and Ideha (Dewa), where they served at least as late as the Emishi rising of 878, and were even increased in numbers in 815. The list assumes that they withered away about 900. Oh-yumi artillery remained an essential part of 9th-century armies but were rare in the 10th century because it became increasingly difficult to find skilled men to maintain them. Armies were led by powerful local warlords fighting either for the Imperial court or on their own behalf, and based round relatives and retainers fighting as mounted archers. Most of the troops were still infantry archers, either retainers or conscripts, fighting as archers behind shield-walls as in the earlier period, but less well-trained. Other infantry, conscripted peasants called nimbei sei in the 9th century and banrui “allies” in the 10th, included men with spears and shields but were very unreliable.
Justifications for list D:
Dates –
For the start date, see the last list. The end date is somewhat arbitrary; it must be after the end of Masakado’s rebellion in 940, in which manoeuvring infantry shield-walls were still crucial, but before the first “classic samurai” conflict. When do we start “samurai” warfare? Perhaps the first real “samurai” war of which any detailed account is available might be the rebellion of Taira no Tadatsune in 1027-1031, or even the Former Nine Years’ War of 1055-1062. Shield-bearing infantry were still responsible for a defeat of Jurchen sea-raiders in 1019 (Farris 1992), so perhaps an end-date for this phase as late as 1020/1025 would look plausible.
I also want to exclude warrior-monks from this list; though Turnbull’s title on that subject begins in 949 with “{t}he first major incident involving violence by monks against monks”, that was a small skirmish in which one side numbered only 56, and it is not certain even that weapons were used. In 981 we see a more serious clash involving bands of 2-300; but Turnbull first uses the word “army” only in connection with the attack on Miidera in 1074. I suggest that taking this list up to the 1020s (or even later!) would not necessitate catering for warrior-monk armies, or even warrior-monk elements.
Alternative arrangements would be to incorporate this whole list as an early sub-period of the Early Samurai list; or to split it at 878 or 900, with the earlier part going into the Ritsuryoh list and the later into the Early Samurai. A split at 900 would have the advantage of keeping the existing end-date for Pre-Samurai and start-date for the Early Samurai list. This would need a change in the notional scale and the aggression score for the Ritsuryoh list at 811. At the moment I prefer the arrangement proposed here, though. If the 10th century were to be included in the Early Samurai list, it would need much greater differentiation from later armies than in the published list, to bring out the greater importance of shield-bearing infantry.
Aggression –
Japan was now back to military isolation. Occasional operations against the Ag 0 Emishi, regarded as suppression of rebellions after the Emishi were “pacified” in 811, continued; the last major campaigns were against the rebellion of 878 (which is possibly where the Emishi list should most sensibly end; it’s prolongation to 1100 was to cater for important ethnically-Emishi clans in the north such as the Abe and Kiyowara, who seem however to have been fully assimilated to the samurai class in their fighting style in the 11th century, and perhaps sooner); so I suggest changing aggression from 1 to 0 after that war. I’ve changed at 900 rather than 878 to match the other changes in the list, and prevent a possibly spurious “transition period” in 878-900 with regulars but low aggression. Invasion scares from Silla Korea (Ag 0) bore fruit only in a Silla attack on Tsushima island in 894, and there were occasional sea-raids from tribal Jurchen.
Scale –
Armies were generally now quite small; if anything this is the single main feature distinguishing this period from the previous one. Masakado never fielded more than 8,000 men (“He led an army of 5,000 soldiers”, Shohmonki p.123; “his regular army of 8,000 men had failed to assemble”, Shohmonki p.129), frequently only a few hundred – “just over one hundred mounted soldiers” with an unspecified number of foot on one occasion (Shohmonki p.82). The main battles of the 878 Emishi revolt featured 600 Japanese against 1,000 Emishi, or 560 against 300 Emishi, while it proved difficult to get Michinoku to send 2,500 men to help Ideha provincial forces (Suzuta, under “Battle of Yakeyama” and “Akita Camp”). “Armies” of 5-600 a side are mentioned in a 10th-century incident (Konjaku Monogatari XXV.III; Wilson p.197).
