"Craftsmen and Specialist Troops in Early Modern Chinese Armies"

Full text

1
Chapter eleven of: Civil-Military Relations in Chinese History: From Ancient China to the Communist
Takeover, ed. by Kai Filipiak (New Work: Routledge, 2015), pp.
191-209
(Asian States and Empires; 7).
CRAFTSMEN AND SPECIALIST TROOPS IN EARLY MODERN CHINESE ARMIES
Ulrich Theobald
This paper will analyze Chinese armies of the 17th and 18th centuries from the viewpoint of labor division. While
most people interested in military history are aware that early armies consisted of infantry and cavalry troops, the
position and identity of artillerymen and engineers are far less well known. Even in Europe, there were no formal
artillery units until the late 18th century. In China, engineers and specialists often did not belong to the military corps,
but were recruited according to need during campaigns. It will be shown what kind of work they had to do in the
armies, how they were recruited, and how they were paid. Although the Qing government hailed the military
achievements of the Banner cavalry, there is no doubt that without the support of specialists coming primarily from the
private sector of economy, the victories of the mounted Banner armies could not have been achieved. The paper will
focus on some outstanding examples and will, among other things, give a detailed account of the casting of cannons by
craftsmen and the use of cannons by artillerymen as specialist troops.
Towards a definition of specialist troops
Why does an army need specialists and craftsmen? There are examples in history of armies
that were totally independent of such personnel, like the Roman armies whose troops were
trained to carry their luggage, repair their tent and weaponry, erect fortifications, and
construct siege towers, crossbows, and catapults. They seemed to be all-round talents totally
independent of specialists. This might have to do with the fact that by late Republican and
Imperial Rome, the armies no longer consisted of peasants recruited for service, but of
professionals.
1
As a shoemaker needs to know his profession, so a soldier is in need of basic
knowledge about his weaponry. Yet if it comes to complex weapons like catapults or siege
machines (including fire power artillery), the building of large fortresses, or even the
construction of bridges, pontoons, boats, ships, or roads, the basic knowledge of the common
soldier is not sufficient. There must be specialists in each army that know how to construct,
build, and operate machines and how to destroy fortifications or assail cities.
Miners, sappers and gunners were such specialists in early modern armies and no army,
whether in China or the West, could do without them. When looking at 18
th
century armies in
Europe, the most surprising issue is that the artillery, today perceived as a very ‘normal’ unit,
was still a corps of specialists, meaning tradesmen, not soldiers. Frederick the Great despised
gunners as second-rate soldiers hardly fit for the society of the ‘gentlemen’ who officered the
1
A very basic introduction into such specialist professions of the Roman army is given by Adrian Goldsworthy, The
Complete Roman Army, London: Thames & Hudson (2003), pp. 145-149, 186-197.

2
other branches of the army, especially the cavalry units.
2
While the other members of the
officer corps were often members of the aristocracy, artillerymen were looked down upon, not
only in Prussia, but throughout Europe, and especially in France. In the same way, the rank-
and-file gunners were normally not allowed to carry a musket like the elite troops of the
infantry and cavalry. A qualified gunner, on the other hand, could be regarded as greatly
superior to the enlisted men in the other arms of service because he knew how to operate
complex weapons beyond the ability to handle a standard musket.
3
The ordinary private
artilleryman in Austria was the Büchsenmeister (master gunner), a fully trained technician
who was supported by untrained infantrymen (Handlanger) providing the muscle. For this
purpose, a regiment of artillery fusiliers was created in 1757. No foreigners were to be
enlisted in this unit. The personnel received favored treatment and higher rates of pay (a skill
premium). Literacy was essential. During battle, artillery companies were detached as
batteries for the support of various infantry or cavalry units and by no means fought as a
coherent unit.
While the age of absolutism is commonly called the age of fortifications and sieges,
necessitating the use of siege artillery with mortars and heavy cannons, there was in fact also
a need for mobile field artillery.
4
Siege became less important during the 18
th
century, and
with the Napoleonic wars artillery units became common in all European states. This
development had begun a little bit earlier in France and Austria with Jean Baptiste de
Gribauval’s (1715 – 1789) general reform and standardization of the artillery. Barrels and
carriages were redesigned to reduce weight and increase mobility. Additionally, the so-called
horse artillery was created, small and very mobile units staffed with mounted gunners. The
importance of artillery can be seen in its pure number: In Austria, the number of field guns
rose from 203 in 1756 to 1,257 in 1805.
5
By the end of the Napoleonic wars, a standard
European army consisted of three branches: infantry, cavalry, and artillery, which was often
grouped into corps operating independently from the supreme commander.
6
The personnel for
the artillery units was educated in specialized academies and had classes in arithmetic,
geometry, advanced maths, mechanics and ballistics, topography and surveying, tactics,
2
Philip Haythornthwaite, Frederick the Great’s Army: 3 Specialist Troops, London: Osprey (1992), p. 3.
3
Philip Haythornthwaite, The Austrian Army 1740-80: 3 Specialist Troops, London: Osprey (1995), p. 3.
4
Geoffrey Parker, The Cambridge Illustrated History of Warfare: The Triumph of the West, Cambridge: Cambridge
University Press (1995), p. 168, mentions the common conception of the warfare of that age as ‘revolv[ing] around
sieges, not battles’.
5
David Hollins, Austrian Napoleonic Artillery 1792-1815, London: Osprey (2003), p. 4.
6
Parker, Illustrated History, p. 198.

3
logistics, staff, and adjutant work. The artillery officers were some of the most intelligent
men of the whole army.
7
Engineers were needed for the construction of fortresses. Starting in the early 17
th
century,
European history of thought clearly differentiated between the architect and engineer. While
the architect was more of a civilian – and a man of spirit and the arts – the engineer was a
‘child of Mars’ and dedicated his work to usefulness instead of creativity. The construction of
fortifications is subject to the possibilities and needs of the artillery; it is an applied science of
mathematics, and especially of geometry. The origins of the architectura militaris are to be
found in Italy (with the star fort or trace italienne as its apogee under the Ancien Régime). In
France, the title of engineer was tied to the king, emphasizing his important function in
strategy. An engineer was to be better educated than a civilian architect. Besides being
knowledgeable in construction, he had to be an expert in military affairs, especially in
ballistics.
8
In the 18
th
century, more emphasis was placed on temporary field fortifications and the
outworks of already existing fortresses. The Prussian corps of engineers was relatively small
and scattered over the various fortresses. In 1742 a pioneer regiment was founded, composed
of sappers and miners. The troops were recruited from civilian miners.
9
In Austria, there was
an engineer academy to train the officers of the engineer corps. Each infantry regiment had a
platoon of sappers, and a sapper corps was established in 1760. The miners were originally
part of the artillery units and united into a small miner brigade in 1763, but transferred to the
engineers in 1772. A pioneer battalion was established in 1758 with the duty to clear roads
and erect bridges. These specialist units included not only experts in their own field, but also
those of other professions, such as surgeons, fouriers, and chasseurs. Austria also had two
companies of pontoneers managing 100 wooden and metal pontoons each at its disposal.
10
Foresters and huntsmen were employed as Jägers (‘hunters’) and mainly entrusted with
skirmishing. The Border corps of Austria also had two companies of sharpshooters. Light
troops or hussars, often composed of people from the border nations, such as Pandours or
Croats, served as independently operating support for the normal infantry. Frederick also
despised these kinds of Austrian troops, because they allegedly lacked bravery and discipline.
7
Hollins, Austrian Napoleonic Artillery, p. 11.
8
Hartwig Neumann, Festungsbau-Kunst und –Technik: Deutsche Wehrbauarchitektur vom XV. bis XX. Jahrhundert,
Augsburg: Bechtermünz/Weltbild (2000 [1988
1
]), pp. 146-149.
9
Haythornthwaite, Frederick the Great’s Army, p. 10.
10
Haythornthwaite, The Austrian Army, pp. 9-12.

