In the previous chapter, I introduced the early Ming notion of a bounded empire based on the (imagined) ideal of Han ethno-cultural homogeneity. From the mid-fifteenth century onward, military threats and the growing awareness of ethno-cultural heterogeneity within the imperium brought about the re-envisioning of Ming primarily as a politico-territorial network connected by the bureaucratic order, as shown in my analysis of the mid-fifteenth century comprehensive gazetteer, Da Ming yitong zhi. This chapter focuses on the transformation in imperial mapping of frontiers, showing an increasing emphasis on territorial knowledge throughout the late fifteenth and early sixteenth centuries. Such a cartographic change both reflected and encouraged the shift in emphasis in the imperial authority in frontiers, from more resource-oriented to more territoryoriented: from its ability to extract and manage resources from locales to the ability to learn about, plan and transform territories.
在上一章中，我介绍了明朝早期基于（想象中的）汉族民族文化同质性理想的有界帝国概念。从 15 世纪中叶开始，军事威胁和对帝国内部民族文化异质性的日益认识，使明朝主要被重新视为一个由官僚秩序连接起来的政治领土网络，正如我对 15 世纪中叶综合地名录《大明一统志》的分析所示。本章重点讨论帝国疆域地图绘制的转变，展示了整个十五世纪末十六世纪初对领土知识的日益重视。这种地图绘制上的变化既反映了也鼓励了皇权在边疆的重点转移，即从更注重资源转向更注重领土：从其从当地提取和管理资源的能力转向了解、规划和改造领土的能力。

My examination of Ming frontier maps complicates the established narrative of cartography and the state’s power. Many studies of early modern European cartography emphasize the relationship between the cartographic revolution and the revolution in government of early modern Europe. They highlight the significance of the cartographic technology that newly enabled the rulers to conceive of space as empty, homogeneous, and universal, defined by the geometric lines of latitude and longitude, and therefore to consider it as an object that can be
我对明代边疆地图的研究使制图学与国家权力的既定叙事复杂化。许多关于近代早期欧洲地图学的研究都强调了地图学革命与近代早期欧洲政府革命之间的关系。他们强调了制图技术的重要性，这种新技术使统治者能够将空间概念为空洞、均质和普遍的，由几何经纬线定义的空间，并因此将其视为可以被控制的对象。
occupied, controlled, and exploited. ${}^{147}$${}^{147}$^(147){ }^{147} This epistemological change led to the shift from a medieval notion of a composite monarchy that governed heterogeneous estates to an early modern territorial state that established a unified and centralized rule over a uniform space. ${}^{148}\mathrm{It}$${}^{148}\mathrm{It}$^(148)It{ }^{148} \mathrm{It} sounds ahistorical to ask the question whether the early modern state could enhance its control over frontier spaces without such a scientific advancement. The maps I examine in this chapter are seemingly unchanging in the sense that they did not show a similar radical degree of scientific or epistemological transformations. Yet my analysis still demonstrates a major shift in mapping frontiers, initiated by the Ming court’s efforts using cartography to enhance central supervision over frontier military garrisons and to establish its claim over contested northern territories. These maps facilitated a new territorial understanding of frontiers, seeing them as internally homogeneous with each other and clearly demarcated from the outside.
${}^{147}$${}^{147}$^(147){ }^{147} 这种认识论上的变化导致了从中世纪统治异质庄园的复合君主制概念向近代早期领土国家的转变。 ${}^{147}$${}^{147}$^(147){ }^{147} 这种认识论上的变化导致了从中世纪统治异质庄园的复合君主制概念向近代早期领土国家的转变，近代早期领土国家在统一的空间内建立了统一的中央集权统治。 ${}^{148}\mathrm{It}$${}^{148}\mathrm{It}$^(148)It{ }^{148} \mathrm{It} 如果提出这样一个问题：如果没有这样的科学进步，早期现代国家能否加强对边境空间的控制？我在本章中研究的地图看似一成不变，因为它们没有出现类似程度的科学或认识论变革。然而，我的分析仍然表明，明朝朝廷利用地图学加强中央对边疆驻军的监管，并确立其对有争议的北方领土的主权要求，从而引发了绘制边疆地图的重大转变。这些地图促进了对边疆领土的新理解，将边疆视为内部同质、外部清晰划分的领土。

Such a cartographic transformation is closely related to military crises and emerging defensive interests_beginning from the mid-fifteenth century. In the early fifteenth century, the court’s interest in frontiers focused mainly on the management of strategic resources. Boundaries were rarely demarcated, and information about local terrains and ethno-cultural differences was scarce. These features reflected the early Ming military strategy, which combined preemptive military campaigns with maintaining a selected number of forward garrisons and diplomatic
这种制图转型与始于 15 世纪中叶的军事危机和新出现的防御利益密切相关。15 世纪早期，宫廷对边境的兴趣主要集中在战略资源的管理上。边界很少划定，有关当地地形和民族文化差异的信息也很少。这些特点反映了明朝早期的军事战略，即把先发制人的军事行动与维持一定数量的前沿驻军和外交活动结合起来。

allies. Frontiers did not comprise a contiguous space but remained as a composite of strategic places where the military power was concentrated. ${}^{149}$${}^{149}$^(149){ }^{149}
盟国。疆域并不是一个连续的空间，而是军事力量集中的战略要地的综合体。 ${}^{149}$${}^{149}$^(149){ }^{149}

From the mid-fifteenth century onward, maps emerged as a tool for the court to collect and archive information to supervise northern frontier defense. In the early fifteenth century, the Yongle emperor abandoned the original position of forward garrisons and relocated them to the south to strengthen the defense of a new capital Beijing. This decision shortened the supply lines and reduced the fiscal burden, which had been increased due to the Yongle’s preference to costly military campaigns. Both some later Ming officials and modern scholars have concluded that this decision weakened the Ming military power in checking hostile actions and reduced diplomatic power in the steppe world. When facing a greater threat from the Mongols under the unified leadership of Esen, the Zhengtong Emperor led a military campaign in 1449 that ended with a great failure. Esen’s army completely destroyed the imperial army near at the fortress of Tumu and captured the emperor himself. ${}^{150}$${}^{150}$^(150){ }^{150} This traumatic event, called the Tumu Debacle, gave a clear message to the entire officialdom that the early Ming military strategy was no longer feasible. ${}^{151}$${}^{151}$^(151){ }^{151} Although the growing fragmentation among the Mongols alleviated this threat in the late fifteenth century, smaller-scale raids continued across the northern frontiers. In order to raise alerts before such attacks and defend possible invasion routes, officials constructed an increasing
从十五世纪中叶开始，地图成为朝廷收集和存档信息以监督北方边防的工具。十五世纪初，永乐皇帝放弃了原来的前沿驻军位置，将其迁往南方，以加强新首都北京的防御。这一决定缩短了补给线，减轻了财政负担，而由于永乐皇帝偏爱耗资巨大的军事行动，财政负担有所增加。后来的一些明朝官员和现代学者都认为，这一决定削弱了明朝遏制敌对行动的军事力量，并削弱了明朝在草原世界的外交力量。正统皇帝在 1449 年面临着蒙古人在额森统一领导下的更大威胁时，领导了一次军事行动，但以巨大的失败告终。额森的军队在土木堡附近彻底摧毁了正统皇帝的军队，并俘虏了正统皇帝本人。 ${}^{150}$${}^{150}$^(150){ }^{150} 这一惨痛事件被称为土木之变，它向整个官方发出了一个明确的信息：明朝早期的军事战略已不再可行。 ${}^{151}$${}^{151}$^(151){ }^{151} 尽管在 15 世纪晚期，蒙古人的日益分裂缓解了这一威胁，但小规模的袭扰在北方边境仍在继续。为了在此类袭击之前提高警惕，并保卫可能的入侵路线，官员们修建了越来越多的

number of smaller defense facilities outside major guards and battalions. During the late fifteenth and the early sixteenth centuries, Ming defense strategies focused on building and fortifying fortresses (tunbao) and beacon towers. ${}^{152}$${}^{152}$^(152){ }^{152}
在主要的卫所和大营之外，还有许多较小的防御设施。十五世纪末十六世纪初，明朝的防御战略重点是修建和加固堡垒（屯堡）和烽火台。 ${}^{152}$${}^{152}$^(152){ }^{152}

In the sixteenth century, military crisis continued to worsen in the northern frontiers under the pressure of the reunified Mongol leadership under Dayan Khan (Batumöngke; 14641543), and his grandson Altan Khan (Anda; 1507-1582). When the Ming court declined the Mongols’ request for increased tribute trade, the Mongols raided northern frontiers and even reached the gates of Beijing in 1550. Alarmed, the Ming court continued its endeavors to fortify fortresses and beacon towers, while at the same time planning the construction of larger-scale defense facilities, including extensive walls and trenches. ${}^{153}$${}^{153}$^(153){ }^{153} With the increased military threats and the state’s intensified involvement in frontier environment through construction projects, the court requested more detailed information about local geography, such as where enemies typically intruded and which places they often targeted, what the local terrain looked like, and how Ming soldiers could take geographical advantage in military operations. My analysis
16 世纪，在达延汗（Batumöngke，1464-1543 年）及其孙子阿勒坦汗（Anda，1507-1582 年）领导下的统一蒙古领导层的压力下，北方边疆的军事危机持续恶化。明廷拒绝了蒙古人增加朝贡贸易的要求，蒙古人便袭击了北方边境，甚至在 1550 年攻入了北京城。明廷大为震惊，继续努力加固堡垒和烽火台，同时计划修建更大规模的防御设施，包括大面积的城墙和壕沟。 ${}^{153}$${}^{153}$^(153){ }^{153} 随着军事威胁的增加以及国家通过建设项目加强对边疆环境的参与，朝廷要求获得更多有关当地地理的详细信息，例如敌人通常会入侵哪些地方，他们通常会以哪些地方为目标，当地地形是什么样子的，以及明朝士兵如何在军事行动中利用地理优势。我的分析

demonstrates how the Ming court endeavored to develop a close-up perspective toward frontier spaces, and how the cartographic change during this era shifted the notion of imperial power in frontier spaces.
该书展示了明朝宫廷如何努力发展对边疆空间的近距离观察，以及这个时代的制图变革如何改变了边疆空间的皇权观念。

This chapter is organized in chronological order. Sections two and three focus on how the early Ming court used maps mainly to manage strategic resources in frontiers. The examples I analyze show that imperial sovereignty was defined less territorially during the early fifteenth century. Instead, the court’s power in frontier regions was restricted to strategic places where military garrisons were located. These garrison towns served as points or nodes of an empirewide networks, which enabled the court’s extraction and circulation of people, goods, and information. Courier route maps created by the court visualized such networks. ${}^{154}$${}^{154}$^(154){ }^{154} Sections four and five investigate maps produced after the mid-fifteenth century, with emphasis on how military crises brought about changes in the court’s information-gathering practices. My examination elucidates how maps facilitated the process of territorializing imperial rule in frontier regions. Through collecting and archiving maps, the court accumulated detailed
本章按时间顺序编排。第二节和第三节主要介绍明初朝廷如何使用地图来管理边疆的战略资源。我分析的例子表明，在 15 世纪早期，朝廷主权的领土界定较少。相反，朝廷在边疆地区的权力仅限于驻军所在的战略要地。这些驻军城镇是帝国范围内网络的节点，使宫廷能够提取和流通人员、货物和信息。朝廷绘制的信使线路图将这些网络形象化。 ${}^{154}$${}^{154}$^(154){ }^{154} 第四和第五部分研究了 15 世纪中叶以后制作的地图，重点是军事危机如何给宫廷的信息收集实践带来变化。我的研究阐明了地图如何促进了帝国在边境地区的统治领土化进程。通过收集和存档地图，宫廷积累了详细的
${}^{154}$${}^{154}$^(154){ }^{154} The visualization of a limited local area of strategic importance is not at all distinct and rather universal in premodern and early modern cartography. For instance, early modern Ottoman maps reveal how the atlas of topographic maps (each map has a distinct city view) constructed the limit of the empire, instead of the continuous borderline. The mapping of the contiguous national territory emerged only in the later period with the shifted notion of territories. See Kathryn A. Ebel, “Representations of the Frontier in Ottoman Town Views of the Sixteenth Century,” Imago Mundi 60, no. 1 (2008), 1-22.; Palmira Brummett, “The Fortress: Defining and Mapping the Ottoman Frontier in the Sixteenth and Seventeenth Centuries” in The Frontiers of the Ottoman World, ed. A. C. S. Peacock (New York: Oxford University Press, 2009), 31-56.
${}^{154}$${}^{154}$^(154){ }^{154} 在前现代和早期现代地图绘制中，具有战略重要性的有限局部区域的可视化并不独特，而是普遍存在的。例如，现代早期的奥斯曼地图揭示了地形图集（每幅地图都有一个独特的城市视图）如何构建帝国的界限，而不是连续的边界线。只有到了后期，随着领土概念的转变，才出现了连绵不断的国家领土地图。见 Kathryn A. Ebel, "Representations of the Frontier in Ottoman Town Views of the Sixteenth Century," Imago Mundi 60, no. 1 (2008), 1-22; Palmira Brummett, "The Fortress：十六世纪和十七世纪奥斯曼帝国边境的定义和绘图"，载于《奥斯曼帝国世界的边境》，A. C. S. Peacett 编辑，第 1 卷（2008 年），第 1-22 页。A. C. S. Peacock（纽约：牛津大学出版社，2009 年），31-56。
knowledge of local geography. The knowledge produced in the format of maps testified to and created the court’s belief in its power to control and transform frontier landscapes. In other words, the maps helped to redefine imperial power in frontier regions in terms of space. Power became spatialized.
当地地理知识。以地图形式产生的知识证明并创造了朝廷对其控制和改变边疆景观的权力的信念。换句话说，地图有助于从空间的角度重新定义帝国在边疆地区的权力。权力变得空间化了。

In 1397, in a letter sent to his third son, the Prince of Jin (whose posthumous title is Prince Gong of Jin 晋恭王, 1358-98), the Hongwu emperor expressed his fury over the prince’s recent horse grazing several hundred $li$$li$lil i away from Kaiping garrison. ${}^{155}$${}^{155}$^(155){ }^{155} The emperor’s letter was provoked by the complaints of neighboring nomadic tribes who informed the emperor of the prince’s intrusion into their grazing lands. Interestingly, Hongwu’s reaction focused instead on the fact that the prince’s action endangered his horses far outside the safety zone. His letter said:
1397 年，洪武皇帝在给他的第三个儿子晋恭王（谥号晋恭王，1358-98 年）的一封信中，对晋恭王最近在开平驻军几百 $li$$li$lil i 外放牧表示愤怒。 ${}^{155}$${}^{155}$^(155){ }^{155} 洪武皇帝的这封信是由邻近游牧部落的抱怨引起的，这些部落向洪武皇帝告发太子闯入了他们的牧场。有趣的是，洪武的反应却集中在太子的行为危及了他的马匹，而且远远超出了安全区。他在信中说：