Generals –
Armies still manoeuvred en masse, using drums and gongs to signal advance and retreat in Chinese style (for instance Shohmonki pp.74-75). However I assume that with no indication of systematic training or drill, this is not enough to earn regular classification for generals or troops. In the 9th century, the military forces of a province were recruited and led by officials called ohryohshi, but these were ad hoc appointments of local officials or locally powerful “strongmen”, not career officers. The Imperial armies that fought against Masakado and Sumitomo in the 10th century are described as “troops raised through the provincial headquarters”, shokoku no heishi, but in effect were composed (just like the rebels) of the privately raised retainers of locally powerful military leaders serving under court-issued warrants and given titles like ohryohshi or tsuibushi. One of the few indications that we have of battlefield organisation in this late period is that a 4,000-strong force under Fujiwara no Hidesato in 940 was in three divisions, which seems to have been thought unusually sophisticated for the period.
Tsuwamono –
In the 9th century provincial governors developed private bands of armed retainers to enforce (or abuse) their authority. In practice these seem to have been little different from the bandits except for the powerful patronage they enjoyed. Increasingly, provincial officials “did not like reading books, but studied the horse and the bow”. Similar powerful local figures, including members of the Minamoto and the Taira lineages (cadet branches of the Imperial house), became warrior clients of various court nobles, becoming the class known as the “warriors of the capital”, miyako no musha, or as samurai – the first usage of the word, derived from saburau, “to serve”. By the 10th century, powerful local gentry lived in fortified residences, eisho, and led forces called ikusa or “war-bands”; their members were called tsuwamono; literally “warriors”, these were mounted archers recruited from the war-leaders’ relatives, innen, and retainers, juhrui.
(The kondei still existed into the tenth century, but probably could not be distinguished from other horse-archers.)
10th-century cavalry would engage in archery duels with their counterparts between the lines, galloping past each other to shoot at point-blank range and reining in to return for another pass, or would charge swiftly when the enemy formations fell into disorder: in his final battle, Masakado launched a cavalry charge only when a sudden wind blew down the shield-walls. The horsemen are now often described as an élite, “each a match for a thousand of the enemy”. They were still primarily archers, swords being rarely used from horseback.
Retainers supporting cavalry –
These would, if included, be Ax (I) double-based with the cavalry (under DBMM, as suggested for the Early Samurai army; under DBM, they’d have to be supporting psiloi). But I haven’t included any.
Turnbull suggests that cavalry were already supported by armed attendants on foot operating closely with the horsemen, as in 12th-century warfare. In Samurai Warfare (1996, p.20) he says, in the context of Masakado’s infantry, “Many of the warriors who fought on foot would not have been simple peasants, but would have been acting in the capacity of warriors’ attendants (genin or shojuh), who had particular responsibilities towards their masters ... In painted scrolls the attendant usually appears in a simpler form of armour than that of the samurai, and he is often bare-legged. He may wear an eboshi (cap) rather than a helmet, and his weapon is frequently the curved-bladed naginata. Even though their primary duties were those of servants to the individual samurai, they did take part in the fighting, as painted scrolls such as the Heiji Monogatari Emaki attest. Such figures as are available suggest an equal number of attendants to samurai. Thus the Konjaku Monogatari describes a force of 70 mounted to 30 foot, while two other sources mention bands of 15 or 16 horsemen accompanied by 20 or more infantry, and seven or eight horsemen with ten or more foot.”
But it isn’t as clear as this account suggests. The painted scrolls mentioned are 12th-century or later, so not good evidence for the 10th. The naginata, indeed, is not definitely attested until 1146. The Konjaku Monogatari passage – “As to his war-band, there were more than seventy mounted men and thirty on foot” (Konjaku Monogatari XXV.VI; Wilson p.207) – is set in the late 10th century, but the numbers suggest that horsemen did not yet all have a footsoldier-attendant each, and it gives no clue as to how these footmen fought; the account of the subsequent battle mentions only mounted archery. Among the ambushed enemy of this force, “others seized their shields and tried to fight” (p.208) which suggests, if anything, that the expected rôle of foot-soldiers was still to fight in the shield-wall. In Shohmonki and Konjaku Monogatari I can’t see any trace of foot-soldiers directly attendant on horsemen or mixing into horsed melées; the only possible exception is when a mounted general retreats “with his shield in front of him” – perhaps carried by an attendant on foot (Shohmonki p.127).