4
General staff, medical services, transport, and the commissariat, as well as the Freikorps
(privately organized units), can also be seen as specialist troops, not to forget the ‘second
arms,’ the Navy.
In summary, specialist troops were engaged in the operation of military activities that required
extraordinary knowledge and a different mode of fighting. They were deployed for the
support of regular troops in field battle (cannoneer units), during siege (bombardiers, miners,
sappers), on the way to the battlefield and back (engineers, pioneers, pontoneers), after the
battle (medical services), or for the organization of the campaign itself (general staff,
commissariat and transport).
I have made this excursion to show what we know about specialist troops in Western armies.
By comparing this information with contemporary China, we will see that there were in fact
similar concepts, but with different preferences. Unfortunately, an equivalent to the abundant
treasure of written and visual sources we have at our disposal to study warfare in the West
cannot be found for China. There are very few illustrations of warfare or warriors before the
19
th
century, and there are virtually no technical descriptions of the skills a gunner needed.
Therefore, we must rely on administrative regulations and official chronicles and archival
documents to gather information about specialist troops supporting the army.
While specialist troops (including physicians and chaplains) were a regular part of the army,
the military also needed civilian specialists who were hired just for the duration of the
campaign. Craftsmen were specialized persons serving the army in their field of knowledge
during a campaign, from bakers to carpenters and miners. Craftsmen who were ordered to
produce uniforms, tents, muskets, or gunpowder in peacetime will not be considered here, nor
will the people doing unskilled labor for the army, like drivers or porters.
Specialist troops among the Banner units
First, we will look at the composition of the troops of Qing China, as reflected in the
regulations for the fixed numbers (ding’e 定額) of troops in the various garrisons and units
provided by the Qingchao huidian zeli 清朝會典則例, a collection of administrative
regulations for the Qing state, in order to gain an impression of what types of specialist troops
existed in the traditional
11
army of Qing China. This is much easier for the Banner troops
(baqibing 八旗兵, Manbing 滿兵) than for the common Green Standard troops (lüyingbing 綠
營兵, Hanbing 漢兵). Among the Banner troops garrisoned in the Capital, three units can be
11
‘Traditional’ means, before the creation of the new local militia during the mid-nineteenth century.

5
called specialist troops, namely the Firearms Brigade (huoqiying 火器營), the Scouting
Brigade (jianruiying 健銳營) and the Vanguard Brigade (qianfengying 前 鋒 營 ). The
specialization of the last two was rather in tactics and function than in their instruments of
fighting.
The provincial Banner troops had the following specialists mentioned in the Baqi tongzhi 八
旗通志,
12
the official statutes of the Eight Banners, at their disposal:
1) Artillery specialists
- Artillery regimental commander (huo[qi]ying canling 火[器]營參領): 1 person in Qiqihar
齊齊哈爾. Interestingly enough, this commander did not directly command any gunners,
but only not further specified troops (bing) of Manchus, Soluns, Dagurs, Barhus, and
Chinese. It might be that this regimental commander was dispatched from the Capital
artillery regiment.
- Artillerymen (pao xiaoji 礟 驍 騎 ):
13
672 in provincial garrisons in China proper (not
Manchuria).
- Gunners (paoshou 礟手): 237 in provincial garrisons in China proper (not Manchuria). The
titles pao xiaoji and paoshou seem to refer to the same task, because they never both
appear in the same garrison.
14
The paoshou, nevertheless, appears to have occupied a lower
position.
15
There were thus in total 909 gunners in the provincial Banner garrisons, at least nominally.
2) Musketry specialists
- Musketry regimental commander (niaoqiangying canling 鳥槍營參領): 1 person in Ula 烏
拉, Jilin. This person commanded 48 musketry officers (niaoqiang lingcui 鳥槍領催) and
626 musketeers (niaoqiang xiaoji).
- Manchu and Mongol vanguard musketry [officers] (Manchu Menggu niaoqiang qianfeng
[xiaoji?] 滿洲蒙古鳥槍前鋒[驍騎]): 12 men. There persons are only mentioned among
the staff of the garrison of Dezhou 德 州 , Shandong. They belonged to the imperial
bordered and plain yellow Banners and commanded 188 musketeers (niaoqiang xiaoji).
12
Baqi tongzhi 八旗通志, Changchun: Jilin wenshi chubanshe (2002), fascicles 32-35.
13
Charles O. Hucker, A Dictionary of Official Titles in Imperial China, Stanford: Stanford University Press (1985), has not
translated the term xiaoji as a separate rank. In the Capital and the Northeastern provinces, xiaoji seemed to be a kind of
elite cavalryman, but still a common soldier, not an officer.
14
The garrisons staffed with paoshou were located in Fujian, Guangdong, Shaanxi, and Gansu. This might hint to a
peripheral status of these provinces. There were no gunners at all in Shandong and Jehol.
15
The subordinates of xiaojixiao ‘lieutenants’ are either called xiaoji, like in the Baqi tongzhi, or bingding 兵丁 ‘common
troops’, like in the listings in the war expenditures code Junxu zeli 軍需則例, in: Xuxiu siku quanshu 續修四庫全書,
Shanghai: Shanghai guji chubanshe (2002), vol. 857.

6
- Musketry officers (niaoqiang lingcui):
16
To be found in the provincial Banner garrisons in
China proper, with a total number of 314. Musketeer units are to be found in the garrisons
of Shanxi, Henan, Zhejiang, Hubei, and Gansu, reinforced by the 48 musketry officers in
Ula, Jilin. In some places, the ethnic origin (Manchu, Mongol, Chinese) is specified.
- Mounted musketeers (niaoqiang xiaoji 鳥槍驍騎): Most musketeers, totaling 4,084, were
commanded by the officers just mentioned (except the vanguard officers in Dezhou). There
are two exceptions, namely 680 musketeers in the garrisons of Haiguan 海關 and Miyun
密雲, Zhili, that seem to have been incorporated into the non-musketeer platoons, with no
special musketry officer commanding them.
- Musketry men (niaoqiang bing 鳥槍兵): 200 troops. They are only to be found in the
garrison of Fuzhou, and were directly commanded by non-musketry officers. Like with the
paoshou gunners, it seems that the niaoqiang bing had a lower status than the niaoqiang
xiaoji.
- Infantry musketeers (niaoqiang bubing 鳥 槍 步 兵): 400 men in Ningxia 寧夏, Gansu,
commanded by non-musketry officers. The special mention of the term ‘infantry’ suggests
that Banner musketeers normally disposed of a horse and were mounted musketeers.
17
3) Naval specialists
- Supervisor-in-chief of the naval forces (shuishiying zongguan 水師營縂管): 1 person, in
Qiqihar, Heilongjiang.
- Assistant commandants of the naval forces (shuishiying xieling 水師營協領): 3 persons, in
Lüshun 旅順, Shengjing (modern Liaoning), Fuzhou, and Guangzhou.
- Company commanders (shuishizing zuoling 水師營佐領), lieutenants (shuishiying
xiaojixiao 水師營驍騎校), and platoon commanders of the naval forces (shuishiying
fangyu 水師營防禦): These officers commanded the naval forces in Jilin, Heilongjiang,
Fujian, and Guangdong.
– Overseers of dockyards (chuanchang guanli zaochuan guan 船廠管理造船官): These
officials were actually not military personnel and are therefore only referred to with their
16
The lingcui is commonly regarded as a kind of NCO, or ‘corporal’. H[ippolit] S[emenovich] Brunnert and V. V.
Hagelstrom, Present Day Political Organization of China, London: Kelly and Walsh (1912), p. 326. Yet because these
persons do not have a superior in the garrisons, they must be considered officers.
17
Compare the characters for the Banner rank xiaoji 驍騎 that are written with the horse radical 馬. This word appears in the
Former Han period (206 BCE – 18 CE) and is explained as ‘fierce rider’, cf. Luo Zhufeng 羅竹風 (ed.), Hanyu da cidian
漢語大詞典, Beijing: Hanyu da cidian chubanshe (1993), vol. 12, p. 882. The Manchurian counterpart of this term is
aliha cooha, meaning ‘Manchu or Mongolian cavalryman’, cf. Jerry Norman, A Concise Manchu-English Lexicon,
Seattle: University of Washington Press (1978), p. 14.