While having vast lands of six thousand $li$$li$lil i from Liaodong to Gansu [throughout the northern frontier], our dynasty has only 100,000 warhorses. ${}^{156}$${}^{156}$^(156){ }^{156} However, it is difficult to dispatch horses [to the northern frontier] quickly from the metropolitan area [the area near Nanjing], Henan, and Shandong. Accordingly, only 20,000 horses beyond the Beiping Pass 北平口 can be mobilized swiftly in event of emergency… Recently you
$li$$li$lil i 从辽东到甘肃[整个北部边疆]，我朝拥有广袤的六千 $li$$li$lil i 土地，却只有十万匹战马。 ${}^{156}$${}^{156}$^(156){ }^{156} 然而，从大都（南京附近地区）、河南、山东等地迅速调派战马[前往北疆]是很困难的。因此，北平口外只有两万匹战马可以在紧急情况下迅速调动...... 最近您

[the Prince of Jin] led several thousand horses and went 300-400 li away from Kaiping. While staying in a wild plain beyond the stronghold (曠塞), you even did not send light cavalry as scouts who could alert you about the enemies’ movements. What if a few ten thousand northern $hu$$hu$huh u soldiers secretly approached in the daytime and made a sudden attack at night? How can our few thousand horses be a match for those several ten thousand enemies? ${}^{157}$${}^{157}$^(157){ }^{157}
[晋王]率领数千骑兵，离开开平三四百里。在 "曠塞 "之外的荒野平原上停留，卻不派輕裝騎兵作為斥候，警惕敵人的動向。如果有几万北军白天偷偷接近，晚上突然袭击怎么办？我们的几千骑兵怎么可能是那几万敌人的对手？ ${}^{157}$${}^{157}$^(157){ }^{157}

After this criticism, the emperor commanded that “since we do not have enough horses, in moving back and forth among camps you are allowed to travel no more than 20-30 li away from the garrison seat.”
在这次批评之后，皇帝命令说："由于我们没有足够的马匹，在营地之间来回走动时，你们不得离开驻军所在地超过二三十里"。

This letter reflects well the emperor’s perception of frontiers, where Ming soldiers were outnumbered by “several ten thousand” hostile enemy forces. According to the emperor’s own words, the enemy could annihilate his precious horses and soldiers at any moment if they wandered outside the areas adjacent to Ming strongholds. This perception indicates the reality that the early Ming state had neither the ability nor the willingness to establish territorial control over its military frontier. Instead, the state’s control was limited to a few strategic areas, where military garrisons (wei or suo), forts (bao), and courier stations (yi or pu) were established. ${}^{158}$${}^{158}$^(158){ }^{158}
这封信很好地反映了皇帝对边疆的看法，在那里，明朝士兵的人数比敌军多 "几万"。根据皇帝自己的说法，如果敌军离开明朝据点附近的地区，随时都可能歼灭他的珍贵兵马。这种看法表明，早期明朝国家既没有能力也没有意愿对其军事边疆建立领土控制。相反，国家的控制仅限于少数几个战略要地，在这些地方建立了驻军（卫或所）、堡垒（堡）和信使站（驿或铺）。 ${}^{158}$${}^{158}$^(158){ }^{158}

In order to protect its limited power far away from the center, the court made serious efforts to keep track of strategic resources registered to each military unit. Particular attention was paid to military registers (junji or junren banji 軍人版籍), or the records of hereditary
为了保护其远离中央的有限权力，朝廷认真记录各军事单位登记的战略资源。朝廷尤其重视军籍（军人版籍）或世袭罔替的记录。

military households attached to each garrison. ${}^{159}$${}^{159}$^(159){ }^{159} Emperor Hongwu established the Ming military system based on hereditary soldiers and military agricultural colonies (tuntian 屯田). In theory, families registered as military households were required to provide one adult male for military service in each generation. Military garrisons were expected to be self-sufficient, with their food provided by agricultural colonies. While this institutional design mainly intended to reduce the fiscal burden of the military, the limited agricultural productivity in frontiers combined with frequent violent conflicts and local officers’ mismanagement caused many soldiers to desert.
每个驻军都有附属的军户。 ${}^{159}$${}^{159}$^(159){ }^{159} 洪武皇帝建立了以世袭士兵和军事农业殖民地（屯田）为基础的明朝军事制度。理论上，登记为军户的家庭每代必须有一名成年男子服兵役。驻军应自给自足，其粮食由农业殖民地提供。虽然这一制度设计的主要目的是减轻军队的财政负担，但边疆地区有限的农业生产力加上频繁的暴力冲突和地方官员的管理不善，导致许多士兵开小差。

Facing the increasing problem of desertion, early Ming information-gathering in frontiers concentrated on tracing the changes in the number of soldiers. Upon assuming a throne after a civil war with his nephew, the Jianwen emperor (r. 1398-1402), the Yongle emperor (r. 14021424) issued an imperial edict that required investigation of the condition of military garrisons. While criticizing military officers who failed to report the number of deserters in order to avoid responsibility, Yongle sent censorial officials to investigate the actual condition of the garrisons and to renew outdated military registers.
面对日益严重的逃兵问题，明朝早期的边疆信息收集工作主要集中在追踪士兵人数的变化上。与侄子建文帝（在位 1398-1402 年）内战后即位的永乐皇帝（在位 14021424 年）下诏要求调查驻军情况。永乐在批评军官为逃避责任而不上报逃兵人数的同时，还派遣佥事调查驻军的实际情况，并更新过时的军籍。

A great number of soldiers all around the empire ran away during the reign of the Jianwen emperor. Military officers in charge of garrisons failed to trace and arrest these deserters. Send executive staff from the office of scrutiny (jishi zhong 給事中) and have
建文帝时期，帝国各地有大量士兵逃跑。负责驻军的军官未能追查和逮捕这些逃兵。从监察御史中丞处派出执行人员，并让他们

them investigate… Also order them to make maps depicting the difficult and easy features of the terrain and the travel distances by road and submit them [to the ministry of war]. Also request them to report the actual figures for provisions, weapons, boats, and wagons [in each garrison] to the ministry. ${}^{160}$${}^{160}$^(160){ }^{160}
命令他们调查......还命令他们绘制地图，标明地形的难易程度和路途距离，并将地图[提交给战争部]。还要求他们将[各驻军]的粮草、武器、船只和马车的实际数字上报给军部。 ${}^{160}$${}^{160}$^(160){ }^{160}

In the above edict, even though the updating of the actual number of soldiers took priority, the registers included other strategic information, such as the number of horses, ships, and wagons assigned to each garrison. The garrison commanders also submitted the registers together with maps. In this edict, the emperor did not clarify why he wanted his officials to compile new maps of garrisons. One possible interpretation is that the court requested the same format for military registers by attaching a map to the registers, similar to the yellow registers (huangce), or civilian household registers. ${}^{161}$${}^{161}$^(161){ }^{161} The maps attached to the yellow registers were pictorial images of agricultural resources for tax purposes. Considering the main purpose of the military registers was to monitor available resources, the maps attached to the military registers probably were intended to track the state’s agricultural colonies parallel to the yellow registers for civilian lands. ${}^{162}$${}^{162}$^(162){ }^{162}
在上述诏书中，尽管更新士兵的实际人数是优先事项，但登记册还包括其他战略信息，如分配给每个驻军的马匹、船只和马车的数量。驻军指挥官还将登记表与地图一并提交。在这份诏书中，皇帝并没有说明为什么要让官员们绘制新的驻军地图。一种可能的解释是，朝廷要求采用与黄册或民间户籍类似的格式，在军事登记册上附上地图。 ${}^{161}$${}^{161}$^(161){ }^{161} 黄册所附的地图是农业资源的图像，用于征税。考虑到军事登记簿的主要目的是监控可用资源，军事登记簿所附的地图很可能是为了追踪国家的农业殖民地，与针对平民土地的黄册类似。 ${}^{162}$${}^{162}$^(162){ }^{162}
${}^{160}$${}^{160}$^(160){ }^{160} Ming Taizong shilu, 10-xia: 6.
${}^{160}$${}^{160}$^(160){ }^{160} 《明太宗实录》十夏：6。
${}^{161}$${}^{161}$^(161){ }^{161} In 1381, Emperor Hongwu requested every county and prefecture to submit one register (ce 冊) for every 110 households ( $li$$li$lil i 里), with the attachment of a map at the front page. Ming Taizu shilu, 135: 4.
${}^{161}$${}^{161}$^(161){ }^{161} 1381年，洪武皇帝要求各州县每110户（ $li$$li$lil i 里）上报一份户籍，并在首页附图。明太祖实录》，135：4。
${}^{162}$${}^{162}$^(162){ }^{162} Despite the edict’s asking for maps “depicting the difficult and easy features of the terrain,” the map probably did not specify the actual shape of the lands. The “xianyi (險易)” was common expression for describing any pictorial images $(tu)$$(tu)$(tu)(t u) depicting a space, regardless of their actual function or visual features.
${}^{162}$${}^{162}$^(162){ }^{162} 尽管诏书要求地图 "描绘地形的难易特征"，但地图可能并没有说明土地的实际形状。 $(tu)$$(tu)$(tu)(t u) "险易 "是描述任何描绘空间的图像 $(tu)$$(tu)$(tu)(t u) 的常用语，不论其实际功能或视觉特征如何。

Maps not only served as records of the actual resource distribution, but also legitimized the emperor’s economic interest against other competing powerholders. As in the case of the Prince of Jin, the Hongwu emperor enfeoffed his sons as princes and established principalities in frontier regions to check the autonomous power of local military officers. ${}^{163}$${}^{163}$^(163){ }^{163} However, these princes were not always aligned with the emperor’s priorities but pursued their own interests. One of major issues was the limited availability of horse grazing lands. In 1397, the same year he became enraged toward the Prince of Jin, Hongwu warned that the livestock of his sons and sons-in-law should not encroach upon the lands assigned to local garrisons. The edict sent by the emperor was accompanied by an illustration of the northern and western frontiers to prevent the princes from taking over the lands that were designated for local soldiers’ farming, herding, and foraging for firewood, which could disrupt soldiers’ livelihoods. This illustration was particularly intended to “make my sons and descendants maintain the boundaries [enforced by him] in future generations.” ${}^{164}$${}^{164}$^(164){ }^{164}
地图不仅是实际资源分配的记录，还使皇帝的经济利益合法化，以对抗其他相互竞争的权力拥有者。与晋王世子的情况一样，洪武皇帝册封自己的儿子为亲王，并在边疆地区建立公国，以遏制地方武官的自主权力。 ${}^{163}$${}^{163}$^(163){ }^{163} 然而，这些亲王并不总是与皇帝的优先事项保持一致，而是追求自己的利益。其中一个主要问题是牧马的土地有限。1397 年，也就是洪武对金王子发怒的那一年，洪武警告他的儿子和女婿不得侵占分配给当地驻军的土地。皇帝在下达诏书的同时，还绘制了北疆和西疆的示意图，以防止皇子们侵占当地士兵耕种、放牧和拾柴的土地，以免扰乱士兵的生计。这一说明尤其是为了 "使我的子孙后代维护[他所强制执行的]疆界"。 ${}^{164}$${}^{164}$^(164){ }^{164}

While none of the above-mentioned maps survive, the remaining records suggest that the early Ming court’s information-gathering activities focused on managing its strategic resources, including soldiers, animals, agricultural colonies, and grazing lands. The emperors used maps to keep track of the changing military conditions and protect the economic base of frontier defense against other imperial agents’ profit-seeking activities, including those of imperial kinsmen. However, the main attention was devoted not to the territories per se but on the resources of particular frontier areas and who had rights over those resources. The next section analyzes how early fifteenth-century maps manifested a distinct conception of emperorship based on the regulated movement of goods, people, and information between the imperial center and peripheries, before the sovereignty became territorialized.
虽然上述地图无一存世，但剩余的记录表明，明初宫廷的信息收集活动主要集中在管理战略资源上，包括士兵、牲畜、农业殖民地和牧场。皇帝们使用地图来跟踪不断变化的军事情况，保护边防的经济基础，防止其他帝国代理人（包括皇亲国戚）的逐利活动。然而，主要关注的不是领土本身，而是特定边疆地区的资源以及谁对这些资源拥有权利。下一部分将分析十五世纪早期的地图是如何在主权领土化之前，在帝国中心与外围地区之间的货物、人员和信息流动管制的基础上，体现出一种独特的皇帝概念的。

During the late Hongwu and Yongle periods, the Ming court produced a number of courier route maps visualizing the journey of official travelers. These courier route maps demonstrate the court’s interest in extracting and circulating the resources using the official transportation network across the empire. In his book, Marco Polo introduces the Yuan courier system that enabled speedy and reliable communication across the far-flung empire.
在洪武晚期和永乐年间，明朝宫廷制作了许多信使线路图，将官方旅行者的行程形象化。这些信使线路图显示了朝廷对利用官方交通网络在帝国境内提取和流通资源的兴趣。马可-波罗在他的书中介绍了元朝的信使系统，该系统能够在遥远的帝国境内实现快速可靠的通信。

Mongolian) as a form of military training (or often an aristocratic sport). David Robinson points out that Prince Yan, the future Yongle Emperor, followed the steppe tradition in diplomacy and enjoyed hunting. For Ming royal hunting, see David Robinson, Martial Spectacles of the Ming Court, 28-84.
蒙古人）的一种军事训练形式（或通常是一种贵族运动）。大卫-罗宾逊（David Robinson）指出，未来的永乐皇帝燕亲王在外交方面沿袭了草原传统，并喜欢打猎。关于明朝皇家狩猎，见 David Robinson, Martial Spectacles of the Ming Court, 28-84。

Now you must know that from this city of Cambaluc [Yuan capital Dadu, at the site of modern Beijing] proceed many roads and highways leading to a variety of provinces, one to one province, another to another, and each road receives the name of the province to which it leads; and it is a very sensible plan. And the messengers of the Emperor in travelling from Cambaluc, be the road whichsoever they will, find at every twenty-five miles of the journey a station which they call Yamb [or Yam in Mongolian], or as we should say, the “Horse Post House.” And at each of those stations used by the messengers, there is a large and handsome building for them to put up at, in which they find all the rooms furnished with the fine beds and all other necessary articles in rich silk, and where they are provided with everything they can want. If even a king were to arrive at one of these, he would find himself well lodged. ${}^{165}$${}^{165}$^(165){ }^{165}
你们要知道，从坎巴鲁城[元朝都城大都，在今北京所在地]出发，有许多道路和公路通往各省，一条通往一个省，另一条通往另一个省，每条路都有它所通往的省的名字；这是一个非常合理的计划。皇帝的信使们从坎巴卢克出发，无论走哪条路，每走 25 英里就会发现一个驿站，他们把这个驿站称为 "羊布"（蒙古语为 Yam），或者我们应该说是 "马驿站"。在信使们使用的每一个驿站，都有一幢宽敞漂亮的楼房供他们住宿，房间里摆放着精美的床铺和所有其他必需品，用的都是富丽堂皇的丝绸，他们在那里可以得到想要的一切。即使是国王来到这里，也会发现自己住得很好。 ${}^{165}$${}^{165}$^(165){ }^{165}

The Ming empire reinstituted the courier system in 1368. ${}^{166}$${}^{166}$^(166){ }^{166} These transportation networks were used for official communication, linking the capital to local civilian and military units around the empire. By means of the courier stations, the Ming court also transported necessary military supplies to frontier garrisons. ${}^{167}$${}^{167}$^(167){ }^{167} The same routes also served for diplomatic purposes when foreign envoys arrived and traveled to the Ming court.
1368 年，明帝国恢复了信使制度。 ${}^{166}$${}^{166}$^(166){ }^{166} 这些交通网络用于官方通信，将首都与帝国各地的地方文职和军事单位联系起来。通过信使驿站，明朝宫廷还将必要的军事物资运送到边疆驻军。 ${}^{167}$${}^{167}$^(167){ }^{167} 同样的路线也用于外交目的，当外国使节抵达明朝并前往明朝宫廷时。