So I am not at all convinced of fighting footsoldier-attendants in the 10th century. Is it worth including them as an option, though?
Foot-archers –
In Shohmonki we hear of shield-walls being formed for battle, troops advancing and withdrawing with their shields slung on their backs “in the proper manner”, or in retreat with shields dragged behind them, routers dropping their shields as they flee, and defeated troops repairing their shields ready for the next fight (Shohmonki pp.79, 81, 82, 87, 90, 95, 130).
The fate of the shield-wall of archers often decided the battle. At Tsuka in 935, when Masakado’s opponents formed a shield-wall, he sent infantry against them and thus won the battle, killing enemy soldiers and their horses with arrows (Shohmonki p.82, Varley p.15; that the infantry attack involves archery is not evident in Rabinovitch’s translation, but is brought out in Varley’s). At his final battle in 940, Masakado launched the cavalry charge that almost won the battle only when a gust of wind blew the opposing shield-walls down, the direction of the wind meaning that the enemy shields fell onto their soldiers, disordering them (Shohmonki p.130, Farris (1992) p.147). In 1019 a force of shield-bearing foot-soldiers defeated a landing of 3,000 seaborne Jurchen raiders in Kyushu (Farris 1992 p.191).
One account of a 10th-century battle in Konjaku Monogatari suggests that the exchange of archery began when the sides were about 110 metres/144 DBM paces apart, and the archers would then normally move their shield-walls forward to closer ranges (Konjaku Monogatari XXV.III; Wilson pp.197-198):
“Each side had about five or six hundred men, and they set up their shields in lines about a hundred and twenty yards apart ... both sides began shooting arrows at the arranged signal ... After that each side moved their shields closer, but just as they were about to shoot at each other at close range, word was passed from Yoshifumi’s side to that of Mitsuru, ‘There is no fun in today’s battle if each of us makes his war-band engage with arrows. You and I alone should try to test each other’s skill...’.”
Some of these foot-archers will have been warrior-retainers of the leaders, of lower status than the horsemen. One incident in Shohmonki mentions a young servant of peasant origins who was serving as a porter but also doing guard duty, suggesting that such lower-status retainers did have a military rôle as foot-soldiers. In return for betraying his master he was offered a post as a mounted soldier in an enemy leader’s retinue. Clearly this was seen as a promotion from his current rôle, but it may indicate that such men could reasonably be expected to be competent archers already. Other foot-archers may have been conscripts of some sort, though most of these will be included in the Hordes.
Although generals tried to manoeuvre such men using drums, gongs and flags, there is no evidence that they trained en masse and they were prone to flee easily, so cannot be treated as regulars any more.
See Shield walls later on for whether these archers should be Irr Bw (I), or upgraded to Bw (O) because of their shields.
Hordes –
In 9th-century armies the gentry and their retainers were sometimes supplemented by peasant infantry, nimbei sei or ninpu, still conscripted but in smaller numbers and on a less systematic basis than the old provincial militias, simply as an extension of their forced labour obligations (Farris 1992 p.116 with p.414 n.117; Friday 1992 pp.125-8, mentioning 1,000 in one incident in 883). While they may have been armed from provincial armouries, they would not have received much if any training.
Banrui were infantry raised in the 10th century from the better-off peasants and others who were not direct dependents of the war-leaders:
“... and allies (banrui)... Unlike dependents, an ally was relatively free and only partially reliant on the leaders’ economic and social functions... Minor local strongmen from the families of district magistrates were undoubtedly also included among the allies. Allies provided the bulk of all Kanto fighting forces; they were foot soldiers who carried spears, shields, and other weapons into battle” (Farris 1992 pp.150-151). “In the view of Hayashi Rokuroh, the banrui were a mobile, independent body of influential peasants that included the dogoh, or powerful landowners residing in the province. In Hayashi’s view, they rallied in support of a specific cause or event, but had no enduring relationship with a specific leader” Rabinovitch p.75). Banrui troops were unreliable: “just before the battle, his allies (banrui) scattered like divining-sticks” (Farris’ translation, 1992 p.137; Rabinovitch p.88 has “...his soldiers fled the scene soon after combat began, scattering in all directions like spilt counting-sticks”.) These should probably therefore be treated as Hordes.