7
official rank (like sipin 四品 ‘4
th
rank,’ wupin 五品 ‘5
th
rank,’ etc.). They were nominally
garrisoned in Qiqihar but were responsible for dockyards in the province of Jilin.
- Shipmen or mariners (shuishou 水手) were to be found in more places than those just
mentioned. Just like the musketeers not commanded by a musketry officer, these seamen
or ‘rivermen’ were apparently commanded by cavalry officers. This is true for the nominal
1,929 seamen in the various garrisons in Jilin and Heilongjiang, but also in Zhapu 乍浦,
Zhejiang, as well as in Fuzhou, and Guangzhou. The 60 seamen in Lüshun were
commanded by 50 officers or NCOs (non-commissioned officers, lingcui), which is quite a
surprising ratio and shows that it is difficult to generally translate lingcui as ‘NCO,’
‘officer,’ or ‘common soldier.’ In this case they seem to belong more to the lower ranks.
4) Scribes and clerks
Although scribes cannot be considered ‘troops’ because they do not fight (sappers or miners
also do not fight directly against the enemy), they were of great importance for the
administration of their garrison and the organization of warfare. They were experts in
accounting and all the paper bureaucracy needed to exchange information between
commanders, the general staff and the central government, and during the march to the front,
when the troops were to be equipped with the necessary food and transport tools, and they had
to care for the supplies in the camp. The omnipresence of scribes (Manchu bithesi, Chinese
translit. bitieshi 筆帖式) in the official and archival documents demonstrates that the army
could not do without them. The post of bithesi was a kind of lower official rank that is
mentioned in the biographies of many high officers and therefore often seems to have been a
kind of trainee position to learn about the organization of a campaign, before being promoted
to the post of a ‘fighting’ officer.
Nearly all registers and lists of military staff mention scribes among the officers, and the
officers are never without scribes. Their official rank ranged from 7 to 9. In many cases their
role seems to have been that of a quartermaster or that of an accountant, as the following titles
show: A yinwu bithesi 印務筆帖式 or suiyin bithesi 隨印筆帖式 controlled the seals, a
dangfang bithesi 檔房筆帖式 took care of the garrison archive, a lixing bithesi 理刑筆帖式
was entrusted with penal matters, the yinku bithesi 銀庫筆帖式 with the funds of the garrison,
and a cang bithesi 倉筆帖式, taizhan bithesi 臺站筆帖式, or juntai bithesi 軍臺筆帖式 with
the provisions and the logistics. There were also translators, the fanyi bithesi 繙譯筆帖式.
Other types of specialist troops did not exist. Unfortunately, similar information cannot be
obtained so easily for the Green Standard troops. Therefore, it is difficult to estimate how

8
great a proportion of the troops used muskets, or of how many cannons all types of troops
had. Such data have to be extracted from sources than the official regulations for the garrisons
(bingzhi 兵制). The Green Standard garrisons of Xuanhua 宣化 and Datong 大同, Shanxi, for
instance, were staffed with 1,000 gunners (paoshou).
18
The information above show that there were, scattered across various places in the empire,
artillery and musketry units that were sometimes commanded as separate units, and were
sometimes integrated into other units. They operated as part of a corps and could be deployed
whenever there was a need for such specialist troops.
A question I am still battling with is how many troops had muskets at their disposal. Looking
at the directives in the Baqi tongzhi, it seems as if the elite Banner troops in the northeastern
provinces did not make use of muskets at all, while up to 50 per cent of the troops of the
banner garrisons in the central provinces were equipped with muskets (see table below), while
all others only used bows and arrows. This raises the question which armament was that of
specialists: Was it the muskets, a seemingly advanced but highly unreliable and imprecise
weapon? Or was it the bow, which demanded much higher skill and greater bravery but had a
better rate of hitting the target effectually, and was even an efficient weapon even when fired
at full speed? Bows and arrows were extensively used in war, as can be seen in the copper
plate engravings of the Qing wars produced in the 18
th
century, as well as in the accounting
book of the second Jinchuan war (1771-1776), which states that almost 500,000 arrows were
used in a five-year war,
19
in contrast to 2,525,086 jin (1,507 metric tons) of lead for bullets
and cannonballs.
20
Yet the use of muskets as a convenient and easy-to-manufacture weapon
must have been so widespread at the end of the 18
th
century that one can no longer speak of
musketeers as specialists. Even during the rebellion of Wu Sangui 吳三桂, contingents of
2,000 Green Standard musketeers seem to have been quite available.
21
18
Pingding Zhunggar fanglüe 平定准噶 爾方略 , in: Yingyin Wenyuange Siku quanshu 影印文 淵閣四 庫 全書, Taibei:
Shangwu yinshuguan, 1986 (repr. by Shanghai: Shanghai guji chuban he, 1987; oldest facsimile by Shanghai: Shangwu
yinshuguan, 1934), vol. 357-359, Zhengbian 正編, fascicle 1 (QL 19/run4/yihai).
19
Pingding Liang Jinchuan junxu li'an 平定两金川军需例案, comp. by Zheng Qishan 鄭棲山, ed. by Xizang shehui kexue
yuan 西藏 社 會科學院 , in: Xizangxue Hanwen wenxian huike 西藏學漢文文獻彙刻, series 2, Beijing: Quanguo
tushuguan wenxian suowei fuzhi zhongxin (1991), fascicle 2, fol. 185a.
20
Pingding Liang Jinchuan junxu li'an, 2, fol. 184b. The figure above would result in a daily firing rate of more than 34,000
charges each and every day over a five-year period (with an average bullet calibre of 6.4 qian or 24 g). The use of no
more than 273 arrows per day is also purely statistical and does not correspond to the real importance of bow and arrow
in battle.
21
Pingding sanni fanglüe 平定三逆方略, in: Yingyin Wenyuange Siku quanshu 影印文淵閣四庫全書, Taibei: Shangwu
yinshuguan, 1986 (repr. by Shanghai: Shanghai guji chubanshe, 1987; oldest facsimile by Shanghai: Shangwu
yinshuguan, 1934), vol. 354, fascicle 42 (KX 17/12/yisi).

9
Permanently employed craftsmen among the Banner units
A lot of Banner garrisons had a fixed number of permanently employed craftsmen (jiang 匠)
at their disposal. Most of them were entrusted with special tasks, but the regulations often
merge them under the term gong-jian-tong-tie jiang 弓箭銅鐵匠, ‘craftsmen for bows,
arrows, brass, and iron.’ In the Capital and the garrisons in the province of Shengjing,
ironsmiths were prevalent, and no other craftsmen are mentioned at all. Their main task might
have been producing horseshoes, arrowheads, and weapons. More detailed information is
provided for some garrisons in Heilongjiang and some provincial garrisons in China proper.
Bowmakers (gongjiang 弓匠) and arrowmakers (jianjiang 箭匠) were part of many provincial
garrisons. Their high numbers show how important bows and arrows were as weapons, in
contrast to muskets, which were still seen as special weapons in the early 18
th
century, at least
ideologically. In the copper plate engravings of the great 18
th
century wars, therefore, bows
and arrows are the most prevalent weapons of the Banner units, whose troops fought as
mounted archers. Mounted musketeers are only seen among the Dzungars. Qing musketeers
were exclusively infantrymen, at least in these illustrations.
22
Hangzhou Fuzhou Jingzhou Chengdu
Gunners 160 28 80 48
Musketeers 736 240 1824
Infantrymen 322 320 700 256
Craftsmen 96 120 168 96
Trainees 128 220 400 144
Relations of different arms and functions in selected provincial Banner garrisons
A comparative table
23
shows how important a part the craftsmen played among the troops. In
two garrisons in Gansu, Huining 惠寧 and Huiyuan 惠遠, the term jiangyi 匠役 is used,
suggesting that these craftsmen were not permanently employed but recruited from among the
local trades according to need. Yet in fact, the dated term yi 役 simply points at someone
working for the government or an individual official (officer) belong to a government
institution, like the genyi 跟役 ‘manservants’ allotted to each soldier and officer during war.
Garrisons disposing of naval or fluvial personnel also employed boat-carpenters (chuanjiang
船匠).
22
A series of these copperplate engravings is to be found on http://crossasia.org/digital/schlachten-bilder/index/english-start,
from the collection of the Berlin State Library, as well as in the private, yet scholarly, online collection ‘Battle of
Qurman’, see http://www.battle-of-qurman.com.cn/e/hist.htm.
23
The numbers do not include officers.