The early Ming court’s production of courier route maps and books indicates its efforts to supervise official transportation routes. In the previous section, Yongle emperor’s edict requested military officers to report “the travel distances by road” in addition to updated military registers. The early Ming court regularly updated information about courier stations. Published in 1394, the first official route book, titled Major Routes of the Realm (Huanyu tongqu 寰宇通衢),
明初朝廷制作的信使路线地图和书籍表明了其对官方交通路线的监督力度。在上一节中，永乐皇帝下诏要求武官除更新军籍外，还要上报 "路途远近"。明初朝廷定期更新信使驿站的信息。1394 年，第一部官方路线书《环宇通衢》出版、

describes the routes from the imperial capital of Nanjing to the thirteen provincial seats, and from each provincial seat to smaller civilian administrative units (prefectures and counties) and military garrisons. ${}^{168}$${}^{168}$^(168){ }^{168} The Huanyu tongqu’s organizing principles and contents reflect its restricted usage for the court’s management of courier routes. Corresponding to a vertical bureaucratic hierarchy, each section is organized by the route information from the capital to each provincial seat, which is followed by those from each provincial seat to lower-order administrative units belonging to the same province. ${}^{169}$${}^{169}$^(169){ }^{169} This book excludes any other horizontal travel routes among administrative units, such as those between provinces or lower administrative units, bypassing their immediate superior (either the capital or the provincial center).
${}^{168}$${}^{168}$^(168){ }^{168} 《环宇通衢》描述了从帝都南京到十三省治所，以及从各省治所到较小的民事行政单位（府、县）和驻军的路线。 ${}^{168}$${}^{168}$^(168){ }^{168} 《环宇通衢》的组织原则和内容反映了它在朝廷管理信使路线方面的局限性。与垂直的官僚等级制度相对应，每个部分都按从京城到各省治所的路线信息编排，然后是各省治所到同属一个省的下级行政单位的路线信息。 ${}^{169}$${}^{169}$^(169){ }^{169} 本书不包括行政单位之间的任何其他横向旅行路线，例如省或下级行政单位之间绕过其直接上级（首都或省中心）的旅行路线。

Huanyu tongqu’s administratively determined travel networks contrast with those of later route books commercially published beginning in the late sixteenth century. The later books incorporate new route information for increasingly popular private travel, prioritizing routes between commercial and transport hubs instead of administrative centers (though these two sometimes overlapped with each other). ${}^{170}$${}^{170}$^(170){ }^{170} In addition, while Huanyu tongqu only includes the names of courier stations and the distance between them, commercial route books incorporate pragmatic knowledge for actual travel, such as “the availability and affordability of lodging, the conditions of harbors and fords, the sources of commodities, the quality of road maintenance, the navigability of rivers and public security.” ${}^{171}$${}^{171}$^(171){ }^{171} Moreover, Huanyu tongqu’s restricted circulation suggests that it was not written for and used by actual travelers, but instead for the court officials who supervised the empire-wide courier networks.
环宇通衢》中由行政部门确定的旅行网络与后来从 16 世纪晚期开始商业化出版的线路书形成了鲜明对比。后来的书籍为日益流行的私人旅行提供了新的路线信息，优先考虑商业和交通枢纽之间的路线，而不是行政中心（尽管这两者有时会相互重叠）。 ${}^{170}$${}^{170}$^(170){ }^{170} 此外，《环宇通衢》只包括信使驿站的名称和它们之间的距离，而商业路线书籍则纳入了实际旅行的实用知识，如 "住宿的可用性和可负担性、港口和岔道的条件、商品的来源、道路维护的质量、河流的可通航性和社会治安"。 ${}^{171}$${}^{171}$^(171){ }^{171} 此外，《环宇通衢》的发行范围有限，这表明它不是为实际旅行者撰写和使用的，而是为监管帝国范围内信使网络的朝廷官员撰写的。

Similar to the early Ming route books, route maps focus on official transportation networks mainly for the limited court audience instead of for actual travelers, regardless of whether or not their travel was official. My analysis is based on the following four route maps currently held in National Palace of Museum, Taipei (NPM).
与明朝早期的路书类似，路书的重点是官方交通网络，主要面向有限的宫廷受众，而不是真正的旅行者，无论他们的旅行是否官方性质。我的分析基于台北故宫博物院（NPM）现藏的以下四幅路线图。

Anonymous, “Route Map from Nanjing to Gansu” (Nanjing zhi gansu yipu tu南京至甘肅驛舖圖；figure 21），early ${15}^{\text{th}}$${15}^{\text{th }}$15^("th ")15^{\text {th }} century， color on paper，handscroll， $55\ast 2432$$55\ast 2432$55**243255 * 2432 cm, NPM. ${}^{172}$${}^{172}$^(172){ }^{172}
佚名，《南京至甘肅驛舖圖》（《南京志-甘肅驛舖圖》）， ${15}^{\text{th}}$${15}^{\text{th }}$15^("th ")15^{\text {th }} 世紀初，紙本彩色，手卷， $55\ast 2432$$55\ast 2432$55**243255 * 2432 厘米，故宮博物院藏。 ${}^{172}$${}^{172}$^(172){ }^{172}

Anonymous, “Route Map from Taiyuan to Gansu” (Taiyuan zhi gansu yipu tu太原至甘肅驛舖圖; figure 22)", color on paper, handscroll, $54.5\ast 664\mathrm{cm},\mathrm{NPM}.{}^{173}$$54.5\ast 664\mathrm{cm},\mathrm{NPM}.{}^{173}$54.5**664cm,NPM.^(173)54.5 * 664 \mathrm{~cm}, \mathrm{NPM} .{ }^{173} Anonymous, “Route Map from Yuezhou to Longzhou” (Yuezhou zhi longzhou yipu tu岳州至龍州驛舖圖; figure 24 and 25), color on paper, handscroll, $54.5\ast 1851\mathrm{cm}$$54.5\ast 1851\mathrm{cm}$54.5**1851cm54.5 * 1851 \mathrm{~cm}, NPM.
图 22）"，纸本彩色手卷， $54.5\ast 664\mathrm{cm},\mathrm{NPM}.{}^{173}$$54.5\ast 664\mathrm{cm},\mathrm{NPM}.{}^{173}$54.5**664cm,NPM.^(173)54.5 * 664 \mathrm{~cm}, \mathrm{NPM} .{ }^{173} 佚名，《岳州至陇州路线图》（《岳州志-陇州驿图》岳州至陇州驛舖圖；图 24 和 25），纸本彩色手卷， $54.5\ast 1851\mathrm{cm}$$54.5\ast 1851\mathrm{cm}$54.5**1851cm54.5 * 1851 \mathrm{~cm} ，故宫博物院。
Anonymous, “Map of Four Courier Routes in Sichuan” (Sichuan sheng silu guanyi tu四川省四路關驛圖; figure 26 and 27), color on paper, handscroll, 55*1085 cm, NPM.
佚名，《四川省四路关驛图》，纸本设色，手卷，55*1085 厘米，故宫博物院藏。

Although we cannot assume that these four maps are representative examples of all the courier maps that the court ever produced, the surviving specimens reveal the court’s interest in maintaining official communication with its frontiers (figure 28). The “Route Map from Nanjing to Gansu” depicts the route from the capital of Nanjing to Suzhou garrison in western Gansu, on the northwestern frontier. The other three maps focus more on communication within frontier zones. The “Route Map from Taiyuan to Gansu” illustrates the routes from Taiyuan (Shanxi provincial seat) to Datong garrison (seat of the Shanxi military commission) and from Tiancheng to Dongsheng garrison, which is one of the forward garrisons established in the Ordos plateau. ${}^{174}$${}^{174}$^(174){ }^{174} The “Route Map from Yuezhou to Longzhou,” from Hunan province to the Ming southwestern
虽然我们不能认为这四幅地图是朝廷制作的所有信使地图的代表作，但现存的标本揭示了朝廷对保持与边疆官方联系的兴趣（图 28）。南京至甘肃路线图 "描绘了从首都南京到西北边陲甘肃西部肃州驻军的路线。其他三幅地图更侧重于边疆地区内的交通。从太原到甘肃的路线图 "绘制了从太原（山西省省会）到大同驻军（山西军委所在地）以及从天城到东胜驻军的路线，东胜驻军是在鄂尔多斯高原建立的前沿驻军之一。 ${}^{174}$${}^{174}$^(174){ }^{174} 《岳州至陇州路线图》，从湖南省到明西南部
these route maps depicted the routes before the capital was relocated to Beijing. Xi, Zhongguo Gudai Ditu Wenhua Shi.
这些路线图描绘的是迁都北京之前的路线。僖宗，《中古度度文华录》。
${}^{173}$${}^{173}$^(173){ }^{173} These map titles have been added by a museum curator. Based on my analysis, the title of “Route Map from Taiyuan to Gansu” is misleading, since the depicted area is limited within Shanxi Province (figure 23). The name is probably based on the textual annotation at the left end of the handscroll, saying “toward the west, two thousand li to Ningxia” (西至寧夏二千里) to indicate the direction, without including the actual route, to Ningxia.
${}^{173}$${}^{173}$^(173){ }^{173} 这些地图标题是由博物馆馆长添加的。根据我的分析，"太原至甘肃线路图 "的标题有误导之嫌，因为所绘区域仅限于山西省境内（图 23）。该名称可能是根据手卷左端的文字注释 "向西，至寧夏二千里"（西至寧夏二千里）来表示方向的，并没有包括实际的路线。
${}^{174}$${}^{174}$^(174){ }^{174} The Dongsheng garrison was abandoned as a result of the retreat of the forward military garrisons to reduce costs during the late Yongle reign. Later Mongols entered this area.
${}^{174}$${}^{174}$^(174){ }^{174} 永乐末年，为降低成本，前沿驻军撤退，东胜驻军被废弃。后来蒙古人进入这一地区。
border with Vietnam, reflects the Ming court’s interest in supplying military provisions through a river route in preparation for the Ming invasion of Vietnam launched in 1406. ${}^{175}$${}^{175}$^(175){ }^{175} Lastly, the “Map of Four Courier Routes in Sichuan” includes four routes within Sichuan province, all radiating from its provincial seat of Chengdu. Among them, two routes go to the Ming-Tibetan frontier in the west, and the other two toward its northern provincial boundary with Gansu. This map also incorporates a large number of tea producing mountains (chashan 茶山) and tea-tax stations (chaguan 茶關), suggesting the map’s intended purpose either for tax collection or management of the tea-horse trade routes leading toward Tibet and Central Asia (figure 27). ${}^{176}$${}^{176}$^(176){ }^{176}
${}^{175}$${}^{175}$^(175){ }^{175} 最后，《四川四路信使图》包括四川省内的四条信使路线，均从省城成都出发。 ${}^{175}$${}^{175}$^(175){ }^{175} 最后，《四川四路信使图》包括四川省内的四条路线，均从省城成都辐射出去。其中，两条通往西部的明藏边境，另外两条通往北部与甘肃的省界。这幅地图还包含了大量的产茶山（茶山）和茶关（茶关），表明该地图的用途是征税或管理通往西藏和中亚的茶马贸易路线（图 27）。 ${}^{176}$${}^{176}$^(176){ }^{176}

The focus of all these route maps is on official courier networks, appearing as a red line more or less at the center connecting each courier station. All these maps were made on handscrolls, which are limited vertically (all four of them are approximately the same height of 54.5 to 55 cm ) but extendable horizontally without limit (the longest is 24.32 meter in width). Within a map, compilers frequently changed the orientations in order to present an entire courier route on a single horizontally-extendable surface. As a result, viewers cannot see relative locations among places or the actual shape of the empire. Instead, place names appear as stages
所有这些线路图的重点都是官方快递网络，或多或少以一条红线为中心连接每个快递站。所有这些地图都是在手卷上绘制的，手卷的垂直高度有限（所有四幅手卷的高度大致相同，均为 54.5 至 55 厘米），但可无限制地水平延伸（最长的手卷宽度为 24.32 米）。在一张地图中，编纂者经常改变方向，以便在一个可水平延伸的平面上展示整条快递路线。因此，观众无法看到各地点之间的相对位置或帝国的实际形状。相反，地名显示为阶段性的

on a linear travel route. In her analysis of Tokugawa Japan’s route maps, Kären Wigen calls such maps “diagrammatic itineraries,” which delineate one or more transportation routes without considering actual geographic shape or distance between places. As one diagrammatic itinerary, Torigai Dōsai’s 鳥飼洞斎（1721-1793）“Handy Guide to the Roadways of Great Japan” (jp. Dai Nihon dōch ū kōtei saiken ki 大日本道中行程細見記; figure 29 and 30) of 1770 depicted multiple travel routes from Edo as a series of simple and parallel lines. ${}^{177}$${}^{177}$^(177){ }^{177} Similarly, early Ming route maps transformed a courier route into the simpler pictorial form of a line, with each map focused on the representation of a single courier route instead of incorporating multiple routes in the same plate as in Dōsai’s map. ${}^{178}$${}^{178}$^(178){ }^{178}
线性旅行路线。凯伦-维根（Kären Wigen）在分析德川时代的日本路线图时，将这种地图称为 "图解路线"，它划定了一条或多条交通路线，而不考虑实际的地理形状或地方之间的距离。鸟饲洞斋（Torigai Dōsai）的《大日本道路指南》（Handy Guide to the Roadways of Great Japan）（jp.大日本道中行程细见记》（Dai Nihon dōch ū kōtei saiken ki 大日本道中行程细见记；图 29 和 30）描绘了从江户出发的多条旅行路线，如同一系列简单而平行的线条。 ${}^{177}$${}^{177}$^(177){ }^{177} 同样，早期的明代路线图将信使路线转化为更简单的线条图像形式，每幅地图都侧重于表现一条信使路线，而不是像道中的地图那样将多条路线纳入同一版图。 ${}^{178}$${}^{178}$^(178){ }^{178}

Early Ming official route books and route maps must have had a close relation in the production stage. In order to produce route maps, the complete textual information has to be gathered and organized according to stages on route. For instance, an unfinished route map (Wuzi yipu tu 無字驛舖圖; figure 31) is a draft map in a similar style with other route maps, with rectangular boxes in the map left blank without texts. Although it is unknown why this map remains unfinished, it is a valuable example for showing the process of producing route maps.
明朝早期的官方路线书和路线图在制作阶段一定有着密切的关系。为了绘制线路图，必须收集完整的文字资料，并按照线路的不同阶段进行整理。例如，一幅未完成的线路图（《无字驛铺图》，图 31）是一幅草图，风格与其他线路图相似，图中的长方形方框留白，没有文字。虽然不知道这幅地图为何仍未完成，但它是展示线路图制作过程的珍贵范例。

Once the route information had been collected, court painters drew an illustration, and either painters or other officials filled out blank boxes with textual information, such as the names of stations and the distances between stations.
收集路线信息后，宫廷画师绘制插图，画师或其他官员在空白框内填写文字信息，如车站名称和车站之间的距离。