Masakado collected 370 spears and shields, probably for issue to banrui, in 935 (translated “halberds” in Rabinovitch p.87; but Farris p.137 has “spears”, and Varley p.14 specifically identifies the weapons as hoko). The association of the two weapons may be a rare confirmation that spearmen were expected to carry shields, and one account of items captured after Masakado’s defeat indeed mentions tedate, “hand shields”; this may mean hand-held shields, but other versions just have heidate, “flat shields”, which could simply mean the usual standing shield (Friday 2004 pp.185-6, n.78).
Heian spears varied in length, from short examples that could easily be used with shields to longer shafts reminiscent of earlier spears. Several examples, perhaps 9th-century, survive in the Kasuga shrine at Nara (Kasuga 1990). One type, with a broad iron head, ranges from 1.27 to 2.73 metres long; another with a small, narrow head can be from 1.73 up to 3.4 metres long. A third style has an asymmetric wooden head, so may be a practice or ceremonial weapon.
Naval –
The Shikoku pirate leader Fujiwara Sumitomo is said to have led 1,000 ships in 936 and 400 in 940, and to have lost 800 in a battle in 941 (Farris 1992 pp.143, 148, 149). Nothing is known of their size or type, but if the numbers are anything like correct, most will have been fairly small. However the “ships” of even 12th-century warfare were mostly open boats, some with a single mast, raised platforms at bow and stern and/or shields as defences, while some were lighter still (Turnbull Fighting Ships..., especially pp.23, 41-2). I have come across no reference to horse-transports at this period.
Camp fortifications –
These were probably still in occasional use, but so far I haven’t found any definite reference.
Northern heishi –
Despite the increased emphasis on cavalry, infantry and oh-yumi remained important parts of 9th-century armies. The provincial gundan were still retained in the “frontier” provinces in the early Heian period. After 811 the size of the gundan was reduced, in Michinoku/Mutsu to 2,000, though restored to 6,000 in 815. The northern gundan still provided troops against the Emishi rebellion of 878, but the armies fielded were now much smaller than before 811: it proved difficult to get Michinoku to send 2,500 men to help Ideha/Dewa provincial forces, and most of the fighting was done by forces numbering in the hundreds (see Suzuta).
Getting rid of this option at 900 is fairly arbitrary, because I simply don’t know what happened to the northern provinces’ regiments after 878. I would suspect they just withered away.
Kenshi –
The chinpei also served in the north until the changes of 815, when they were disbanded because of their poor morale – they were “like women” when compared to the men of the local gundan. I don’t think them worth mentioning for their brief and ineffectual existence in this period.
They were replaced by 2,000 kenshi “strong warriors”, professionals recruited from veteran soldiers who garrisoned the main forts (Farris 1992 p.112). It may have been these men who used the 100 shudo hand-held crossbows and 29 oh-yumi heavy crossbows taken from one fortress in 878 (Farris 1992 p.115; Friday 2004 p.74).
Emishi –
As many as 1,000 loyal Emishi troops took a “rebel” Emishi village in 811. After 878 I have found no specific reference to them, and I would assume they were gradually absorbed into the Japanese system. Transplanted Emishi settled elsewhere in Japan are treated as indistinguishable from Japanese tsuwamono cavalry.