10
Artillery units in action
In the following part of the paper, I will examine the activities of the artillery units, as well as
the types of cannon used by the Qing armies.
Artillery pieces in early and mid-Qing China were cast of iron or brass. Only a few military
units possessed artillery weapons at all, and the most important of these was the Firearms
Brigade (huoqiying) in the Capital. Uniaxial mountings, which Europe used for field artillery
from the 16
th
century onwards at the latest, and which allowed the artillery units enormous
mobility, were not totally unknown in China, but rather rare. Of the twenty-four types of
cannon and howitzer presented in the picture volumes of the Da-Qing huidian 大清會典
‘Collected statutes of the Great Qing,’ only seven were mounted uniaxially, most others
biaxially.
24
Most barrels (paotong 砲筒) were installed on square biaxial carriages (paoche 砲
車) with four often very small wheels, like in Western naval artillery, that did not allow long-
range movement. These cannons were solely made for defense purposes in fortifications.
Light cannon types with a barrel weight of 100 jin (59 kilos) or so could be transported on the
back of a mule or by two porters. Some heavier pieces could be carried by camels (therefore
sometimes called ‘camel guns,’ tuopao 駝礮, or probably more generally ‘cannons carried by
beasts’),
25
but for the heavy artillery, the only transport possibility was melting them down
and recasting them near the front. This was the custom during three wars, namely the First
and Second Jinchuan campaigns (1747 – 1749, 1771 – 1776) and the campaigns against
Burma/Myanmar (1766 – 1770). There were also some types of cannon whose barrels
consisted of pieces which could be screwed together (jiujiepao 九節炮), yet those cannons
only had a very small caliber, and the construction could diminish the effect of the propellant
gases.
Therefore, all materials for cannons and equipment, from gun metal to cleaning shovels, had
to be procured in the home garrisons and be transported by porters or mules to the war theatre.
Iron for cannonballs (paozi 炮 子) and barrels was bought on the free market and was
therefore called ‘merchant iron’ (shangtie 商鐵).
26
Brass and lead (for bullets) were often
confiscated from among the metals earmarked for the provincial mints.
27
24
Da-Qing huidian tu 大清會典圖, Taibei: Wenhai chubanshe (1992), fascicle 98, [fol. 18-22]; 100, [fol. 3-25].
25
Pingding Zhunggar fanglüe, Zhengbian, fascicle 77 (QL 24/8/xinchou).
26
Pingding Liang Jinchuan junxu li'an, 2, fol. 43a-b.
27
Pingding Liang Jinchuan fanglüe 平定兩金川方略 (Military annals of the war against the two Jinchuan), in: Yingyin
Wenyuange Siku quanshu 影印文淵閣四庫全書, Taibei: Shangwu yinshuguan, 1986 (repr. by Shanghai: Shanghai gu ji
chu ban she, 1987; oldest facsimile by Shanghai: Shangwu yinshuguan, 1934), vol. 356, fascicle 48, fol. 18a-19a (QL
38/1/xinhai). Pingding Liang Jinchuan junxu li'an, 2, fol. 46a-46b.

11
When the materials necessary to cast the cannons arrived at the designated location, the
commanding officer had to determine an adequate place to install a camp foundry (suiying
paoju 遂營炮局, ‘cannon foundry following the camp’). Here, the furnaces were erected in
which the metal was melted down and the cannons, howitzers, and cannonballs were cast. But
casting was only part of the work. The casting molds were made of clay and sand and
therefore gave the cannonballs and barrels only a very crude shape. In a workshop, all parts of
the barrel could be processed correctly, especially the ‘soul’ inside the barrel, whose surface
has to be as smooth as possible and fit the size of the cannonballs exactly in order to let as
little of the propellant gas as possible escape. In the Jinchuan mountains, producing a high-
quality soul in the artillery pieces was impossible. Therefore, the barrels had to be cast as
ready-to-use pieces without a rifled bore, the cannonballs had to be adjusted to the concave
diameter, and both—cannonballs and barrel—had to be made as smooth as possible.
Regulations like the Gongbu junqi zeli, the Ministry of Works’ regulations for military
equipment, and the Da-Qing huidian tu contain details concerning not only the exact
dimensions of the particular types of guns, but also the weight of the projectiles and the
amounts of gunpowder required. Thus, the quality of the guns cast on the spot cannot have
been very high – the destructive force and range of the cannons were below the standard of
the guns produced in the normal state workshops in the provincial capitals or the garrisons.
Among the materials needed for casting cannons were tools like drill heads (zuantou 鑽頭),
but barrels were cast with a hollow core (paoxin 炮心), without applying any drilling or
rifling process, except probably boring the touchhole.
28
When cannons and ammunition were ready, they were transported to a battery platform that
cannot have been very far from the furnaces, since the cannons were cast on-site because they
were not easy to transport. As can be seen in the copper plate engravings, wheel-furnished
mounts served to pull the guns to the batteries for just a short piece of way, where the road or
a path could be made broad and even enough for the purpose. Yet there were also some
barrels that were simply mounted on wooden balks. A device for adjusting the barrel to the
target seems not to have existed in such constructions. The battery itself was protected by
redoubts against shots and attacks of the rebels that would try to harass the gun crew while
loading, cleaning, and adjusting. The dimensions of such a battery and its surrounding could
be enormous, as some figures from written sources prove: For one battery, seven wooden
redoubts had to be erected. Within the surroundings giving enough protection to the gunners
and the other staff, the battery was erected with a dimension of six to seven zhang (about 20
28
Pingding Liang Jinchuan junxu li'an, 2, fol. 64a.

12
meters).
29
As the cannon was meant to fire down from the battery platform, it must be
assumed that the cited dimension corresponds to the height. Yet this would mean that the
batteries were erected as a kind of ramp inclined along the slope of the hills just above the
fortifications of the enemy. That the construction of such large platforms consumed time and
manpower can be attested by several documents: Several thousand troops were used to erect
wooden constructions and to pile up a battery platform.
30
For this purpose, trees were cut
down to erect palisades for protection.
31
The batteries ideally had to reach at least the same
ground level as the fortification of the enemy.
32
When the batteries were high enough, the
enemy’s hideouts would be easier to aim at. Other battery platforms were only two to three
zhang high (about 8 meters),
33
or four to five zhang (about 14 meters).
34
Because of the inferior quality of the barrels, the range of fire cannot have been very
impressive: According to several statements it was only the distance of an arrowshot, or even
less.
35
The guns and howitzers proved most effective when placed above the target. This meant that
the batteries often had to be erected on a mountain ridge and the cannons fired down on the
enemy’s war towers.
36
This made the transport of cannons even more difficult, but it also
shows either that the imperial gunners did not know their job very well, or that the gunpowder
and the cannons cast in the mountains were of really poor quality. Howitzers (chongtianpao
衝天砲) were rarely used in spite of the steep angle which made them, with their highly
inclined parabolic trajectory, more effective in bombarding the enemy towers and trenches
than cannons, with their less inclined trajectory.
There are many proofs that the quality of the guns was nothing to write home about: For
example, the melting furnaces of the camp foundry were not allowed to be set up too near the
battery. Bursting gun barrels seem to have been such a commonplace incidence that the
generals took precautions to prevent too many soldiers or workers from being injured or killed
by metal fragments. Once workers or porters had taken the cannonballs and gunpowder to the
battery, they had to leave immediately and go back to the foundry.
37
The imperial artillery had
a major problem: The cannons cast on the spot regularly exploded and caused damage to
29
Pingding Liang Jinchuan fanglüe, 63, fol. 4a (QL 38/6/yiyou).
30
Pingding Liang Jinchuan fanglüe, 19, fol. 13b-14a (QL 37/2/guiwei).
31
Pingding Liang Jinchuan fanglüe, 50, fol. 1b-2a (QL 38/2/bingyin).
32
Pingding Liang Jinchuan fanglüe, 51, fol. 17a (QL 38/2/jiashen); 57, fol. 24a-24b (QL 36/r3/yiyou).
33
Pingding Liang Jinchuan fanglüe, 58, fol. 8b-9a (QL 38/4/bingshen).
34
Pingding Liang Jinchuan fanglüe, 61, fol. 17a-17b (QL 38/6/dingyou).
35
Pingding Liang Jinchuan fanglüe, 52, fol. 1a (QL 38/3/gengyin); 54, fol. 16a (QL 38/3/jiwei).
36
Pingding Liang Jinchuan fanglüe, 21, fol. 5a (QL 37/2/guisi); 22, fol. 2b (QL 37/3/renyin), 12b-13a (QL 37/3/ yisi); 48,
fol. 10a (QL 38/1/bingwu).
37
Pingding Liang Jinchuan junxu li'an, 2, fol. 43b-44b.