If these route maps simply visualized the textual information that had already been compiled in route books, what were the differences between maps and books in terms of their functions and purposes? Compared to the book format, a handscroll suffers from several restrictions in production, usage, and circulation. Creating a painting several meters long as in route maps requires much more time, cost, and specialized skill. Once drawn, handscrolls are also not easy to update or modify. In the case of a book, an editor can easily update the information or correct errors by unbinding the book, replacing the old page with a new one, and rebinding it. In contrast, when a route was changed, a scroll required a more extensive revision by cutting off and redrawing an entire section, if not the entire scroll. ${}^{179}$${}^{179}$^(179){ }^{179} For users, a book requires a smaller surface space for reading compared to a handscroll that needs a much larger surface to see even a section. It is also more convenient to find information in a book, by flipping through the pages and skimming through section titles of province names that the travel destination would belong to. ${}^{180}$${}^{180}$^(180){ }^{180} In contrast, a several-meter-long painting is unwieldy for such reference use unless a viewer intends to see the entire route. In order to find the information that they want, the viewers should unroll the scroll from the beginning. Considering its extensive width, it is highly unlikely that the entire map was ever displayed. Since the limited surface
如果这些线路图只是将已编入线路书的文字信息视觉化，那么地图和书籍在功能和目的上有什么区别？与书籍的形式相比，手卷在制作、使用和流通方面受到诸多限制。绘制一幅几米长的地图需要更多的时间、成本和专业技能。手卷一旦绘制完成，也不容易更新或修改。如果是一本书，编辑只需拆开装订，用新页面替换旧页面，然后重新装订，就能很容易地更新信息或纠正错误。相比之下，如果要更改路线，则需要对卷轴进行更广泛的修改，即使不是整个卷轴，也要截去整个部分并重新绘制。 ${}^{179}$${}^{179}$^(179){ }^{179} 对于用户来说，书籍需要的阅读空间更小，而手卷则需要更大的空间才能看到哪怕是一个部分。此外，在书中查找信息也更加方便，只需翻阅书页，略读旅行目的地所属省名的章节标题即可。 ${}^{180}$${}^{180}$^(180){ }^{180} 相比之下，一幅几米长的油画对于这种参考用途来说并不方便，除非观众打算看完整条路线。为了找到想要的信息，观众需要从头开始展开画卷。考虑到其宽度，整幅地图被展示出来的可能性很小。因为有限的表面

allows viewers to see only a section at a time, the viewers also cannot have a quick glimpse of the entire route information.
观众每次只能看到一个路段，也无法快速浏览整个路线信息。

Considering all these factors, these maps are costly to produce and inconvenient to be used as a reference tool. Such characteristics become even more obvious when we compare these early Ming route maps with later maps inserted in commercial route books for referential use. Two double-folio maps are included at the beginning of Comprehensive Routes in Maps and Notes (Yitong lucheng tuji 一統路程圖集)，which are respectively entitled “Map of Routes from Beijing to the Thirteen Provinces and Frontiers” (Beijing zhi shisansheng gebian lutu北京至十三省各邊路圖) and “Those from Nanjing” (Nanjing zhi shisansheng gebian lutu 南京至十三省各邊路圖). These two maps are schematic diagrams of the empire-wide transportation networks (figure 32-1 and 32-2). ${}^{181}$${}^{181}$^(181){ }^{181} “Those from Nanjing” draw transportation routes on the background image of the Ming empire, which help readers to easily recognize the spatial relations among local places. ${}^{182}$${}^{182}$^(182){ }^{182} Such maps can serve as an effective visual summary of empire-wide travel routes. The readers can also easily go back and forth to cross-check between the map and the textual information of the routes included in the same book. While early Ming route map handscrolls likely were based on textual information or a route book, once they were
考虑到所有这些因素，这些地图的制作成本很高，而且不便用作参考工具。当我们将这些早期的明代路线图与后来插在商业路线书中用作参考的地图进行比较时，这些特点就更加明显了。两幅双一统路程图集》开头收录了两幅对开地图，分别题为 "北京至十三省边疆路程图 "和 "北京至十三省边疆路程图"。北京至十三省各边路图》和《南京至十三省各边路图》。这两幅地图是全帝国交通网络的示意图（图 32-1 和 32-2）。 ${}^{181}$${}^{181}$^(181){ }^{181} "南京人 "在明帝国的背景图上绘制了交通路线，这有助于读者轻松识别各地的空间关系。 ${}^{182}$${}^{182}$^(182){ }^{182} 这种地图可以有效地直观概括整个帝国的交通路线。读者还可以很方便地来回对照地图和同一本书中的路线文字信息。虽然明朝早期的路线图手卷很可能是根据文字信息或路线书绘制的，但一旦它们被

produced, it is unlikely that they would be paired with the book as a reference. Instead, these route maps were more likely viewed independently on a separate occasion from the book reading. If it was not for a reference, who used these route maps for what occasions?
因此，这些地图不太可能与书籍搭配起来作为参考。相反，这些线路图更可能是在阅读书籍之外的另一个场合独立观看的。如果不是作为参考，那么是谁在什么场合使用这些线路图呢？

The route maps’ contents, painting styles and physical features allow us to speculate about the intended purposes. As for the contents, the route maps include additional information about official sites, including town walls, office buildings, royal tombs and state-authorized temples, such as city god temples (chenghuang miao 城隍廟), altars for soil and grain (sheji tan社稷壇), and round mound altars (yuanqiu 圓丘). The highly standardized images of these buildings are similar to depictions of cities in gazetteer maps, which symbolize the Ming state’s control over local areas. Such official geography is incorporated into a panoramic, boundless view of the vast expanse of the imperial realm. Borrowing from the conventional landscape painting style, the route maps provide a visually pleasing experience of an imagined journey across courier routes. Other than the components of official geography, a number of landscape markers are included, such as mountains, hills, rivers, smaller streams, and lakes. Some of them are named and can serve as additional geographic references in addition to the names of courier stations and administrative units. Stylistically, painters used the blue-and-green coloring (qinglü靑綠）， which was traditionally associated with the land of immortals and the idea of a distant ideal past. ${}^{183}$${}^{183}$^(183){ }^{183} The maps’ idealized landscape depictions are likely intended to express the
通过线路图的内容、绘画风格和实物特征，我们可以推测其预期目的。在内容方面，线路图包含了官方遗址的额外信息，包括城墙、办公楼、皇家陵墓和国家授权的寺庙，如城隍庙、社稷坛和圆丘坛。这些高度标准化的建筑形象类似于地名图中对城市的描绘，象征着明朝国家对地方的控制。这种官方地理学被纳入了一幅全景图中，无边无际地展现了帝国领域的广袤无垠。借用传统山水画的风格，路线图提供了一种穿越信使路线的视觉享受。除了官方地理的组成部分，地图中还包括许多景观标记，如山脉、丘陵、河流、小溪流和湖泊。其中一些还被命名，除了快递站和行政单位的名称外，还可以作为额外的地理参考。在风格上，画家们使用了青绿色彩（qinglü靑绿），这在传统上与仙境和遥远的理想过去有关。 ${}^{183}$${}^{183}$^(183){ }^{183} 地图上理想化的风景描绘很可能是为了表达 "仙境 "的概念。

magnificence of the imperial space. Decorative mountains and streams also make the scroll space less monotonous and more dynamic. Along with the red-colored travel route, a series of mountains, often overlapping horizontally or appearing in a sequence, provide additional visual guidance for viewers to follow the footsteps of an imagined traveler. ${}^{184}$${}^{184}$^(184){ }^{184}
帝王空间的壮丽。装饰性的山峦和溪流也使画卷空间不再单调，更具动感。伴随着红色的旅行路线，一系列山峦往往水平重叠或依次出现，为观众提供了额外的视觉引导，让他们跟随想象中旅行者的脚步。 ${}^{184}$${}^{184}$^(184){ }^{184}

However, unlike landscape paintings, route maps are not primarily intended for the visual appreciation of the depicted landscape. Instead, the major attention is paid to the routes and administrative geography, while most landscape features remain unidentified, appearing in the background as highly standardized images. Mountains are miniaturized, appearing in the foreground or background spaces and rarely intersecting with official buildings or routes in the middle ground. In comparison with landscape features, official buildings, especially town walls, occupy a significant space at the center of the map. Such emphasis contrasts with conventional landscape paintings, where natural features often dominate most of the space while man-made structures are either concealed amidst forest or miniaturized. In this respect, these route maps are similar to city maps in gazetteers, which depict towns surrounded by but still separate from the mountains. These city maps provide a clear view of town walls and office buildings in an enlarged size located at the center of the map, while placing miniaturized mountains at the margins as background features. The mountains in the route maps are almost always spatially
然而，与风景画不同的是，线路图的主要目的并不是对所描绘的风景进行视觉欣赏。相反，主要关注的是路线和行政地理，而大多数景观特征仍未确定，只是作为高度标准化的图像出现在背景中。山脉被小型化，只出现在前景或背景空间，很少与官方建筑或中间的路线相交。与山水画相比，官方建筑，尤其是城墙，在地图中心占据了很大的空间。这种强调与传统山水画形成鲜明对比，在传统山水画中，自然景观往往占据大部分空间，而人工建筑要么被掩盖在森林之中，要么被缩小。在这方面，这些线路图与地名录中的城市地图相似，后者描绘的城镇被群山环绕，但仍与群山分离。在这些城市地图中，位于地图中心的城墙和办公楼被放大，一览无余，而微型化的山脉则作为背景特征被置于地图边缘。线路图中的山脉几乎总是在空间上
green colors in Chinese painting tradition. see Quincy Ngan, “The Materiality of Azurite Blue and Malachite Green in the Age of the Chinese Colorist Qiu Ying (ca. 1498 - ca. 1552)” (PhD diss., University of Chicago, 2016).
见 Quincy Ngan，"The Materiality of Azurite Blue and Malachite Green in the Age of Chinese Colorist Qiu Ying (ca. 1498 - ca. 1552)"（芝加哥大学博士论文，2016 年）。
${}^{184}$${}^{184}$^(184){ }^{184} For the notion of “armchair travel” in the late Ming, see Si-yen Fei, Negotiating Urban Space: Urbanization and Late Ming Nanjing, 124-56, which examines topographical illustrations of Nanjing in tour guides compiled during the sixteenth and seventeenth centuries.
${}^{184}$${}^{184}$^(184){ }^{184} 关于晚明 "扶手椅旅行 "的概念，参见费思彦，《协商城市空间》：该文研究了 16 世纪和 17 世纪编纂的旅游指南中的南京地形图。
separated from the structures at the center in order to allow the viewers’ maximum visibility of the town without any concealment by the natural landscape.
这些建筑与中心的建筑分开，以便让观赏者最大限度地看到小镇，而不会被自然景观所掩盖。

Combining both the landscape painting and city map conventions, route maps visualize the Ming as a spatially coherent unit interconnected by its official communication network. Maps recreate courier stations, towns, and villages as nodes of official communication as well as points along a journey, each of which is embedded in bucolic and idealized landscapes. The standardized depictions of both official buildings and landscape features enhance the idea of spatial coherence across a geographically vast empire. In frontiers, some features are more frequently found, including beacon towers and passes, and mountains whose peaks were painted with white color that hint at high elevations. However, in general, frontier areas are not depicted as exotic or foreign, and it is difficult to differentiate them from the internal regions of the empire without looking at actual place names. Any cultural or ethnic references are missing, which contrasts with their more frequent incorporation in later fifteenth and sixteenth century maps depicting frontiers. ${}^{185}$${}^{185}$^(185){ }^{185} At the end of the map, the boundaries between Ming and the outside world remain unclear, as if the same landscape features would continue to appear beyond the scroll. Only the discontinued route and administrative buildings indicate the end of both the journey and imperial rule.
结合山水画和城市地图的惯例，线路图将明朝视觉化为一个由官方通信网络相互连接的空间连贯单元。地图将信使驿站、城镇和村庄再现为官方通信的节点以及旅程中的各个点，每个点都被嵌入田园风光和理想化的风景中。对官方建筑和景观特征的标准化描绘增强了地理上庞大帝国空间一致性的概念。在边疆地区，一些地貌特征更为常见，包括烽火台和关隘，以及山峰被涂成白色，暗示着海拔较高。然而，总体而言，边疆地区并没有被描绘成异域或外国地区，如果不查看实际地名，很难将其与帝国内部地区区分开来。地图中没有任何文化或种族的提法，这与十五世纪后期和十六世纪描绘边疆的地图中更频繁地出现这些提法形成了鲜明对比。 ${}^{185}$${}^{185}$^(185){ }^{185} 在地图的末尾，明朝与外部世界的边界仍然不清晰，似乎同样的地貌特征会在卷轴之后继续出现。只有中断的路线和行政建筑表明了旅程和帝国统治的结束。

The handscroll format is particularly useful to facilitate viewers’ imagining of travel across the empire based on the court’s regulated routes. As a typical handscroll painting, these maps were likely stored rolled and only unrolled when they were viewed. At the beginning of the scrolls at the right end (since it is viewed from right to left), the maps’ journey originates from
手卷的形式尤其有助于观众根据宫廷规定的路线想象穿越帝国的旅行。作为典型的手卷绘画，这些地图很可能是卷着存放的，只有在观看时才会展开。在卷轴右端的开头（因为是从右向左看），地图的旅程从

the capital (Nanjing), the provincial seat (Taiyuan in Shanxi or Chengdu in Sichuan) acting as the representatives of the imperial court at the provincial level, or the city more geographically proximate to the capital (Yuezhou) toward the destination farther away (Longzhou). Due to its unwieldy size, viewers can only unroll a section of the scroll at a time. After finishing the viewing of one section, viewers can continue rolling and unrolling and proceed until they reach the end of the scroll. This procession imitates the movement of official travel, from the imperial center(s) toward the peripheries. ${}^{186}$${}^{186}$^(186){ }^{186} Once reaching the destination at the end of the scroll, the viewers reroll the entire handscroll from left to right, reviewing once again the map in reverse order. This reversed movement simulates official travel back from the peripheries to the center.
在这幅画卷中，画卷的内容包括：京城（南京）、作为朝廷省级代表的省城（山西太原或四川成都），或距离京城较近的城市（越州）以及距离京城较远的目的地（龙州）。由于卷轴体积庞大，观众每次只能展开其中的一部分。看完一段后，观众可以继续滚动和展开，直到卷轴的末尾。这一过程模仿了官方旅行的移动方式，即从皇室中心向外围移动。 ${}^{186}$${}^{186}$^(186){ }^{186} 到达卷轴末尾的目的地后，观众从左到右重新卷起整张手卷，以相反的顺序再次查看地图。这种反向运动模拟了从外围回到中心的正式旅行。