Artillery –
Oh-yumi remained an essential part of 9th-century armies. Provincial governors kept oh-yumi in the armouries, though often in poor condition, and the Court sent oh-yumi specialists to the provincial administrations – in the 860s there was a drive to appoint oh-yumi experts in coastal provinces threatened by Korean invasion scares. The Korean attack on Tsushima in 894 was defeated when the defenders set up their shield-wall and deployed oh-yumi (Farris 1992 p.115). By the early 10th century, however, these specialists had a poor reputation: in 914 Miyoshi Kyoyuki wrote: “Oh-yumi instructors today are permitted to buy their offices. Only the price is negotiated, with no questions as to the candidate’s ability. At the time of their appointments, those named do not even know the existence of the weapon called the oh-yumi, still less how to work the springs and bowstrings”. However he could still say that “of all the weapons of this Court, the oh-yumi is God”, and though probably becoming rare in actual use by the time he wrote – oh-yumi are not mentioned in the fighting of Masakado’s rebellion shortly afterwards, for instance – the weapon was still occasionally used in the late 12th century, as late as 1189 (Turnbull 1998; Farris 1992 p.316 etc). A decline in the number of oh-yumi available somewhen around 900 therefore seems about right.
*******
Impact on the Paekche list (II/75):
This list at the moment makes special provision for pre-500 AD Japanese allies to replace the cavalry in the published list III/7 with Bw (S). That will no longer be necessary.
Impact on the Early Samurai list (III/54):
Change the start-date from 900 to 1020, to fit with the end-date suggested here for new list III 7/D (unless the whole III/7D list, or the latter part of it, becomes a part of III/54).
*******
Shield-walls
It was standard practice – certainly in the ritsuryoh and early Heian periods, probably in the Kofun, and conceivably even in the Yayoi era – for archers to engage from behind a wall of large standing shields. There seems to be no way that these can be explicitly represented in either DBM or DBMM, so we are obliged to include them in the Bow elements. TF are not appropriate, because these shield-walls were clearly tactically mobile – Konjaku Monogatari (XXV.III; Wilson pp.197-198) describes two 10th-century armies setting up their shield-walls about 120 yards (c.145 paces) from each other for the initial exchange of arrows, and then preparing to move the shields forward for close-range shooting – and they would not have such an effect against a charge as TF do. DBMM’s Portable Obstacles have an effect in hand-to-hand but not against archery, which is precisely the opposite of what we need.
Failing any sort of portable fortification, some archers in existing lists are upgraded precisely because they are protected by pavises or similar standing shields. In the published lists Sinhalese and Early Samurai Bw (I) can be upgraded to Bw (O) because of such shields. Similarly the yugei guards in the published version of this very list owe their Bw (S) status at least in part to their use of such large shields...
Upping archers by a grade because of their shield-walls would make the heishi conscripts of the ritsuryoh armies into Reg Bw (O), whereas they are (I) at the moment. Would this be a bad thing? They are well-equipped – with sword and dagger, and often issued with armour – and periodically trained in both weapon-handling and formation work. Though the conscript regiments have a bad reputation, it is not entirely justified; the quality and treatment of conscripts seems to have declined but may originally, when the regiments were set up, have been quite high. In particular their disbandment in 792 was not because of military incapacity but to reduce the tax burden; this is likely to have been the genuine reason, not just an excuse, since the provinces where fighting was expected were precisely those where the regiments were kept on. One other reason for (I) status in the published list was to allow the chimpei and sakimori, longer-service and hence supposedly more experienced troops picked from the conscripts, to be a higher grade as Bw (O). Unfortunately the chimpei at least seem to have been demoralised by their long service and to have been regarded as inferior to the heishi of the northern provincial regiments!
The same (O) grading, but irregular, might need to be extended to the retainer or levied archers of the 10th century.
However the same logic would make the nobles and retainers of 4th-7th century armies into Bw (S). These are professionals of a warrior class, well-armoured and carrying good swords. On that basis they should be Bw (O). If they lack a particularly powerful bow, should fighting behind a wall of standing shields upgrade them to Bw (S)? It does for the yugei guards – who are basically a group of these same men, regularised by their guard service. The published Paekche list allows this classification for pre-500 archers in a Japanese allied contingent. (Though note that pavises or pavise-bearers don’t upgrade professional, armoured European crossbowmen from Bw (O) to Bw (S).) I feel a little uneasy about this, I must confess, since I have been campaigning for samurai to dismount as Bw (O), not (S), and here I am arguing for earlier Japanese to be Bw (S)....