13
equipment and people. It became evident that the brass was of inferior quality because it
contained too much slag (kuangxing 礦性), and that it was useless as gun metal.
38
Yet the
need for cannons and howitzers was so urgent that the commanding generals nevertheless
ordered artillery pieces to be cast in the camp foundries. It was ordered that only when not the
slightest impurity (shayan 砂 眼) was to be seen, was the material suitable for casting.
Therefore, it was a better method to spend time refining the metal than to cast the pieces
prematurely and lose time through recasting the exploded cannons—and lose the lives of
gunners and workmen to boot.
39
Half of the burst cannons was ‘as black as soot,’ due to the
very high amount of lead compounds (qiansha 鉛沙) contained in the material.
40
The staff of the camp foundry was to be recruited from the logistics stations. These persons
did not only bring the material for the cannons—brass bars or rods or ‘crab shells’ (xieke 蟹
殻) produced in Leshan 樂山
41
– from Chengdu 成都 to Jinchuan, but also had to transport
the finished barrel along with the ammunition from the camp foundries to the battery and thus
provided the gun crew with new supplies.
Head of the camp foundry was a foreman or ‘engineer’ (suiyingju zhangfu 遂營局長夫) who
had to supervise the cannon casting, which means that he was a skilled workman and an
expert in metallurgy. The workers serving in the foundry and later transporting the guns and
the cannonballs to the battery were allowed to walk back to the foundry without having a
specified task to do and were thus paid for a non-productive period. The foreman, who as an
expert craftsman was much more expensive than the unskilled laborers, was not allowed to go
back to the foundry without doing productive work. If this regulation was not just made for
principle, it seems clear that the distance between the foundry and the battery must have been
at least several li. One reason for this great distance might be that the locations of battery and
foundry were not calculated according to the danger arising from bursting guns or from
enemy snipers, but rather according to topography – there had to be a location convenient
enough to establish a foundry as near the battery as possible – and to the danger that the
enemy might be able to make a counter-attack. Should the enemy conquer the battery or some
cannons—which happened several times during the war—, at least the foundry would not fall
into their hands. The distance between the foundry and a battery could be up to ten stations, or
day-travels, which seems possible when it was only a question of transporting cannonballs,
38
Pingding Liang Jinchuan fanglüe, 19, fol. 19a-19b (QL 37/1/guiwei).
39
Pingding Liang Jinchuan fanglüe, 51, fol. 19a (Q 38/2/jiashen); 53, fol. 7a-7b (QL 38/3/jiachen).
40
Pingding Liang Jinchuan fanglüe, 57, fol. 1a (QL 38/r3/dingchou).
41
Pingding Liang Jinchuan fanglüe, 58, fol. 7a-7b (QL 38/4/xinmao).

14
but not if a freshly cast cannon had to be pulled forward to the battery.
42
Nevertheless, there
was a long period of time during the first half of the war when the finished barrels had to be
transported up to the top of mountain ridges to fire down on the enemy’s castles. This practice
also seems to have been common during other wars, like in Mongolia, where fortified
batteries were set up on top of mountains in Uliastai 烏里雅蘇台.
43
Whenever the front advanced, the battery had to be relocated too, and ideally the foundry
followed it at a corresponding distance. During long periods of the war, the front advanced
only very slowly, and therefore the camp foundries often remained in the same place for
months. When the battery was moved, the commanding official had to decide which cannon
could be pulled forward on its mount and which cannons should be melted down, either
because they were too heavy to be transported, because the mountain path was too narrow or
too steep, or because a cannon had become unusable. The responsibility for this decision was
borne by the supervising Grand Minister (zongli dachen 總理大臣), because the cannon brass
was of high value and any loss would have to be refunded to the ‘owner,’ in most cases the
provincial mint.
Metal that is cast and recast several times will of course lose much of its original quality, and
in order to regain the original pureness of the brass, the foreman had to add eight or nine jin of
‘pure’ brass to every 100 jin of remelted brass. If the brass had already been remelted and
recast several times, or had been buried in order to hide it from the enemy and was therefore
heavily corroded, the foundry master had to add up to 16.375 jin (16 jin 6 liang) to every 100
jin of the corroded material.
We will now take a look at the ‘normal case,’ in which cannons were not cast on the spot but
brought with the troops from the garrisons. Unfortunately, there is no evidence of how the
cannons and howitzers were produced in the garrisons. It might be that the furnaces of the
provincial mints were used for this purpose. Artillery pieces were cast according to about two
dozen models described in detail in the Da-Qing huidian tu 大清會典圖,
44
or according to
local custom, so that in the end, each provincial garrison possessed its own types of guns.
Standard cannons were not common. Even in the West, standards for field and siege artillery
were only adapted in the course of the 18
th
century.
42
Pingding Liang Jinchuan junxu li'an, 2, fol. 44a-45b.
43
Pingding Zhunggar fanglüe, Qianbian 前編, 34 (YZ 11/4/gengshen); Zhengbian, 53 (QL 23/4/dingmao).
44
Da-Qing huidian tu, fascicles 98-100.

15
The campaign against Wu Sangui made it necessary to cast a special type of cannon, the
‘great red-coat cannon’ (da hongyi pao 大紅衣炮, actually ‘great Dutch cannon’).
45
Twenty
pieces were cast especially for this war, apparently in Beijing or in a larger provincial
garrison, and sent to the war theatre in Changsha 長沙, accompanied by two officials, one
from the Ministry of War, and one from the Ministry of Works.
46
The efforts of the Jesuit
missionary and scholar Ferdinand Verbiest (Nan Huairen 南懷仁, 1623 – 1688) to cast
cannons were well remembered at that time.
47
During the campaigns against the Dzungars, cannons were brought from the garrisons, along
with other equipment or foodstuff. The same caravan of mules that brought rice also
transported cannons to the front. Three camels (or other beasts of burden?) could be tied
together side by side to transport one cannon (mei pao yi wei pei tuo san zhi 每礮一位配駝三
隻), but the copper plate engravings show that one camel could also carry one smaller gun.
48
Horses were also used to transport cannons. Of a total contingent of 5,150 horses, for
instance, 150 beasts were used for the purpose of carrying artillery pieces.
49
The numbers
were surely not very high: General Siju 席柱, for example, was allowed 30 pieces of mother-
and-child cannons (zimupao 子母礮), General Fiyanggu 費揚古 50 pieces.
50
It seems to have
been very common to join small batteries with a platoon of musketeers (qian pao qi fa 鎗礮齊
發),
51
a method that can also be seen in many scenes on the copper plate engravings.
Garrisons in the conquered territory were immediately equipped with artillery, like the four
batteries of zimu cannons in Ulaanggom 烏蘭古木.
52
The cannons were to be procured by the
Ministry of Works (gongbu 工部), which can only mean that the cannons were produced in a
foundry in Beijing and then sent to the troops’ garrisons in the east, like the 40 pieces of
cannon sent to Ningxia 寜夏 in 1724.
53
Yet not all cannons used by the various garrisons were
45
Joseph Needham, Science and Civilisation in China, vol. 5, Chemistry and Chemical Technology, part 7, Military
Technology; The Gunpowder Epic, Cambridge: Cambridge University Press (1986), p. 392, and Nicola di Cosmo, ‘Did
Guns Matter? Firearms in the Qing Formation,’ in Lynn A. Struve (ed.), The Qing Formation in World-Historical Time,
Cambridge, MA: Harvard University Press (2004), p. 140, describe the advent and early use of the hongyi cannon in
China. The word yi 衣 ‘clothes’ was used by the Manchus to avoid the taboo word yi 夷 ‘barbarian’. The word hongyi 紅
夷 ‘red [hair] barbarian’ referred to the Dutch. See also Lu Xixing 陸錫興 (ed.), Zhongguo gudai qiwu da cidian 中國古
代器物大詞典, Bingqi xingju兵器型具, Shijiazhuang: Hebei jiaoyu chubanshe (2002), pp. 137-138.
46
Pingding sanni fanglüe, 29 (KX 16/3/yisi).
47
Pingding sanni fanglüe, 8 (KX 13/8/dingyou).
48
Pingding Zhunggar fanglüe, Qianbian, 53 (QL 16/10/yiyou).
49
Ibid.
50
Pingding Zhunggar fanglüe, Qianbian, 2 (KX 54/6/renshen).
51
Pingding Zhunggar fanglüe, Qianbian, 5 (KX 57/run8/bingwu); similar statements in 8 (KX 59/10/yimao), 9 (KX
60/9/guichou), etc.
52
Pingding Zhunggar fanglüe, Qianbian, 10 (KX 61/7/dingyou).
53
Pingding Zhunggar fanglüe, Qianbian, 15 (YZ 2/11/gengxu).