In other words, the handscroll route map is a powerful medium to imitate and rehearse the journey for the intended viewers supposedly sitting at the imperial center. Imagined imperial touring was an important political tradition of defining the imperial space and envisioning the sovereignty. The legendary figure Yu the Great’s touring upon delivering humanity from the Great Flood in high antiquity is a well-known example. Through his travels, Yu the Great not only pacified the unruly landscape, but also established a permanent basis for imperial rule by demarcating the boundaries of the Nine Provinces (jiuzhou) and determining the
换句话说，手卷线路图是一种强大的媒介，可以为假定坐在帝国中心的目标观众模仿和演练旅程。想象中的帝王巡游是界定帝王空间和憧憬主权的重要政治传统。传说中的大禹治水就是一个著名的例子。通过巡游，大禹不仅平定了桀骜不驯的山水，还划定了九州的疆界，确定了帝国统治的永久基础。

types of tribute that each province has to submit to the court. ${}^{187}\mathrm{Yu}$${}^{187}\mathrm{Yu}$^(187)Yu{ }^{187} \mathrm{Yu} the Great’s travels, described as a “downward” movement (xia 下) from the imperial center to peripheries, is followed by the reversed “upstream” movements (shang 上) of local tribute, symbolizing the political submission of the localities to the center. The route map handscrolls allow viewers, most likely the emperor and his close advisors, to recreate similar movements in both directions between the center and periphery. The constant and regulated flow of information, goods, and people following the courier routes centered on the emperor residing at the capital represent imperial sovereignty over the depicted spaces. ${}^{188}$${}^{188}$^(188){ }^{188}
各省必须向朝廷提交的贡品种类。 ${}^{187}\mathrm{Yu}$${}^{187}\mathrm{Yu}$^(187)Yu{ }^{187} \mathrm{Yu} 大帝的旅行被描述为从帝国中心到外围的 "下行 "运动，紧随其后的是地方进贡的反向 "上行 "运动，象征着地方对中心的政治服从。通过路线图手卷，观众（很可能是皇帝和他的亲信）可以再现中央和外围之间类似的双向流动。以居住在首都的皇帝为中心，信息、货物和人员按照信使路线持续有序地流动，体现了帝国对所描绘空间的主权。 ${}^{188}$${}^{188}$^(188){ }^{188}

In summary, my analysis of route maps demonstrates the following three aspects: first, together with route books, route maps served their limited court audience as a means of enhancing the court’s supervision over courier routes. Considering the cost and specialized skill required for their production, the route maps’ circulation would be more restricted than that of official route books, possibly strictly for the emperor and immediate officials. Second, on the route maps, the standardized and idealized depiction of administrative towns, buildings, and landscape features represent the imperial space as homogeneous, coherent, and interconnected by
总之，我对路线图的分析表明了以下三个方面：首先，路线图与路线书一起，作为加强朝廷对信使路线监督的一种手段，为其有限的朝廷受众服务。考虑到其制作成本和所需的专业技能，路线图的流通范围比官方路线书更有限，可能仅限于皇帝和直属官员。其次，在线路图上，行政城镇、建筑和景观特征的标准化和理想化描绘代表了帝国空间的均质性、连贯性和相互关联性。
${}^{187}$${}^{187}$^(187){ }^{187} Also see the explanation of Yugong geography in chapter one of this dissertation. The legend of Yu the Great’s survey and demarcation of boundaries played an important role in defining the trans-temporal imperial space.
${}^{187}$${}^{187}$^(187){ }^{187} 参见本论文第一章中对禹贡地理的解释。大禹治水勘界的传说对确定跨时空的帝国空间起了重要作用。
${}^{188}$${}^{188}$^(188){ }^{188} The recent book by Aurelia Campbell explores a similar issue from the perspective of architectural history, examining how the Yongle emperor established his authority over the geographically far-reaching empire through collecting, extracting, and circulating the goods and people. Aurelia Campbell, What the Emperor Built: Architecture and Empire in the Early Ming (Seattle: University of Washington Press, 2020).
${}^{188}$${}^{188}$^(188){ }^{188} 奥瑞莉娅-坎贝尔（Aurelia Campbell）最近的著作从建筑史的角度探讨了类似的问题，研究了永乐皇帝如何通过收集、提取和流通货物和人口，在地域辽阔的帝国中树立权威。奥莱丽亚-坎贝尔（Aurelia Campbell），《皇帝建造了什么？明初的建筑与帝国》（西雅图：华盛顿大学出版社，2020 年）。
the court’s regulated routes. Third, the route maps symbolized the imperial sovereignty over depicted areas, not only for the purpose of extracting and circulating goods, people, and information, but also for possessing geographical information, rearranging it in an orderly fashion, and recreating the regulated course of movement across the empire. Maps were a microcosm of the imperium, and the act of unrolling the scrolls reenacted the progress of imperial power through an imagined journey from the center to the periphery, and from the peripheries toward the capital.
朝廷规定的路线。第三，路线图象征着帝国对所描绘地区的主权，这不仅是为了提取和流通货物、人员和信息，也是为了掌握地理信息，以有序的方式对其进行重新安排，并再现整个帝国的规范移动路线。地图是帝国的缩影，展开卷轴的行为通过从中心到外围、从外围到首都的想象之旅，再现了帝国权力的进程。

In contrast to the early fifteenth century maps that focused mainly on resource management, the later maps placed new emphasis on visualizing spatial relations of military facilities. Such maps provided visual aids for the court’s understanding of frontier spaces based not just on the vertical bureaucratic order between superior garrisons and inferior forts, but also on horizontal relations among neighboring units, which had to coordinate military actions. In addition, maps promoted imperial claims over contested territories through invoking the memory of Ming military conquest and (re)inscribing local places as part of the Ming imperium.
15 世纪早期的地图主要侧重于资源管理，与此不同的是，后期的地图将新的重点放在了军事设施空间关系的可视化上。这些地图为朝廷理解边境空间提供了视觉辅助，其基础不仅是上级驻军和下级堡垒之间的纵向官僚秩序，还有必须协调军事行动的相邻单位之间的横向关系。此外，地图还通过唤起对明朝军事征服的记忆，并（重新）将当地描述为明朝统治的一部分，从而促进了朝廷对有争议领土的要求。

In the late fifteenth century, the Ming found itself besieged by hostile enemies on its northwestern frontiers. The geopolitics of the moment are reflected in Zhu Ying’s 朱英 (14171486) memorial on “how to guard the border (預計安邊疏),” submitted when he was serving as Gansu Grand Coordinator between 1474 and 1475. Zhu wrote,
十五世纪末，明朝发现自己被西北边境的敌军围困。朱英（1417-1486 年）在 1474 年至 1475 年担任甘肃巡抚时提交的 "预计安边疏 "奏折反映了当时的地缘政治。朱英写道

In my opinion, Gansu undoubtedly is the most important strategic area among the three major frontier garrisons of Shaanxi [Yansui 延綏, Ningxia 寧夏, and Gansu]. Located
在我看来，甘肃无疑是陕西三大边防军（延绥、宁夏、甘肃）中最重要的战略要地。位于
across the [Yellow] River on the other side of Lan County 蘭縣, ${}^{189}$${}^{189}$^(189){ }^{189} on the south this area faces the Western Fan barbarians, who rely on the difficult terrain of the Snow Mountains; on the north, it confronts the Northern $Hu$$Hu$HuH u barbarians, who march from the desert. Our territories lie between these two as one long thread extending more than two thousand li. The fortresses of Ganzhou 甘州, Liangzhou 凉州, and other garrisons stand isolated in remote areas, cut off from the rest of the empire… ${}^{190}$${}^{190}$^(190){ }^{190}
${}^{189}$${}^{189}$^(189){ }^{189} 南面是依托雪山险要地形的西蕃蛮族，北面是从大漠进军的北蛮 $Hu$$Hu$HuH u 。我们的领土就在这两者之间，绵延两千余里。甘州要塞、凉州要塞和其他驻军孤立于偏远地区，与帝国其他地区隔绝...... ${}^{190}$${}^{190}$^(190){ }^{190} 。

Produced in the following decade, the composition of “the Regional Map of Gansu” (Gansu difang tuben 甘肅地方圖本; hereafter Gansu difang tu) confirms Zhu Ying’s concerns about the vital importance of the Gansu region in Ming frontier defense in the late fifteenth century.
甘肃地势图本》（Gansu difang tuben 甘肃地方图本，以下简称《甘肃地势图》）在随后的十年中问世，证实了朱英对 15 世纪末甘肃地区在明朝边防中至关重要性的担忧。

As the extreme northwestern frontier, Gansu consisted of the Hexi corridor, which is a narrow stretch of arable plains between the Gobi Desert and the Qilian Mountains on the north and the Hehuang valley, the Yellow, and the Huang River basins located south of the Qilian Mountains. The Hexi corridor was the main ancient Silk Road connecting North China and Central Asia. The Hehuang valley is on the Qinghai route, which is connected to Miran, an oasis town on the southern rim of the Taklamakan desert, and at times served as an alternative Silk Road. The political instability and military conflicts that ensued after the collapse of the Mongol Yuan empire undermined the thriving Silk Road trade under the Pax Mongolica. Although the Yongle emperor instituted tribute and trade relations with Central Asian states, later Ming rulers restricted the tribute trade due to increasing budgetary concerns. Ming restrictions on tribute missions threatened the economic survival of Central Asian oasis cities and strained relations with those cities, including Turfan. In protest, the Moghulistan ruler Yunus Khan (r. 1462-1484, also known as Hajji 'Ali), who controlled Turfan at that time, attacked Hami in 1473, a buffer
甘肃地处西北边陲，由河西走廊组成，河西走廊是位于戈壁滩和祁连山以北，以及祁连山以南的河套、黄河和黄河流域之间的一片狭长的耕地平原。河西走廊是古代连接华北和中亚的主要丝绸之路。和黄河谷位于青海路线上，与塔克拉玛干沙漠南缘的绿洲城镇米兰相连，有时也是另一条丝绸之路。蒙元帝国灭亡后，政局动荡，军事冲突不断，破坏了蒙古大政下繁荣的丝绸之路贸易。虽然永乐皇帝与中亚各国建立了朝贡和贸易关系，但后来的明朝统治者出于日益增长的预算考虑，限制了朝贡贸易。明朝对朝贡使团的限制威胁到中亚绿洲城市的经济生存，并使包括吐鲁番在内的这些城市关系紧张。作为抗议，当时控制吐鲁番的莫卧儿斯坦统治者尤努斯汗（Yunus Khan，1462-1484 年在位，又名哈吉-阿里）于 1473 年袭击了哈密。
${}^{189}$${}^{189}$^(189){ }^{189} Lan County was promoted to Lanzhou from the mid-Chenghua reign (r. 1464-1487).
${}^{189}$${}^{189}$^(189){ }^{189} 兰县自成化中期（1464-1487 年）升格为兰州。
${}^{190}$${}^{190}$^(190){ }^{190} Chenzhou zongzhi 椰州總志, 35: 7b-8a.
${}^{190}$${}^{190}$^(190){ }^{190} 《郴州总志》，35：7b-8a。
zone between Ming Gansu and Central Asia. After Yunus Khan’s death, political turmoil continued to worsen, and eventually brought about the Ming withdrawal from Hami in the early sixteenth century. ${}^{191}$${}^{191}$^(191){ }^{191}
哈密是明代甘肃和中亚之间的一个地区。尤努斯汗死后，政治动荡不断加剧，最终导致明朝于 16 世纪初撤出哈密。 ${}^{191}$${}^{191}$^(191){ }^{191}

Gansu difang tu shows the court’s emerging defense concerns in Gansu in the late fifteenth century (figure 33). ${}^{192}$${}^{192}$^(192){ }^{192} This map is huge, being 2.39 meters in height by 2.85 meters in width, mounted on a scroll of 2.48 by 3.25 meters. Considering its large size and the uniform direction of the inscribed texts, the map was likely to be hung on the wall. At first sight, the Gansu difang tu shares some similarities with earlier courier route maps in respect to its blue-and-green landscape painting style and the emphasis on administrative geography. However, in
甘肃防御图》展示了 15 世纪晚期朝廷在甘肃新出现的防御问题（图 33）。 ${}^{192}$${}^{192}$^(192){ }^{192} 这幅地图非常巨大，高 2.39 米，宽 2.85 米，镶嵌在 2.48 米乘 3.25 米的卷轴上。考虑到其巨大的尺寸和刻有文字的统一方向，该地图很可能是挂在墙上的。乍一看，《甘肃地方图》与早期的信使路线地图在蓝绿山水画风格和强调行政地理方面有一些相似之处。然而，在
${}^{191}$${}^{191}$^(191){ }^{191} In 1404, the actively expansionist Yongle emperor subordinated the state of Qara Del (or Kara Del) in Kumul and established Hami prefecture. Rather than governing it directly, the Ming court let the existing royal house continue to rule and send tribute envoys. The city suffered attacks from the Oirat Mongols by the 1440s, and later from the Turfans throughout the late fifteenth and early sixteenth century. Morris Rossabi, "The Tea and Horse Trade with Inner Asia during the Ming,"39-57; Morris Rossabi, "The Ming and Inner Asia,"8:143-200.
${}^{191}$${}^{191}$^(191){ }^{191} 1404年，积极扩张的永乐皇帝将库木勒的喀喇德尔（或喀喇德勒）州划归自己管辖，并设立了哈密府。明朝朝廷没有直接对其进行统治，而是让现有的王室继续统治，并派遣贡使。到 14 世纪 40 年代，这座城市遭受了卫拉特蒙古人的攻击，后来在整个 15 世纪末和 16 世纪初又遭受了土尔扈特人的攻击。莫里斯-罗沙比：《明朝与内亚的茶马贸易》，39-57；莫里斯-罗沙比：《明朝与内亚》，8:143-200。
${}^{192}$${}^{192}$^(192){ }^{192} The title of this map appears on the back of the scroll (figure 34). I agree with Li Xiaocong that this may not be the map’s original title, since during the late fifteenth century the area was more often called as the Shaanxi military commission (shaanxi xingdusi). Li believes that the scroll might have been named during the Qing establishment of the Hexi corridor as an independent province of Gansu since 1665. I believe the name could derive from the late Ming practice of calling this area “Gansu” when mapping the entire northern border areas under the nine northern border garrison spatial system (jiubian zhen). In either case, the name Gansu difang tu would be a later addition. Li Xiaocong’s argument is cited in Sheue-Yann Lu, “Ming Caihui Ben Gansu Difang Tu Kao,” Gugong Xueshu Jikan 29, no. 1 (2011): 101.
${}^{192}$${}^{192}$^(192){ }^{192} 这幅地图的标题出现在卷轴的背面（图 34）。我同意李孝聪的观点，即这可能不是该地图的原标题，因为在 15 世纪晚期，该地区更常被称为陕西军委（shaanxi xingdusi）。李认为，这幅画卷可能是在 1665 年清朝将河西走廊作为甘肃的一个独立省份时命名的。我认为这一名称可能源于明末在绘制整个北部边疆地区的地图时，根据九边戍边的空间系统将这一地区称为 "甘肃 "的做法。无论哪种情况，甘肃地防图的名称都是后来加上去的。李孝聪的论点引自 Sheue-Yann Lu，《明采会本甘肃地方图考》，《古贡学报》第 29 期（2011 年）：101.
contrast to the route maps’ frequent change in direction to place the routes at the center throughout long handscrolls, the Gansu difang tu map fixes its orientation by placing the south on the top. In addition, unlike the route map handscrolls that only show a section at a time, the Gansu difang tu map, displayed on the wall, provides an all-at-once view of the entire Gansu region.
在长长的手卷中，路线图经常改变方向，将路线置于中心位置，与此形成鲜明对比的是，《甘肃方图》将南方置于顶部，从而固定了方向。此外，与每次只显示一个部分的路线图手卷不同，挂在墙上的《甘肃地方图》可让人一目了然地看到整个甘肃地区。