The other option that might be suggested is to allow Bw (X) for bodies of archers behind shield-walls. One problem with this is that there is no evidence that any of the archers were fronted by spearmen (it is a particularly tempting suggestion for the heishi squads – but one for which I can’t find evidence; militiamen did use spears, but how and in what numbers?), and that is required by the definition. The other problem is that these archers should probably, on the scant evidence we have, be only one element deep. Single-based Bw (X) are not only not all that good on the table when compared with bows who can deploy two elements deep (at least under DBM: they seem to be improving under DBMM), they also lack any further differentiation (unlike double-based Bw (X), who can have (O) or (S) back ranks) – so yugei guards and heishi conscripts would all be Reg Bw (X), kofun-period nobles and 10th-century peasants all Irr Bw (X). This seems to me to be lumping too many types together.
Spearmen
Spearmen were a minority infantry type throughout the whole period covered. Spears were generally very long, but it is not at all clear whether or not they were generally used with shields. Ironically in DBx terms, the best evidence we have for the use of spears and shields together comes from Masakado’s era, for troops who should probably be not Spearmen but mere Hordes under the rules.
Nor is it clear how the spearmen fought. There is no evidence for deep formations, even the eight ranks or so represented by two DBM/M Spear elements, let alone the 10-12 ranks represented by three-deep Ax(X)/Pk(F). On the other hand, we have very little evidence on depths or formations at all.
There is at least some evidence for shields in the Yayoi and Kofun periods, though we cannot be completely sure that any of the men who used those shields were in fact spearmen. Nor can we be at all certain that, even if the spearmen carried shields, they formed the “rigid shield-walls” necessary for Spears classification under DBM or DBMM rules. They could perhaps be classed as Blades, on the assumption that they might have fought as individuals relying on skill at arms, but it just feels wrong to class men with three- or four-metre spears essentially as swordsmen.
I have therefore opted above for Spears (Inferior) in the early lists when there is at least some evidence for shields and a case for comparison with contemporary Koreans, Ax(X)/Pk(F) for the heishi of the ritsuryoh army when we have an incident with an apparently shieldless spearman, and Hordes for the later armies on grounds of quality. As noted above under the ritsuryoh list, neither Bw (X) nor even Pk (X) seems completely impossible; or there’s even a case for classing the lot as Ax (S).
Spearheads throughout this period were the socketed hoko, sometimes confusingly translated “halberd” because a few hoko have a secondary hook-like blade protruding from the base of the spearhead, curving back down towards the socket; most, however, are purely conventional spears (Friday 2004 p.85). The glaive-like naginata is first reliably mentioned in 1146 (though there are 11th-century references which may mean naginata, but are ambiguous), and the yari, with a tanged straight-edged spearhead, in 1334 (Friday 2004 pp.86-87); so both are comfortably outside this period.
*******
References (for all four lists):
Asakawa, Kan’ichi, The Early Institutional Life of Japan: A Study in the Reform of 645 A.D. (Tokyo, 1903; Paragon Book Reprint Corp., New York, 1963) – online at http://www8.big.or.jp/~yabukis/asa2005/taika-en.pdf
Aston, W G, Nihongi: Chronicles of Japan from the Earliest Times to AD 697 (Japan Society, 1896, as 2 volumes; Charles E Tuttle, Rutland, Vermont, 1972)
Barnes, Gina L, “Discoveries of iron armour on the Korean Peninsula”, Papers of the British Association for Korean Studies 5 (1994)
– “Archaeological Armor in Korea and Japan: Styles, Technology and Social Setting”, Journal of East Asian Archeology 2, 3-4 (2000)