16
new. Some pieces were retained after the demise of the Ming and continued to be used during
the campaigns in Xinjiang, much to the discontent of the leading generals Huang Tinggui 黃
廷桂 and Arigun 阿里衮, because the four large ‘divine’ cannons (da shenpao 大神礮) used
in the camp of General Yarhašan 雅爾哈善 were already worn out (duanlie 斷裂) and too
dangerous to be fired.
54
Probably as a result to such experiences, as well as the fact that the
transport of cannons from Beijing to the ‘Wild West’ was a complicated matter, it was
decided that new cannons were to be cast in Liangzhou 凉州, Gansu. This case from the
Jinchuan wars was treated as a precedent in the campaign against Muslim insurgents of the
Tarim Basin, and cannons were cast directly in the war theatre. Yarhašan asked to send
craftsmen (jiangyi) as well as all materials needed to the camps, as well as to cast cannons in
the camp. For the casting of 20 to 30 new cannons, experts were to be hired, and Generals
Sunggari 嵩阿禮 and Huang Tinggui were to strictly observe if they put the right spirit into
their work.
55
Apparently, the cannon casters were either hired on the labor market, for
example from blacksmiths or producers of agrarian tools, or selected from workers employed
in state-owned workshops like the provincial mint. For this extraordinary task expert state
officials and craftsmen ‘from far away’ were especially sent to Liangzhou. The experience the
Generals Jaohui 兆惠 and Šuhede 舒赫徳 had gained during the Jinchuan war was now to be
useful in producing cannons to fight the rebel Khwāja-i Jahān (‘Hojijan’ 霍集占).
56
The
craftsmen were not sent directly home (to the indefinite neidi 内地, ‘inside the provinces’)
after the necessary cannons had been cast, but remained with the army and casted further
pieces or artillery in Aksu 阿克蘇, where 120 gunners operated the batteries.
57
Cannons and
other weapons were also produced in Suzhou 肅州, Gansu.
58
The brass used to cast the
cannons was not brought back, neither in the shape of cannons nor as brass ingots or rods;
instead, the material was to remain on the spot for the casting of new, Chinese-style coins in
Xinjiang. The brass thus became part of a complex reorganization of the whole coinage
system in the newly conquered territory. Therefore, the casters remained for one or two years
to realize these monetary politics.
59
Many cannons were also to remain in the conquered
territory as a means of protecting the garrisons that were to be established or manned to
prevent future uprisings, like Hami 哈密, Barkol 巴里坤, Yarkant 葉爾羌, Kašgar 喀什噶爾
54
Pingding Zhunggar fanglüe, Zhengbian, 56 (QL 23/5/guichou and jiayin).
55
Pingding Zhunggar fanglüe, Zhengbian, 57 (QL 23/6/xinwei); 59 (QL 23/7/wushen).
56
Pingding Zhunggar fanglüe, Zhengbian, 60 (QL 23/8/bingyin).
57
Pingding Zhunggar fanglüe, Zhengbian, 64 (QL 23/11/wuxu); 66 (QL 23/12/wuyin).
58
Pingding Zhunggar fanglüe, Zhengbian, 72 (QL 24/5/yihai).
59
Pingding Zhunggar fanglüe, Zhengbian, 75 (QL 24/7/dingwu).

17
or Ili 伊 犁 . The contingents left in these garrisons included a considerable number of
craftsmen: Namely, there were 160 officers (lingcui), 320 vanguard cavalrymen (qianfeng 前
鋒), 2,800 normal cavalry troops (mabing 馬兵), 600 infantry troops (bubing 步兵), 40
gunners (paoshou), 80 craftsmen (jiangyi), and 240 trainees (yangyubing 養育兵).
60
The ratio
of gunners to craftsmen in Ili (40 : 80) was thus similar to other provincial capitals like
Chengdu (48 : 96) or Jingzhou 荊州 (80 : 168), Hubei, but they disappeared behind the large
number of cavalry troops.
Different types of craftsmen employed and hired by the army
A sufficient number of different cannons seem to have been available in the western
garrisons, so that on each campaign against the Dzungars and the Uighur city-states in the
Tarim Basin, cannons could be brought along with the equipment. These outposts were
extremely important for supplying the campaigning troops. General Yue Zhongqi 嶽鍾琪 was
entrusted with the fortification of Barkol, where ramparts and a number of batteries of an
unknown size were built.
61
The defense of a fortification was taken over by specialized
troops, namely musketeers and gunners. For this type of fortification, laborers (fuyi 夫役)
were recruited from the provinces to undertake the construction work (chenggong 城工). At
least some of the laborers seem to have been convicts (renfan 人犯) sentenced to do this hard
work for 60 days. Their penalty fulfilled, they were allowed to become soldiers or were free
to return to their home villages.
62
For wooden fortifications, craftsmen of princely
establishments or from workshops in the capital could also be dispatched.
63
The irrigation
canals of the military colonies (tuntian 屯田 ) were also planned and dug by craftsmen
(gongjiang), while the fields themselves were worked by the troops.
64
As an expression of
imperial power, steles with inscriptions (bei 碑) were erected on the sites of victory. The
inscriptions were incised by craftsmen who were especially brought to the sites of the
battlefields, like Yarkant.
65
Specialists in metal processing were dispatched to the military
colonies to help the troops in the colonies cast their own tools for working the fields. There
were several different ways to obtain iron. The most convenient way was to melt down
60
Pingding Zhunggar fanglüe, Xubian 續編, 31 (QL 30/7/renchen). The yangyubing, Manchurian term hūwašabure cooha,
were ‘young men brought up at state expense’ that were destined for military service, cf. Norman, Manchu-English
Lexicon, p. 142.
61
Pingding Zhunggar fanglüe, Qianbian, 24 (YZ 9/6/gengxu).
62
Pingding Zhunggar fanglüe, Qianbian, 27 (YZ 9/11/renxu).
63
Pingding Zhunggar fanglüe, Qianbian, 31 (YZ 10/9/guichou).
64
Pingding Zhunggar fanglüe, Zhengbian, 45 (QL 22/10/gengwu).
65
Pingding Zhunggar fanglüe, Zhengbian, 82 (QL 24/11/yisi).

18
weapons (‘swords to ploughshares’), but the craftsmen also cooperated with local experts in
prospecting different places where rocks contained a sufficient amount of iron ore to make
iron of. The tradesmen were to train the colonist troops in working the iron, and were then
sent back home.
66
Such experts did not necessarily come from the private labor market. The
Mongol Banner troops (or Dzungars?) settled down in the colony of Khobdo 科卜多,
Mongolia, were first to be trained how to use agrarian tools. Green Standard troops, already
experienced with the military colony system, were sent to Khobdo, along with several experts
in woodcutting, stonecutting, and smithery (mu-shi-tie jiang 木石鐵匠), to provide the basic
work for the colonists.
67
Green standard troops, like the Banner troops, apparently included a
handful of craftsmen, as another statement about the colonization of Ili shows: Craftsmen
from the Green Standard troops were to be selected to purchase the necessary materials and
prepare the first sets of agrarian tools that the colonists would later use.
68
This incident proves
that the Green Standard garrisons, like the Banner garrisons, also had regularly employed
craftsmen at their disposal.
During the war against Wu Sangui and his allies, boats played an important role in the
deployment of troops. Therefore, boat carpenters (duogong 舵工, chuanjiang 船匠) were
often hired to prepare tools that would allow the army to cross rivers and waterways,
69
but
war ships (jian 艦) also played an important role in many battles.
70
The army often used
pontoon bridges (fuqiao 浮橋), and they sometimes had to be repaired by specialists.
71
They
played a very important role in the Jinchuan wars, during which craftsmen were once sent out
at night to connect the two banks of a mountain creek.
72
Rope bridges (suoqiao 索橋) or
makeshift bridges (pianqiao 偏橋) were also often used. An important bridge south of an
enemy castle at Giyaržisang 甲爾日磉 was destroyed by divers from Hunan, who tore down
the wooden pillars of the construction.
73
Craftsmen were also to be used to produce camel saddles (luotuo anxie 駱駝鞍屉). Many
families in the province of Shanxi specialized in the processing of wool and the making of
ropes. These were to be entrusted with the production of the tools needed by the army, and
66
Pingding Zhunggar fanglüe, Xubian, 17 (QL 27/5/renzi).
67
Pingding Zhunggar fanglüe, Xubian, 22 (QL 28/9/wuyin).
68
Pingding Zhunggar fanglüe, Xubian, 27 (QL 29/11/gengwu).
69
Pingding sanni fanglüe, 25 (KX 15/7/jiyou), or 31 (KX 16/6/renzi)
70
Pingding sanni fanglüe, 42 (KX 17/11/jiachen).
71
Pingding sanni fanglüe, 42 (KX 17/11/yihai), or 59 (KX 20/10/dingwei).
72
Pingding Liang Jinchuan fanglüe, 12 (QL 36/12/bingzi).
73
Junjichu yue zhe bao 軍機處月摺報 2776.156, no. 37513 (QL 40/8/16, memorial by Artai 阿爾泰), quoted in Zhuang Jifa
莊吉發, Qing Gaozong shiquan wugong yanjiu 清高宗十全武功研究, Beijing: Zhonghua shuju (1987), p. 164.