The map strives to provide a comprehensive list of defense facilities in this region. As in figure 35, guards and battalions are depicted as triple concentric rectangular shapes from the aerial view, circumscribed by a white-colored crenellated parapet on the outer edge of the wall. The space inside the wall is filled in with bright red color, which contrasts with the mountains and rivers painted in green and blue. Other inferior military defense facilities, such as fortresses (bao 堡 or cheng 城), courier stations, watch towers (sai 塞), and military camps (ying 營), are illustrated as two concentric rectangular shapes in a relatively smaller size compared to that of the guards and battalions. Passes are depicted as a singular rectangle without walled structures. Beacon towers (duntai 墩臺) are drawn from a frontal view as tall structures with a red flag on the top, either attached to other fortresses and garrisons, or appearing independently. All beacon towers are painted in white, which distinguishes them from both other red-colored military facilities and blue and green landscape features.
该地图力求提供该地区防御设施的全面清单。如图 35 所示，从鸟瞰图上看，卫兵和营被描绘成三重同心矩形，由城墙外缘的白色雉堞围成。城墙内部空间用鲜艳的红色填充，与绿色和蓝色的山川形成鲜明对比。其他下级军事防御设施，如堡垒（堡或城）、信使驿站、瞭望塔（塞）和军营（营），则绘成两个同心长方形，尺寸相对小于卫兵和营。关隘则被描绘成一个没有城墙结构的长方形。烽火台（duntai 墩台）从正面看是高大的建筑，顶部有一面红旗，或与其他堡垒和守备部队相连，或独立存在。所有烽火台都涂成白色，使其与其他红色军事设施以及蓝色和绿色景观区别开来。

According to Lu’s research based on the names of administrative units, the Gansu difang $tu$$tu$tut u map can be dated between 1480 and 1488 (or it could be a later copy of the original map that reflects the military conditions of the 1480s). ${}^{193}$${}^{193}$^(193){ }^{193} This gives an important clue to contextualize the map’s creation in relation to the emerging defense needs and the changing practice of map
根据 Lu 根据行政单位名称所做的研究，《甘肃地方图》 $tu$$tu$tut u 的绘制时间可推定为 1480 年至 1488 年之间（也可能是反映 1480 年代军事状况的原图后期副本）。 ${}^{193}$${}^{193}$^(193){ }^{193} 这为我们提供了一条重要线索，可将该地图的绘制与新出现的防御需求和地图绘制实践的变化联系起来。
${}^{193}$${}^{193}$^(193){ }^{193} Ibid., 109-29.  ${}^{193}$${}^{193}$^(193){ }^{193} 同上，109-29 页。
submission. From the mid-fifteenth century, the Ming court requested the regular submission of maps from northern frontier garrisons. According to the Wanli edition of the Collected Statutes of the Ming Dynasty (hereafter Ming huidian 明會典), it was in 1465 that the court required the submission of military maps once every three years, similar to civilian administrative units. ${}^{194}$${}^{194}$^(194){ }^{194} However, the Ming shilu, on which the Ming huidian account was based, suggests that the practice of regular submission had already existed. Instead, the imperial decree of triennial submission was a response to a request from the Grand Coordinator of Shaanxi, who complained about the excessive administrative burden caused by annual submission of both registers and maps. ${}^{195}$${}^{195}$^(195){ }^{195} Although exactly when the regular map submission began remains unclear, the debate regarding the frequency of map submissions suggests that regular submission was a recent invention that frontier officers considered burdensome.
提交。从十五世纪中叶开始，明朝朝廷要求北方边防军定期提交地图。据万历版《明会典》（以下简称《明会典》）记载，1465 年，朝廷要求每三年提交一次军事地图，与民事行政单位类似。 ${}^{194}$${}^{194}$^(194){ }^{194} 不过，《明会典》记载所依据的《明实录》表明，定期提交地图的做法早已存在。 ${}^{194}$${}^{194}$^(194){ }^{194} 《明会典》所依据的《明实录》表明，定期上报的做法早已存在，而每三年上报一次的诏令是对陕西巡抚抱怨每年上报登记册和地图造成过重行政负担的请求的回应。 ${}^{195}$${}^{195}$^(195){ }^{195} 虽然定期提交地图的确切时间尚不清楚，但有关提交地图频率的争论表明，定期提交地图是边防官员认为负担过重的新发明。

The submitted maps aided court officials to understand the spatial relations among the major garrisons and the increasing number of smaller defense facilities. Although textual lists of local defense facilities had existed before, they simply listed the names of smaller defense facilities categorized by their superior garrisons in charge of supervision. Although this information was helpful for clarifying the chain of bureaucratic command, it was hard to understand their relative positions, especially if they were attached to different superior units. Maps allowed a bird’s-eye view of military networks in the region, enabling viewers to easily understand the relative distance and location of each military facility.
提交的地图有助于朝廷官员了解主要驻军与日益增多的小型防御设施之间的空间关系。虽然以前也有地方防卫设施的文字清单，但它们只是按照负责监管的上级驻军分类列出了较小防卫设施的名称。虽然这些信息有助于明确官僚指挥系统，但很难了解它们的相对位置，尤其是当它们隶属于不同的上级单位时。通过地图可以鸟瞰该地区的军事网络，使观众能够轻松了解每个军事设施的相对距离和位置。

The Gansu difang tu map also includes detailed information on the soldiers and animals assigned to each military facility (figure 35). For guards and battalions, the map includes the total number of registered soldiers and horses, the soldiers assigned to military activities (操練) (and ready to mobilize in emergency), and those for other purposes (such as city defense, scouting and farming), those stationed at the garrison (實有) and those transferred to other places （調到），among which were those who had already arrived（實有已到）at the facility and those who had not (事故未到). For beacon towers, the name, the distance from the supervising garrison, and the number of soldiers (qijun 旗軍) and horses stationed at the tower are recorded.
甘肃地方图》还包括分配到各军事设施的士兵和牲畜的详细信息（图 35）。对于卫队和营队，该图包括登记在册的兵马总数、分配从事军事活动（操练）（并在紧急情况下随时准备调动）的兵马、用于其他目的（如城防、侦察和耕作）的兵马、驻扎在守备部队（实有）的兵马以及调往其他地方的兵马、对于烽火台来说，"实有 "和 "未到 "分别指已到达和未到达设施的人。至于烽火台，则记录了名称、与监督驻军的距离，以及驻守烽火台的士兵（旗军）和马匹的数量。

Such information indicates the military strength of each facility in terms of quantifiable data. In addition, the highly standardized vocabulary and indentation in the textual annotations helped the understanding of the region’s military forces for the intended readers, likely the high-ranking military officers or the court audience.
这些信息以可量化的数据说明了每个设施的军事实力。此外，文本注释中高度标准化的词汇和缩进有助于预期读者（可能是高级军官或宫廷听众）了解该地区的军事力量。

For instance, Table 2 states the troop strength of the Ganzhou Guard recorded within the image of the Ganzhou wall on the Gansu difang tu.
例如，表 2 列出了甘肃地方图上甘州城墙图像中记录的甘州卫兵力。

The map’s textual annotations still focus on the management of resources, such as soldiers and horses assigned to each garrison. While there is a brief explanation of its distance from the neighboring units, it does not specify local geography, such as features of terrain, strategic passes, and vulnerable places. In contrast, the later sixteenth century atlases include more detailed territorial information, as will be analyzed in the next section. Such a difference may be attributed to their different formats: a hanging scroll was probably not the best format design for the close investigation of local topography, compared to smaller-size illustrations of later atlases. The textual annotations included in the Gansu difang tu are not visible at a distance unless a viewer came very close to the map. Still, the selected inclusion of information regarding military
地图上的文字注释仍侧重于资源的管理，如分配给每个驻军的士兵和马匹。虽然地图简要说明了驻军与邻近部队的距离，但并没有具体说明当地的地理环境，如地形特征、战略通道和易受攻击的地方。相比之下，16 世纪晚期的地图集包含更详细的领土信息，下一节将对此进行分析。造成这种差异的原因可能是它们的格式不同：与后来地图集中尺寸较小的插图相比，挂轴可能不是仔细研究当地地形的最佳格式设计。甘肃地方图》中的文字注释在远处是看不到的，除非观众非常靠近地图。尽管如此，有选择地收录有关军事的信息

resources instead of local geographic features suggests that the map’s primary purpose was to spatialize the military networks using the quantifiable data of soldiers and horses assigned to each defense facility.
而不是当地的地理特征，这表明该地图的主要目的是利用分配到每个国防设施的兵马的量化数据来实现军事网络的空间化。

Other than visualizing the military networks, landscape depictions in the Gansu difang tu also served the imperial interest in creating the meaning of local places and historical memories in a contested frontier. The mapmaker collected, (re)arranged, and recreated local place markers to fit into a particular spatial narrative that supported the imperial authority over the region. The Gansu difang tu’s peculiar choice of direction, by placing the south on the top, defines Gansu’s subordinating position in relation to Lanzhou situated at the north or the superior position. As the farthest northwestern frontier, Gansu was controlled by the Shaanxi military commission (Shaanxi xingdusi), administratively independent from, while nominally included within, the boundary of Shaanxi civilian province. By flipping the direction upside down, Gansu is placed south of Lanzhou, which appears at the top left corner of the map looking over Gansu, while Lanzhou is actually located in the southeast part of Gansu. Lanzhou symbolizes Shaanxi’s civilian provincial administration, controlled by the provincial center of Xi’an, which is missing in this map but located southeast across the Yellow River similar to Lanzhou.
除了将军事网络视觉化之外，甘肃地方图中的景观描绘也符合帝国的利益，即在有争议的边疆地区创造当地地方的意义和历史记忆。地图绘制者收集、（重新）安排和再创造当地的地方标记，使其符合特定的空间叙事，以支持帝国对该地区的权威。甘肃地方图》在方向上的特殊选择，将南方置于上方，确定了甘肃相对于北方兰州的从属地位或优势地位。作为西北最远的边疆，甘肃由陕西省军事委员会（陕西行署）控制，行政上独立于陕西省民政厅，但名义上属于陕西省民政厅的范围。通过颠倒方向，甘肃位于兰州以南，兰州出现在俯瞰甘肃地图的左上角，而兰州实际上位于甘肃的东南部。兰州象征着陕西的民政省政府，由省中心西安控制，西安在这幅地图中缺失，但与兰州类似，位于黄河对岸的东南方。

A number of Shaanxi maps other than Gansu difang $tu$$tu$tut u adopts the same composition that assumes a higher/upper position of Xi’an as an imperial representative in relation to other localities, regardless of whether or not Xi’an is included in the maps. For instance, the similar upside-down orientation is found in several printed maps in the mid-sixteenth century Shaanxi provincial gazetteer-the “Map of the Entire Shaanxi Area” (Quanshan jiangyu tu 全陝疆域圖;
除甘肃地方志 $tu$$tu$tut u 外，其他一些陕西地图也采用了同样的构图，即无论西安是否被纳入地图，西安作为帝王的代表，相对于其他地方都处于较高/较高的位置。例如，16 世纪中叶陕西省地名录《全陝疆域图》中的几幅印刷地图也采用了类似的倒置方位；
figure 36), the “Map of the West of the River” (Hexi jiangyu zongtu 河西疆域總圖；figure 37 and 39), and the “Map of the Yellow River” (Huanghe tu 黃河圖；figure 38). Such an
图 36）、《河西舆图》（河西疆域总图；图 37 和 39）以及《黄河图》（黄河图；图 38）。这样的
orientation is in accordance with the ideal notion of the “emperors facing south,” in which the capital as the seat of the emperor was often located at top, and other localities appearing south, regardless of their actual orientation. As the provincial center, Xi’an represents the emperor’s seat in these maps (and Lanzhou in case of Gansu difang tu).
根据 "皇帝坐北朝南 "的理想概念，作为皇帝所在地的首都通常位于上方，而其他地方则位于南方，无论其实际方位如何。西安作为省城中心，在这些地图中代表了皇帝的所在地（甘肃方图则代表兰州）。

Not only did this upside-down orientation fit the ideal political order, but it also transformed a distant and little-known frontier like Gansu into a more familiar place using the existing visual convention of mapping the Yellow River. Studies indicates that most Yellow River handscrolls (changjiang $tu$$tu$tut u 長江圖) depict the image of the river flowing from its western source to the Bohai Sea in the east. ${}^{198}$${}^{198}$^(198){ }^{198} Since a handscroll is viewed from right to left, the river source (or the area closer to the river origin) in the west appears at the right, resulting in the upside-down view of placing the south on top. By adopting the same compositional structure and depicting the river flowing from the right, the Gansu difang $tu$$tu$tut u helped viewers to recognize Gansu easily based on a well-known visual practice and geographic knowledge. As discussed in chapter one, the existing cartographic practices often employed the Yellow River, together with the Yangzi River, as a signifier of the Central Realm. Considering these practices, the Gansu difang tu’s emphasis on the spatial relationship between the Yellow River and Gansu had the visual effect of transforming a contested frontier space into an integrated part of the Ming empire.
这种颠倒的方位不仅符合理想的政治秩序，而且还利用现有的黄河图视觉习惯，将甘肃这样一个遥远而鲜为人知的边疆变成了人们更熟悉的地方。研究表明，大多数黄河手卷（changjiang $tu$$tu$tut u 长江图）都描绘了黄河从西部源头流向东部渤海的形象。 ${}^{198}$${}^{198}$^(198){ }^{198} 由于手卷是从右往左看的，所以西边的河源（或离河源较近的地方）就会出现在右边，从而形成了南边在上的倒置景象。甘肃地画 $tu$$tu$tut u 采用了相同的构图结构，描绘了从右边流过的河流，帮助观众根据众所周知的视觉习惯和地理知识轻松识别甘肃。正如第一章所讨论的，现有的地图绘制实践通常使用黄河和长江作为中原地区的标志。考虑到这些做法，《甘肃地方图》对黄河与甘肃之间空间关系的强调，在视觉上产生了将一个有争议的边疆空间转变为明朝帝国的一个组成部分的效果。

Viewers are supposed to read the Gansu difang $tu$$tu$tut u in a particular order: from the northern exterior of the empire toward the southern interior near Lanzhou, and again from a relatively inner area in the east toward the western edge of the empire, where the sacred places of Yellow River source and Mount Kunlun are located. Since it is an upside-down view, the “north” appears in the map’s foreground as relatively empty beyond the Zhenfan 鎮番 garrison located at the extreme northern edge of the Ming territory. Moving along three upward-flowing streams, in the left, middle, and right of the foreground respectively, the viewers’ eyes meet a series of garrisons (Shandan, Yongchang, Liangzhou, and Gulang from right to left) located between two parallel mountain ranges (possibly Mount Heli 合黎山 at the bottom and Mount Qilian 祁連山 at the top) in the middle ground. Then from the Gulang battalion on the left, a chain of garrisons and fortresses runs upward along with vertically-layered mountains and two river streams (presumably the Zhuanglang and Datong Rivers 大通河). At the map’s top left corner is the Zhenyuan 鎮遠 Bridge of the Yellow River (figure 40). This bridge is also called a pontoon bridge (fuqiao 浮橋), made of boats covered with wooden planking, and only in use during the seasons when the Yellow River was unfrozen (figure 41). ${}^{199}$${}^{199}$^(199){ }^{199} The bridge was considered a major gateway to the “West of the River” region (通河西), or Gansu, linking it with the rest of Shaanxi
甘肃地方图 $tu$$tu$tut u 的阅读顺序是这样的：从帝国的北部外部向兰州附近的南部内部阅读，然后再从东部相对内部的区域向帝国的西部边缘阅读，那里是黄河源头和昆仑山的圣地。由于是倒置图，地图前景中的 "北方 "在明朝领土最北端的镇番驻军之外显得相对空旷。沿着前景左、中、右三条向上流淌的溪流前行，观众的视线会与位于两座平行山脉（可能是山下的合黎山和山顶的祁连山）之间的一系列驻军（从右至左依次为山丹、永昌、凉州和古浪）相遇。然后，从左侧的古浪营开始，沿着垂直层叠的山脉和两条河流（可能是庄浪河和大通河），一连串的驻军和堡垒向上延伸。地图左上角是黄河镇远桥（图 40）。这座桥也叫浮桥，是用木板铺成的船，只在黄河不冻的季节使用（图 41）。 ${}^{199}$${}^{199}$^(199){ }^{199} 这座桥被认为是通往 "通河西 "地区（即甘肃）的主要门户，连接着陕西的其他地区

province closer to the political center. On the map, the viewers’ eyes approach from the opposite direction, looking over the city of Lanzhou, which was under civilian rule, from the other side of the Yellow River. The bridge symbolizes both connectivity and discontinuity. Connecting both sides of the Yellow River, the bridge incorporates Gansu as a part of Ming territory. ${}^{200}$${}^{200}$^(200){ }^{200} At the same time, the bridge blocks enemies’ intrusion from the outer region by cutting off their crossing if needed.
兰州离政治中心更近。在地图上，观众的视线从相反的方向靠近，从黄河对岸俯瞰文官统治下的兰州城。这座桥既象征着连通性，也象征着不连续性。大桥连接黄河两岸，将甘肃纳入明朝版图。 ${}^{200}$${}^{200}$^(200){ }^{200} 同时，这座桥还能在必要时切断敌人的渡河通道，阻止敌人从外部地区入侵。