Batten, Bruce L, “Foreign Threat and Domestic Reform: The Emergence of the Ritsuryo State”, Monumenta Nipponica vol.41 no.2 (1986)
- “The Founding of Dazaifu”, at http://www.fukuokahistory.com/live/content/view/15/52/ (2004)
Bottomley, Ian, “A new Japanese crossbow”, Royal Armouries Yearbook 3 (1998)
Bryant, Anthony J, Early Samurai 200-1500 AD (Osprey, London, 1991)
Farris, William Wayne, Heavenly Warriors: The Evolution of Japan’s Military, 500-1300 (Harvard University Press, 1992)
– Sacred Texts and Buried Treasures: Issues in the Historical Archaeology of Ancient Japan (University of Hawai’i Press, 1998)
Fredholm von Essen, Michael, “Armies and Enemies of Yamatai: Japanese Warriors of the Yayoi Period, 300 BC - AD 300”, Slingshot 200 (November 1998)
Friday, Karl F, Hired Swords: The Rise of Private Warrior Power in Early Japan (Stanford University Press, 1992)
– “Pushing Beyond the Pale: The Yamato Conquest of the Emishi and Northern
Japan” (Journal of Japanese Studies 23:1, 1997)
– Samurai, Warfare and the State in Early Medieval Japan (Routledge, London, 2004)
Hesselink, Reinier H, “The Introduction of the Art of Mounted Archery into Japan”, The Transactions of the Asiatic Society of Japan, Fourth series, volume 6 (1991)
Hong, Wontack, Relationship Between Korea and Japan in Early Period: Paekche and Yamato Wa (Ilsimsa, Seoul, 1988)
Imamura, Keiji, Prehistoric Japan: New perspectives on insular East Asia (UCL Press, London, 1996)
Kanzaki, Ivan Hisao, San Kan Seibatsu: The Yamato Invasion of Korea and the Origins of the Japanese Nation (US Naval Academy Department of History, 2002; online at www.usna.edu/History/honors/2003/KanzakiThesis.doc and http://bbs.zanhei.com/archive/index.php/t-734.html )
The Kasuga Shrine Art Masterpieces and Treasures (Nara, 1990)
Keally, Charles T, “Bad Science and the Distortion of History: Radiocarbon Dating in Japanese Archaeology” (Sophia International Review no. 26, 2004, and at http://www.t-net.ne.jp/~keally/Reports/sir2004.html )
- Yayoi Culture at http://www.t-net.ne.jp/~keally/yayoi.html (2005)
Kidder, J Edward, Japan before Buddhism (Thames & Hudson, London, 1959)
– Early Japanese Art (Thames & Hudson, London, 1964)
– Ancient Japan (Elsevier-Phaidon, Oxford, 1977)
Rabinovitch, Judith N, Shomonki: The Story of Masakado's Rebellion (Sophia University Press, Tokyo, 1986)
Suzuta, Yukinori (“Suzutayu”), Conquest of Emishi at http://www.isn.ne.jp/~suzutayu/MHJapan/ToC.html
Tsunoda, Ryosaku, and L Carrington Goodrich, Japan in the Chinese Dynastic Histories (P D & Ione Perkins, 1951)
Turnbull, Stephen, Samurai Warriors (Blandford, Poole, 1987)
– “Chinese influence on Japanese siege warfare”, Royal Armouries Yearbook 3 (1998)
– Siege Weapons of the Far East (1): AD 300-1300 (Osprey New Vanguard series, London, 2001)
– Japanese Warrior Monks AD 949-1603 (Osprey Warrior series, 2003)
– Fighting Ships of the Far East (2): Japan and Korea AD 612-1639 (Osprey New Vanguard series, London 2003)
Varley, Paul, Warriors of Japan as Portrayed in the War Tales (University of Hawai’i Press, 1994)
Wilson, William Ritchie, “The Way of the Bow and Arrow. The Japanese Warrior in Konjaku Monogatari (Monumenta Nipponica vol. 28 no. 2, 1973)
Yang Hong, “Lamellar Armor and Horse Bardings in Yamato and Koguryo and their Connections with China”, Journal of East Asian Archeology 2, 3-4 (2000)
*******
Orthography: since my long-o and similar characters not only weren’t visible on some peope’s systems, but also didn’t seem to survive being uploaded to the TNE website, I have (at Luke’s suggestion) gone over to using “oh” and “uh” for Japanese long vowels. Hence “oh-yumi” rather than “oyumi” or “o-with-a-line-over-the-top-yumi”. One exception is in titles of books and articles, where I’ve settled for the plain vowel to make online searching for the title easier.
Comments (0)
You don't have permission to comment on this page.