19
paid an appropriate price.
74
Craftsmen for all kind of specified work were often hired from
among the people. Unlike the Roman soldiers mentioned in the beginning, the Qing troops did
not do the crude work. They saw themselves as gentlemen and not as men of the trades or the
workshops. It seems that not even the manservants (genyi) of the troops, who each officer was
served by in large numbers, engaged in woodcutting and construction work. This dirty work
was taken over by craftsmen from among the people.
75
The fine work, however, was done by
experts in the garrisons, like huge amounts of arrows that were sent to the war theatre in
Jinchuan from the nearest Banner garrison, as well as from Beijing. The making of arrows
required real expertise and could only be done by craftsmen who were regularly employed in
the garrisons, as was seen above.
76
Craftsmen (or laborers?) could even be granted a reward that was normally a privilege of
soldiers.
77
Craftsmen (jiangyi) in the camps were so important that the emperor issued a
special edict stressing the need to nourish them sufficiently. The military seems to have
regularly abused the funds for the civilian craftsmen.
78
The pay of hired craftsmen and experts
The regulations for war expenditure, Junxu zeli 軍需則例, promulgated in 1784, provide
detailed instructions about the payment granted to craftsmen hired by the army.
79
Alongside
many less or not specialized persons, like granary workers (cangfu 倉夫) and grain measurers
(douji 斗級), logistics laborers (zhanfu 站夫), and workers constructing sheds and stables
(tagai fupeng/mapeng renfu 搭 蓋 夫 棚 / 馬 棚 人 夫 ), real specialists are also dealt with,
categorized into different professions. The regulations determined that all workers coming
from provinces other than the one where the war took place (in the case of the western
campaigns, Gansu, probably including Shaanxi) were to be granted a so-called baggage
payment (xingzhuangyin 行裝銀) of six silver liang. This money was to cover the travel
expenses. Craftsmen hired in the province where the war took place were given four, five, or
six liang, depending on the distance to the war theatre. All craftsmen were furthermore paid
out a sum of three liang of ‘money for the comfort of the family’ (anjiayin 安家銀). This sum
was to cover the losses incurred if the family’s nourisher could not work for his business. On
74
Pingding Zhunggar fanglüe, Qianbian, 17 (YZ 7/11/bingzi).
75
Pingding Liang Jinchuan fanglüe, 64 (QL 38/6/bingchen).
76
Pingding Liang Jinchuan fanglüe, 57 (QL 38/run3/renwu).
77
Pingding sanni fanglüe, 1 (KX 10/12/jiwei).
78
Pingding Zhunggar fanglüe, Qianbian, 33 (YZ 11/2/gengshen).
79
Junxu zeli, Hubu junxu zeli, 6.

20
the march from their hometown to the border of the province (in case the war took place
beyond the borders of the proper provinces), the craftsmen obtained a daily pay of 0.06 liang
(or six fen) to buy food with, and outside the provincial borders one sheng (1.035 litres) of
rice. This regulation was also common for soldiers. It was apparently easier to distribute rice
than to organize moneychangers and food stalls. In the war theatre, the monthly pay granted
to the craftsmen differed according to their profession. Cannon casters were given three liang
per month, and others, like boat carpenters (chuanjiang 船匠), blacksmiths (tiejiang鐵匠),
stone cutters (shijiang石匠), wood cutters (mujiang木匠), tailors (caifeng 裁縫), and artisans
mounting pictures or maps (biaobei 裱背), were given only two liang, while ferrymen (dufu
渡夫) and boatmen (shuishou 水手) received only one and a half liang. They were also fed
daily with one sheng of rice, and their families were granted a monthly share of three dou of
rice, equivalent to one sheng per day.
80
Other groups of specialists were physicians (yisheng 醫生), hired employees of government
agencies (gongshi 供事), scribes (shushi 書識), painters (huajiang 畫匠), and papermakers
(zhijiang 紙匠). The money for the comfort of the families of physicians was as high as 50
liang, that for the others half of this sum. The baggage pay for the physicians was 30 liang,
that of the others 15 liang. The monthly pay for the physicians was three liang, the same as
for cannon casters. Employees of state agencies were granted a mere one and a half liang, the
others two liang. Physicians and state employees were allowed the assistance of a manservant,
for which they were also given money. All persons of the latter group were not given any
money for their travel to the war theatre.
81
This makes it evident that the army set up a clear
hierarchy of who was really needed, namely the men of the physical trades, with the cannon
casters, who were real professionals in their field and could not be replaced, at the top.
Compared with the rest, their work was also quite difficult, dangerous, and demanding.
Craftsmen of other professions stood in second place, and those with the easiest physical
work, like scribes, painters, and papermakers, were valued less. It is interesting to see that the
labor pay for a physician was not higher than that for a cannon caster, but that he was given
more allowances to carry all his luggage and equipment with him, and granted a higher pay
for his family. It is not certain what kind of work the gongshi ‘state employees’ did (probably
yamen runners, li 吏), but it seems that, in one way or another, they were experts in
80
Junxu zeli, Hubu junxu zeli, 6, paragraph Ge xiang jiangyi anjia gong-shi-lu fei 各項匠役安家工食路費.
81
Junxu zeli, Hubu junxu zeli, 6, paragraph Yisheng gongshi shushi huajia deng zhengzhuang anjia gongshi kouliang 醫生供
事書識畫匠等整裝安家工食口糧.

21
bureaucratic matters and could be hired out from their actual workplace to do some work in
the war theatre.
Finally, there is the interesting question of how long craftsmen and experts were used for the
army. From the regulations just explained it seems as if there was no time restriction and that
they could be hired as long as the campaign lasted. This was different from the peasants, who
were, according to the household registers, recruited to do the porterage labor in the logistics
stations. These porters had to be replaced every five months before they were set free to return
to their fields.
82
Summary
Early modern Chinese armies included several specialist troops, which were mainly
musketeer companies and artillery units. In the course of the 18
th
century, the niaoqiang
musket became so common that one can no longer speak of their handlers as specialists,
although officially, some Banner regiments still continued to use the specialist designation of
‘musketeers’ as a kind of elite troop. The bow and arrow remained the preferred standard
weapon of many Banner units, be it because of tradition or heroism, but surely not because of
lower cost. Artillery units were still very rare at the end of the Qianlong reign (1736 – 1795),
and there were very few field cannons available for the troops to the battlefield. Instead,
cannons were often cast when needed, either in government workshops, in garrisons close to
the battlefield, or even on the spot. Heavy artillery was used against fortresses, mainly in the
shape of howitzers, but light guns were likewise regularly used against enemy troops in
battles. Yet compared to 17
th
and 18
th
century Western armies, artillery played a far less
important role in Chinese warfare. The same is true for musketeers. Other weapons, like bows
and arrows in the Banner troops, or the shields, sabers or pole-arms used by the Green
Standard infantry units, were still very important and even gained more importance with the
increasing use of village militia from the late 18
th
century onwards
.83
In late imperial Chinese warfare, siege warfare did not play a crucial role. There were
virtually no techniques for crushing a fortress by digging tunnels. Accordingly, there was no
specialized miner corps among the Chinese troops of that age. Sappers were needed more
often to do ad-hoc construction work of ramparts, saps, or trenches in the field when
conditions on the battlefield required a static type of warfare. Yet there was also no
82
Junxu zeli, Hubu junxu zeli, 6, paragraph Zhanfu anjia gongshi kouliang 站夫安家工食口糧.
83
For this topic, compare Philipp A. Kuhn, Rebellion and its Enemies in Late Imperial China: Militarization and Social
Structure, 1796-1864, Cambridge, MA: Harvard University Press (1970).