From the top left corner, the flow of the Yellow River brings the viewers toward the right side of the map, which is the western end of the empire. One of its tributaries, the Xining River (or Huangshui 湟水), runs slightly southeast, and along both sides a series of fortresses connected to the Xining Guard is depicted. At the right end of the remaining part of the map there are two lakes, one named “small lake” (xiao haizi 小海子) appearing on a mountain peak and the other, named “big lake” (da haizi 大海子), receiving water from five different streams (figure 43-1 and 43-2). This “big lake” is known as the Lake of the West (Xihai 西海), or the Blue Lake (Qinghai 青海, hereafter Lake Qinghai). ${}^{201}$${}^{201}$^(201){ }^{201} Unfortunately, the right side of the map is damaged. Like the later “Map of the West of the River” in the Shaanxi tongzhi, the Gansu difang
从左上角开始，黄河的流向将观众带向地图的右侧，也就是帝国的西端。其支流之一的西宁河（或称湟水）略向东南流去，沿河两侧描绘了一系列与西宁卫相连的要塞。在地图剩余部分的右端，有两个湖泊，一个名为 "小海子"，出现在一座山峰上；另一个名为 "大海子"，接受来自五条不同溪流的水（图 43-1 和 43-2）。这个 "大湖 "被称为西海湖或青海湖。 ${}^{201}$${}^{201}$^(201){ }^{201} 遗憾的是，该地图右侧已损坏。与后来《陕西通志》中的《河西图》一样，《甘肃地方图

$tu$$tu$tut u may have originally included the source of the Yellow River (Huanghe yuan 黃河源) and Mount Kunlun in this damaged part. ${}^{202}$${}^{202}$^(202){ }^{202}
$tu$$tu$tut u 最初可能将黄河源头和昆仑山包括在这一受损部分中。 ${}^{202}$${}^{202}$^(202){ }^{202}

Both at the beginning and ending points of the map’s visual journey, local iconographic features highlight the region’s historical integration into the Ming. The northern end (at the bottom of the map) of the Gansu difang $tu$$tu$tut u includes the images of three connected lakes, identified as “haizi” (海子). These lakes represent the modern Juyan Lake 居延海, located at the outer edge of the Black River. ${}^{203}$${}^{203}$^(203){ }^{203} The lake’s image reminds viewers of the Ming conquest of
在地图视觉之旅的起点和终点，当地的图标特征都突出了该地区融入明朝的历史。甘肃地方图 $tu$$tu$tut u 的北端（地图底部）包括三个相连湖泊的图像，被称为 "海子"。这些湖泊代表了位于黑河外缘的现代居延海。 ${}^{203}$${}^{203}$^(203){ }^{203} 这个湖的形象让人联想到明朝征服黑龙江的历史。

beyond the right side of the screen (figure 47). Such a map composition probably follows the trend of repositioning both Mount Kunlun and the Yellow River source farther south, as in other later maps of the Ming empire. For the changing positions of these two landscape features, see Dorofeeva-Lichtmann, “A History of a Spatial Relationship: Kunlun Mountain and the Yellow River Source from Chinese Cosmography through to Western Cartography,” 1-31.
图 47）。这样的地图构图很可能遵循了将昆仑山和黄河源头重新定位到更南边的趋势，就像明帝国后期的其他地图一样。关于这两种地貌的位置变化，请参阅 Dorofeeva-Lichtmann 的《空间关系史》：从中国宇宙学到西方地图学中的昆仑山和黄河源"，1-31。
${}^{203}$${}^{203}$^(203){ }^{203} The Black River is also called the Weak River (ruoshui 弱水), known as a mythological river flowing near the western paradise of Mount Kunlun.
${}^{203}$${}^{203}$^(203){ }^{203} 黑河又称弱水，是流经昆仑山西方极乐世界附近的一条神话传说中的河流。

Etsina (亦集乃 in the map; its modern character is 額濟納), also known as Khara-khoto or “Black Water City” (in Chinese Heishui cheng 黑水城). ${}^{204}$${}^{204}$^(204){ }^{204} After having been founded by the Western Xia empire (1038-1227), the city continued to flourish as a trade center under the Yuan. In 1372, the Ming army besieged the city defended by the Mongol general Khara Bator (active in the fourteenth century; meaning the Black General). In order to force his surrender, the Ming cut the city’s water source from outside. ${}^{205}$${}^{205}$^(205){ }^{205} Reflecting this story, the map depicts the lake stream on top of Etzina as being dried out, colored in darkish yellow instead of green (figure 44).
额济纳（Etsina，地图中为亦集乃；现代字为额济纳），又称哈剌忽多或 "黑水城"（汉语为黑水城）。 ${}^{204}$${}^{204}$^(204){ }^{204} 黑水城由西夏帝国（1038-1227 年）建立，元朝时作为贸易中心继续繁荣。1372 年，明军围攻蒙古将军花剌子模（Khara Bator，活跃于 14 世纪，意为黑将军）守卫的城市。为了迫使他投降，明军切断了城外的水源。 ${}^{205}$${}^{205}$^(205){ }^{205} 为了反映这个故事，地图上描绘的额济纳城顶的湖流已经干涸，颜色由绿色变为深黄色（图 44）。

Next to the abandoned city of Etzina, in the middle of the foreground, there is a circular shape hill, identified as Mount Tuohuan 脫歡山 (figure 45), accompanied by two Tibetan Buddhist pagodas on its right side. This mountain similarly commemorates a Ming military success on its northernmost frontier. During the 1430s, the area suffered from frequent raids and violent conflicts between different Mongol tribes, due to the expansive policy of the Oirat Mongols led by Toghon (?-1438?; his Chinese name is Tuohuan, the same characters with the Mount Tuohuan 脫歡). ${}^{206}$${}^{206}$^(206){ }^{206} Born into a family of generations of military officers, Mao Zhong
在废弃的额济纳城旁边，前景中间有一座圆形山丘，被认定为脱欢山（图 45），其右侧有两座藏传佛教佛塔。这座山同样也是为了纪念明朝在其最北部边境的一次军事胜利。1430 年代，由于以脱欢（Toghon，-1438 年；汉名脱欢，与脱欢山同音）为首的花剌子模蒙古人的扩张政策，该地区频繁遭受蒙古部落之间的袭击和激烈冲突。 ${}^{206}$${}^{206}$^(206){ }^{206} 毛仲出生在一个世代从军的家庭。

毛忠（1394 - 1468） was known for several military victories against the Mongols at Mount Yibula 亦不剌山 (current Mount Yabulai 雅布赖山) in 1433 and Mount Tuohuan in 1434. ${}^{207}$${}^{207}$^(207){ }^{207} In 1434, Mao captured an Oirat Mongol general Öljei Temur (完者帖木兒, ?-?; a different person than the Bunyashir Khan [1379-1412] having the same name), who had been pursuing the Adai Khan (1390-1438) of the Alutai (or Arughtai) Mongols and a number of his followers. Mao’s exploit seems to have been a combined result of luck and Ming opportunism to exploit the intertribal conflicts between different Mongols, and did not lead to Ming consolidation of its power over the region. Still, Mao was later promoted to Earl of Fuqiang (伏羌伯) and these two mountains served as the northern jurisdictional boundary of Zhenfan and Yongchang garrisons. ${}^{208}$${}^{208}$^(208){ }^{208} Considering this historical background, the map’s emphasis on Mount Tuohuan celebrates both the individual exploits of Mao Zhong, a well-known general native to this region, and the Ming military victories that legitimized its domination over the area. The same Chinese
毛忠（1394 - 1468）毛忠因 1433 年在亦不剌山（今雅布赖山）和 1434 年在沱湟山对蒙古人的数次军事胜利而闻名。 ${}^{207}$${}^{207}$^(207){ }^{207} 1434年，毛泽东俘虏了追击阿鲁台（或阿鲁格台）蒙古阿台汗（1390-1438年）及其一些追随者的花剌子模蒙古将领窝阔台帖木儿（完者帖木儿，?-?；与同名的布雅思尔汗[1379-1412年]不同）。毛泽东的功绩似乎是运气和明朝利用不同蒙古部落间矛盾的机会主义的综合结果，并没有使明朝巩固对该地区的统治。不过，毛泽东后来还是被晋升为伏羌伯，这两座山也成为镇番和永昌驻军的北部管辖边界。 ${}^{208}$${}^{208}$^(208){ }^{208} 考虑到这一历史背景，该地图对坨环山的强调既是为了颂扬该地区著名将领毛忠的个人功绩，也是为了颂扬明朝的军事胜利，这些胜利使明朝对该地区的统治合法化。同样的中国
the title taishi and continued to expand the territories. He was the same person who captured the Zhengtong Emperor at Fortress Tumu.
太师的称号，并继续扩张领土。他就是在土木堡擒获正统皇帝的人。
${}^{207}$${}^{207}$^(207){ }^{207} See Liu Guang’s memorial in Ming Xuanzong shilu, 113: 4a. Mao Zhong was from a hereditary military officer family registered in Liangzhou. Since his great grandfather had been subordinated to the new Ming Hongwu emperor, his family served for generations as military generals in the region. Mao Zhong’s grandfather died during the Hami campaign, and his father served as a company commander (baihu) of Yongchang wei. Lanzhou fuzhi, 9: 55a-b; Guochao xianzheng lu 國朝獻徵録，juan 9， chapter “the records on earls-first.”
${}^{207}$${}^{207}$^(207){ }^{207} 见《明宣宗实录》113：4a 中的刘光奏折。毛忠出身于凉州世袭军官家庭。由于他的曾祖父隶属于新的明洪武皇帝，因此他的家族世代在该地区担任武将。毛忠的祖父死于哈密战役，父亲曾任永昌卫连长（白虎）。兰州府志》，9: 55a-b；《国朝献征录》，卷 9，"伯爵第 "一章。

characters used to describe the names for Esen’s father Toghon and for this mountain also facilitated the imaginary Ming military victory over the Oirat Mongols, which belied the reality in which the Ming neither captured Toghon nor subdued the Oirat Mongols in this region.
用于描述额森的父亲托格洪和这座山的名称的字符也有助于想象明朝在军事上战胜了卫拉特蒙古人，但这与现实情况不符，在这一地区，明朝既没有攻占托格洪，也没有征服卫拉特蒙古人。

On the western end (the right side of the map), the extant Gansu difang $tu$$tu$tut u ends with Lake Qinghai. Its distinct lake topography contrasts with the barren desert terrain of most of the northwestern frontier. For instance, a book on the northwestern frontier published in 1691, $A$$A$AA Brief Narrative of the Lands Neighboring Shaanxi (Qinbian jilüe 秦邊紀略), describes the wonder and marvel of such a contrast, saying that "it [the lake] is located at the corner of the northwest, which is mostly sand in which a plant cannot grow more than one inch (cun). ${}^{209}$${}^{209}$^(209){ }^{209} [Where there is a lake,] northern water streams all flow over and plentiful forest grows. ${}^{210}$${}^{210}$^(210){ }^{210} The map depicts two islands in the middle of the lake, including Haixin Island (海心山). Used as a residence by Tibetan lamas, Haixin Island was also known as a breeding ground that had once produced excellent steeds (龍種) for the Sui dynasty Emperor Yang (r. 604-618). ${}^{211}$${}^{211}$^(211){ }^{211} This pictorial element evokes the historical memory of political engagement with this region in a strikingly similar way to the Ming in regard to the acquisition of horses for the imperial army. It
在西端（地图右侧），现存的甘肃地仓 $tu$$tu$tut u 以青海湖为终点。其独特的湖泊地形与西北边疆大部分地区贫瘠的沙漠地形形成鲜明对比。例如，1691 年出版的一本关于西北边疆的书《秦边纪略》（ $A$$A$AA ）描述了这种反差带来的奇妙和惊叹，书中说："它（湖泊）位于西北的一角，那里大部分是沙地，植物在其中生长不能超过一寸（厘）。 ${}^{209}$${}^{209}$^(209){ }^{209} [有湖的地方，]北方的水流都流过，生长着茂密的森林。 ${}^{210}$${}^{210}$^(210){ }^{210} 地图描绘了湖中央的两个岛屿，其中包括海心山。海心岛曾是西藏喇嘛的居所，也是隋炀帝（604-618 年）的良种繁育场。 ${}^{211}$${}^{211}$^(211){ }^{211} 这一图像元素唤起了人们对这一地区政治参与的历史记忆，其方式与明朝为皇军购置马匹的方式惊人地相似。它

also was timely considering the weakening Ming influence over the Lake Qinghai basin in the late fifteenth century. In the early sixteenth century, increasing conflicts with the Mongols caused the Ming to lose its control over the region, which it failed to regain before the end of the dynasty. ${}^{212}$${}^{212}$^(212){ }^{212}
考虑到十五世纪末明朝对青海湖盆地的影响力不断减弱，《青海通史》的出版也恰逢其时。16 世纪初，明朝与蒙古人的冲突日益加剧，导致明朝失去了对该地区的控制，并在王朝灭亡前未能重新夺回控制权。 ${}^{212}$${}^{212}$^(212){ }^{212}

Altogether, my analysis demonstrates how the Gansu difang tu defines the meaning of local places and recreates historical memory in order to justify Ming military domination over Gansu during the late fifteenth century. From the mid-fifteenth century onward, the change in military strategy and the court’s intensified efforts to collect information about military networks promoted the map as an instrument to articulate the spatial relations among an ever-increasing number of defense facilities. Such knowledge was expected to enhance the court’s supervision over its military forces in distant frontiers. Providing an all-at-once view over the entire area, the Gansu difang tu map enhanced the viewers’ sense of visibility as they wished. Considering the standard bureaucratic language used in textual annotations, it is likely that the recipient was either the emperor or high-ranking officers in the Ministry of War. Borrowing visual conventions, utilizing local iconography, and reproducing local history, the map legitimized the Ming claim over the contested territories in an era of growing military conflicts.
总之，我的分析展示了甘肃地方图是如何定义地方意义和再现历史记忆，以证明十五世纪末明朝对甘肃的军事统治是合理的。从十五世纪中叶开始，军事战略的变化和朝廷收集军事网络信息的力度加大，促使地图成为阐明日益增多的防御设施之间空间关系的工具。这些知识有望加强朝廷对其在遥远边疆军事力量的监督。甘肃地方图》提供了对整个地区的全景式观察，可随心所欲地增强观看者的可视感。考虑到文字注释中使用的标准官僚语言，接收者很可能是皇帝或兵部的高级官员。借用视觉惯例、利用当地图标和再现当地历史，在军事冲突日益加剧的时代，该地图使明朝对争议领土的主张合法化。