22
specialized sapper corps, and the people digging out trenches or erecting barricades were
hired civilian workers.
Most garrisons had a certain number of tenured craftsmen at their disposal, mainly
blacksmiths and bow- and arrow-makers. During warfare, there was a huge need for all
different kinds of experts. These were mainly hired from the population of the region, and in
many instances came from far away districts. The army hired a lot of craftsmen to do all the
work that surpassed the ‘normal’ job requirements of a soldier. All work with wood and stone
was done by civilian experts, and the army engaged private ferrymen and boat carpenters, as
well as physicians and secretaries. Yet what is most astonishing is that the army did not
employ tenured experts in cannon casting. Although there must have been casters in various
state-owned workshops, like the mints, apart from fortress artillery, the arsenals of the huge
armies of Qing China had only a small array of cannons at hand. Field artillery was rarely
used, and if problems with the material occurred, private experts or semi-experts were hired.
For geodesy, surveying, and even for ballistic calculations, the army made use of Jesuit
mathematicians like Felix da Rocha (1713 – 1781).
84
This is an excellent example to
demonstrate that, although Chinese armies were staffed with experts and specialists, and knew
how to integrate civilian craftsmen into the army during war, any kind of scientific education
for engineers was missing. But the reasons for this are another story.
BIBLIOGRAPHY
Baqi tongzhi 八 旗 通 志 (Comprehensive treatises on the Eight Banners). Changchun: Jilin wenshi
chubanshe (2002).
Junxu zeli 軍需則例 (Regulations for war expenditure), in: Xuxiu siku quanshu 續修四庫全書, Shanghai:
Shanghai guji chubanshe (2002), vol. 857.
Pingding sanni fanglüe 平定三逆方略 (Military annals of the war against the Three Rebels), in: Yingyin
Wenyuange Siku quanshu 影印文淵閣四庫全書, Taibei: Shangwu yinshuguan, 1986 (repr. by
Shanghai: Shanghai guji chubanshe, 1987; oldest facsimile by Shanghai: Shangwu yinshuguan,
1934), vol. 354.
Pingding Zhungar fanglüe 平定準噶爾方略 (Military annals of the war against the Dzungars), in: Yingyin
Wenyuange Siku quanshu 影印文淵閣四庫全書, Taibei: Shangwu yinshuguan, 1986 (repr. by
Shanghai: Shanghai guji chubanshe, 1987; oldest facsimile by Shanghai: Shangwu yinshuguan,
1934), vol. 357-359.
Pingding Liang Jinchuan fanglüe 平定兩金川方略 (Military annals of the war against the two Jinchuan),
in: Yingyin Wenyuange Siku quanshu 影印文淵閣四庫全書, Taibei: Shangwu yinshuguan, 1986
(repr. by Shanghai: Shanghai guji chubanshe, 1987; oldest facsimile by Shanghai: Shangwu
yinshuguan, 1934), vol. 356.
84
Joanna Waley-Cohen, The Culture of War in China: Empire and the Military under the Qing Dynasty, London: Tauris
(2006), p. 58.

23
Pingding Liang Jinchuan junxu li'an 平定两金川军需例案 (Precedents for war expenditure of the war
against the Two Jinchuans), comp. by Zheng Qishan 鄭棲山, ed. by Xizang shehui kexue yuan 西藏
社會科學 院 , in: Xizangxue Hanwen wenxian huike 西藏學 漢 文 文獻彙刻 , series 2, Beijing:
Quanguo tushuguan wenxian suowei fuzhi zhongxin (1991).
Da-Qing huidian tu 大清會典圖 (Illustrations to the collected statutes of the Qing dynasty). Taibei:
Wenhai chubanshe, 1992.
Qingchao wenxian tongkao 清朝文獻通考 (Comprehensive treatises of the Qing dynasty based on various
sources, original title Huangchao wenxian tongkao 皇朝文獻通考), in: Yingyin Wenyuange Siku
quanshu 影印文淵閣四庫全書, Taibei: Shangwu yinshuguan, 1986 (repr. by Shanghai: Shanghai
guji chubanshe, 1987; oldest facsimile by Shanghai: Shangwu yinshuguan, 1934), vol. 632-638.
Brunnert, H[ippolit] S[emenovich] and V. V. Hagelstrom (1912). Present Day Political Organization of
China. London: Kelly and Walsh (repr. by Taipei: Ch'eng Wen, 1971; originally published in
Russian in 1910).
Chen Feng 陳鋒 (1992). Qingdai junfei yanjiu 清代軍費研究 (Military expenditure in Qing China).
Wuchang: Wuhan daxue chubanshe.
di Cosmo, Nicola (2004). ‘Did Guns Matter? Firearms in the Qing Formation,’ in Lynn A. Struve (ed.), The
Qing Formation in World-Historical Time, Cambridge, MA: Harvard University Press, pp. 121-166.
Goldsworthy, Adrian (2003). The Complete Roman Army. London: Thames & Hudson.
Guddat, Martin (1992). Kanoniere, Bombardiere, Pontoniere: The Artillerie Friedrichs des Großen
(Cannoneers, bombardeers, pontoneers: The artillery of Frederick the Great). Herford/Bonn: Mittler.
Hucker, Charles O. (1985). A Dictionary of Official Titles in Imperial China. Stanford: Stanford University
Press.
Haythornthwaite, Philip (1988). Wellington’s Specialist Troops. London: Osprey.
Haythornthwaite, Philip (1990). Austrian Specialist Troops of the Napoleonic Wars. London: Osprey.
Haythornthwaite, Philip (1992). Frederick the Great’s Army: 3 Specialist Troops. London: Osprey.
Haythornthwaite, Philip (1995). The Austrian Army 1740-80: 3 Specialist Troops. London: Osprey.
Hofschroer, Peter (2003). Prussian Staff and Specialist Troops, 1791-1815. London: Osprey.
Hollins, David (2003). Austrian Napoleonic Artillery 1792-1815. London: Osprey.
Kuhn, Philipp A. (1970). Rebellion and its Enemies in Late Imperial China: Militarization and Social
Structure, 1796-1864. Cambridge, MA: Harvard University Press.
Lu Xixing 陸錫興 (ed., 2002). Zhongguo gudai qiwu da cidian 中國古代器物大詞典, Bingqi xingju 兵器
型具 (Great dictionary of ancient Chinese tools. Weapons and penal instruments). Shijiazhuang:
Hebei jiaoyu chubanshe.
Luo Zhufeng 羅竹風 (ed. 1993). Hanyu da cidian 漢語大詞典 (Comprehensive dictionary of Chinese).
Beijing: Hanyu da cidian chubanshe.
Needham, Joseph (1986). Science and Civilisation in China, vol. 5, Chemistry and Chemical Technology,
part 7, Military Technology; The Gunpowder Epic. Cambridge: Cambridge University Press.
Neumann, Hartwig (2000 [1988
1
]). Festungsbau-Kunst und –Technik: Deutsche Wehrbauarchitektur vom
XV. bis XX. Jahrhundert (The art and technique of fortification in Germany, 15
th
to 20
th
centuries).
Augsburg: Bechtermünz/Weltbild.
Norman, Jerry (1978). A Concise Manchu-English Lexicon. Seattle: University of Washington Press.
Parker, Geoffrey (1995). The Cambridge Illustrated History of Warfare: The Triumph of the West.
Cambridge: Cambridge University Press.
Waley-Cohen, Joanna (2006). The Culture of War in China: Empire and the Military under the Qing
Dynasty. London: Tauris.
Zhuang Jifa 莊吉發 (1987). Qing Gaozong shiquan wugong yanjiu 清高宗十全武功研究 (The ‘Ten
Successful Campaigns’ of emperor Qing Gaozong). Beijing: Zhonghua shuju.