In the sixteenth century, military crises continued to worsen. In response, the court continued to collect more detailed information about local geography, such as where enemies often intruded and then places toward which they often headed, what the local terrain looked like, and how Ming soldiers could take advantage of the terrain in military operations. The four Shaanxi atlases that I analyze in this section demonstrate how security crises encouraged the court to develop a close-up perspective toward frontier spaces. The court officials used maps to obtain topographical knowledge, plan military actions, and transform local terrain as needed for effective defense.
16 世纪，军事危机持续恶化。作为回应，朝廷继续收集有关当地地理的更详细的信息，如敌人经常入侵的地方和他们经常前往的地方、当地地形的样子，以及明朝士兵如何在军事行动中利用地形。我在本节中分析的四幅陕西地图集展示了安全危机如何促使朝廷对边疆空间形成一种近距离观察的视角。朝廷官员利用地图获取地形知识、规划军事行动，并根据需要改造当地地形，以实现有效防御。

On the one hand, the court justified its intensified map-making as a way to overcome indolence on the part of its own officials. Maps and documents were expected to be accurate representations of military and fiscal conditions in frontiers. However, the court suspected that most officials simply copied and resubmitted existing maps and documents their predecessors had made. Such suspicions were partly true, though more as a result of a structural problem. The grand coordinators in charge had little incentive to investigate actual military conditions that could antagonize their relations with other military officials in the field. ${}^{213}$${}^{213}$^(213){ }^{213} To solve the problem,
一方面，朝廷认为加强地图绘制是为了克服其官员的懒惰。人们期望地图和文件能准确反映边疆的军事和财政状况。然而，朝廷怀疑大多数官员只是复制和重新提交他们的前任绘制的现有地图和文件。这种怀疑部分属实，但更多的是结构性问题造成的。负责的大协调员没有什么动力去调查实际的军事情况，因为这可能会影响他们与实地其他军事官员的关系。 ${}^{213}$${}^{213}$^(213){ }^{213} 为了解决问题、

the court began to appoint regional inspectors (xun’an yushi) to investigate frontier garrisons and verify the maps and military registers that had been previously submitted by grand coordinators. ${}^{214}$${}^{214}$^(214){ }^{214} However, regional inspectors faced the same problem with the grand coordinators and the court remained unsatisfied. The 1546 guidelines made by the Ministry of War for regional inspectors reflect the court’s effort to obtain more accurate and up-to-date information. As for the background of making such guidelines, the ministry indicated the failure of the current regional inspectors, who "against the original intention, still relied on previous versions in compiling maps and essays and thus perpetuated false information or omissions… Therefore, our officials (of the ministry) had no firm basis to make policy decisions. ${}^{215}$${}^{215}$^(215){ }^{215}
${}^{214}$${}^{214}$^(214){ }^{214} 朝廷开始任命地方巡抚（寻安尉）调查边防驻军，并核实大都督之前提交的地图和军籍。 ${}^{214}$${}^{214}$^(214){ }^{214} 然而，地方巡抚面临着与大都督相同的问题，朝廷仍不满意。兵部为地区督察员制定的 1546 准则反映了朝廷为获得更准确和最新的信息所做的努力。至于制定这些指导方针的背景，兵部表示是由于现任地区巡抚的失职，他们 "违背初衷，在编纂地图和文章时仍然依赖于以前的版本，从而使虚假信息或遗漏长期存在......因此，我们（兵部）官员在做出决策时没有坚实的依据。 ${}^{215}$${}^{215}$^(215){ }^{215}
could lead to serious political peril. Yu’s political adversaries also used the burdens of wall-building to push for forcing Yu’s retirement. See Waldron, The Great Wall of China, 116-8.
可能导致严重的政治危险。于谦的政治对手也利用修筑长城的负担来迫使于谦退休。见 Waldron, The Great Wall of China, 116-8。
${}^{214}$${}^{214}$^(214){ }^{214} When enthroned in 1521, the Jiajing Emperor commanded the regional inspectors to submit maps and registers every three years to the Ministry of War. Ming Shizong shilu, 13: 5. As executive officials of the Censorate, regional inspectors were expected to be relatively independent from the power politics within the frontier regions. Also, their authority to report directly to the throne could bypass the usual bureaucratic chain of document submission that might have caused the intervention of other officials. In addition to the supervision of military officials and the conditions of garrisons, the court let regional inspectors engage more directly in military campaigns, including the recruitment and training of mercenaries for frontier forces in the late sixteenth and early seventeenth centuries. In the Jiajing and Wanli reigns, there were increasing complaints that regional inspectors intervened with or harassed grand coordinators. For the Censorate’s increasing role in northern border affairs, see Charles O. Hucker, The Censorial System of Ming China (Stanford University Press, 1966).
${}^{214}$${}^{214}$^(214){ }^{214} 嘉靖皇帝在 1521 年登基后，命令地方巡抚每三年向兵部提交地图和名册。明世宗实录》，13：5。 作为佥事的执行官员，地方巡抚应相对独立于边疆地区的权力政治。此外，他们直接向皇上汇报的权力可以绕过通常的官僚文件提交链条，而这一链条可能会引起其他官员的干预。除了监督武官和驻军条件外，朝廷还让地区巡抚更直接地参与军事行动，包括在 16 世纪末和 17 世纪初为边军招募和训练雇佣军。嘉靖和万历年间，关于地方巡抚干预或骚扰大员的投诉越来越多。关于佥都御史在北方边境事务中日益重要的作用，见 Charles O. Hucker，《明代中国的佥都御史制度》（斯坦福大学出版社，1966 年）。

On the other hand, the 1546 guidelines demonstrate a new interest in frontier territories that did not exist in maps of the previous century. As shown earlier, both the courier station maps and Gansu difang tu illustrated most landscape features as relatively abstract without providing detailed explanations of local terrain. The textual information focused more on resource management, such as the route information or the number of soldiers and horses assigned to each garrison. In contrast, the 1546 guidelines reveal the ministry’s quest for much more detailed information of local terrains (table 3). ${}^{216}$${}^{216}$^(216){ }^{216} Such information would be useful to foresee future invasion routes, to identify places that required additional fortification, and to devise effective defense strategies in each area.
另一方面，1546 年指南显示了对边疆领土的新兴趣，而上个世纪的地图中并不存在这种兴趣。如前所述，信使驿站图和甘肃地方图都将大部分地貌特征描绘得相对抽象，没有对当地地形进行详细说明。文字信息更侧重于资源管理，如路线信息或分配给每个驻军的兵马数量。相比之下，1546 年的指南则揭示了该部对当地地形详细信息的追求（表 3）。 ${}^{216}$${}^{216}$^(216){ }^{216} 这些信息对于预测未来的入侵路线、确定需要增加防御工事的地方以及在每个地区制定有效的防御策略都非常有用。

Maps 地图
Territorial scope of this garrison
驻军的领土范围
Neighboring garrisons that this garrison borders
与该驻军接壤的邻近驻军
Locations of passes, towers, camps, and fortresses under the jurisdiction and the officials in charge of each facility
所辖关卡、碉堡、营地和要塞的位置，以及负责每个设施的官员
Distance to a place of strategic importance from the main garrison town
从主要驻军城镇到重要战略地点的距离
Places where the Ming side conducts offensive operations, and their distance from the strategic place
明方开展进攻行动的地点及其与战略要地的距离
Places where a large or small group of raiders could enter, and any nearby facilities or obstacles that could block raiders
一大群或一小群突袭者可能进入的地方，以及附近可能阻挡突袭者的设施或障碍物
Distances to the main garrison town from the original locations of passes or fortresses and from the current locations of passes and fortresses (in cases where the location was changed or a new facility was added)
从山口或要塞的原始位置以及从山口和要塞的当前位置（在位置改变或增加了新设施的情况下）到主要驻军城镇的距离
Places which border the neighboring garrisons and whether these garrisons could transfer soldiers under emergency
与邻近驻军接壤的地方，以及这些驻军能否在紧急情况下调动士兵
Locations of northern enemies and whether they were militarily strong or weak
北方敌人的位置及其军事实力是强是弱
Numbers of officials, soldiers, horses and provisions station in smaller fortresses
小型要塞的官员、士兵、马匹和辎重数量

Created more or less at the same time of the above guidelines, the following four atlases exemplify the court’s emerging interest in frontier geography during the sixteenth century. ${}^{218}$${}^{218}$^(218){ }^{218}
以下四幅地图集大致与上述指南同时绘制，体现了 16 世纪宫廷对边疆地理的新兴兴趣。 ${}^{218}$${}^{218}$^(218){ }^{218}

Illustrations and Descriptions for Attack and Defense in Gansu (Gansu zhen zhanshou tulüe 甘肅鎮戰守圖略; figure 48-1, 48-2 and 49), later than 1544, color on paper, atlas of 35 sheets (including 13 maps), 52*90, National Palace Museum, Taipei ${}^{219}$${}^{219}$^(219){ }^{219} Illustrations and Descriptions for Attack and Defense in Shaanxi (Shaanxi zhen zhanshou tulüe 陝西鎮戰守圖略), 57 sheets (35 maps) (hereafter Shaanxi album A) ${}^{220}$${}^{220}$^(220){ }^{220} Illustrations and Descriptions for Attack and Defense in Shaanxi (also titled as Shaanxi zhen zhanshou tulüe), 11 sheets (11 maps) (hereafter Shaanxi album B) Illustrations and Descriptions for Attack and Defense in Ningxia (Ningxia zhen zhanshou tulüe 寧夏鎮戰守圖略), 20 sheets (16 maps)
甘肃镇守图略》（Gansu zhen zhanshou tulüe 甘肅鎮戰守圖略；图 48-1、48-2 和 49），晚于 1544 年，纸本彩色，地图集共 35 张（包括 13 幅地图），52*90，台北故宫博物院， ${}^{219}$${}^{219}$^(219){ }^{219} 《陕西镇守图略》、57 张（35 幅地图）（以下简称陕西画册 A） ${}^{220}$${}^{220}$^(220){ }^{220} 《陕西镇守攻防图略》、宁夏镇守图略》，20 张（16 幅地图

The maps in these atlases adopt a close-up perspective to depict each individual fortress or garrison and its immediate surroundings. Often centered on one particular fortress or garrison as the main subject, each map includes selected nearby forts, courier stations, beacon towers, other defensive barriers, and some landscape features, often identified with labels. Both the high quality of illustration and the inclusion of classified information, such as the location of a secret gate (暗門) on a defensive wall, indicate that the atlases were likely commissioned by a high-
这些地图集中的地图采用特写视角来描绘每个要塞或驻军及其周边环境。每幅地图通常以一个特定的要塞或驻军为中心，包括附近选定的要塞、信使站、烽火台、其他防御屏障和一些地貌特征，通常都有标签标识。高质量的插图和包含的机密信息（如防御墙上暗门的位置）都表明，这些地图集很可能是由高级官员委托绘制的。

${}^{220}$${}^{220}$^(220){ }^{220} This atlas has titled the same as the previous one, but the depicted areas do not overlap with each other. Considering the similarity of the painting style and the types of annotated texts within the map (while those in Ningxia and Gansu show a slight difference), it is possible that the two albums originally belonged to the same atlas. Regionally, Shaanxi album B focuses more on the East Yansui lu（延綏東路； $lu$$lu$lul u is a sub-military division below zhen). Except for the map plate no. 47 of Guide fortress (歸德堡), Shaanxi album A extends from Ningxia Central garrison on the west, throughout the northern frontiers to the Jinglu garrison toward the east, and other garrisons throughout Sino-Tibetan frontier located further south within the Shaanxi province.
${}^{220}$${}^{220}$^(220){ }^{220} 这本图册的标题与前一本相同，但所描绘的地区并不重叠。考虑到该图的绘画风格和图中注释文字的类型相似（而宁夏和甘肃的注释文字略有不同），这两本图册原本可能属于同一本图册。就地区而言，陕西图册 B 更侧重于延綏東路； $lu$$lu$lul u 是镇以下的一个军分区。除第 47 号归德堡图版外陕西图册 A 西起宁夏中卫，纵贯北部边疆，东至靖虏，其他驻军遍布汉藏边疆，位于陕西省境内更南端。
level military officer in the region, possibly the Supreme Commander of Three Defense Commands in Shaanxi (陝西三邊總督), and then submitted to the Ming court. ${}^{221}$${}^{221}$^(221){ }^{221} The uniformity in the size, structure, contents, painting style, and the estimated date of production suggest that all these atlases were likely produced by a homogeneous group of officials, if not commissioned by the same person. ${}^{222}$${}^{222}$^(222){ }^{222} The format of the atlases facilitates viewers’ close investigation of each depicted area. All these atlases are the same size of 52 by 90 centimeters, significantly larger than contemporaneous printed books or gazetteers, which averaged 21.1 by 14.6 cm , and were supposed to be read page by page on a flat surface. All these maps are paired with textual essays, most of which appear on the left page of the same sheet with the map on the right page. ${}^{223}$${}^{223}$^(223){ }^{223} As the 1546 guidelines requested, these textual essays provide detailed information about enemies’ invasion routes, defense strategies according to each invasion scenario, and the areas of regular scouting, as we see in figure 49.
${}^{221}$${}^{221}$^(221){ }^{221} 这些地图册的制作者可能是陕西三边总督，也可能是该地区的高级军事官员，然后提交给明朝朝廷。 ${}^{221}$${}^{221}$^(221){ }^{221} 这些图册在尺寸、结构、内容、绘画风格和估计制作日期方面的统一性表明，所有这些图册即使不是受同一人委托，也很可能是由同一批官员制作的。 ${}^{222}$${}^{222}$^(222){ }^{222} 地图册的格式便于观众仔细观察每个描绘的区域。所有这些地图集的尺寸相同，都是 52 x 90 厘米，比同时代的印刷书籍或地名录要大得多，后者的平均尺寸为 21.1 x 14.6 厘米，而且应该在平面上逐页阅读。所有这些地图都配有文字说明，其中大部分文字说明位于同一张纸的左页，地图位于右页。 ${}^{223}$${}^{223}$^(223){ }^{223} 正如 1546 年指南所要求的，这些文字说明提供了有关敌人入侵路线的详细信息、根据每种入侵情况制定的防御策略以及常规侦察区域，如图 49 所示。

The atlases’ depiction of military features looks similar to that in the Gansu difang tu, indicating the bureaucratic standardization of cartographic symbols. In figure 49, the map on the page titled “Illustrated Essay on Fortress Hongchengzi” (紅城子圖說) highlight the presence of Hongchengzi Fortress, the nearby major city of Lanzhou, and the Golden City Gate（金城關） that defended the entrance to Lanzhou across the Zhenyuan Bridge on the Yellow River. ${}^{224}$${}^{224}$^(224){ }^{224} In an enlarged size compared to other defense facilities and landscape features, all three places are colored in vivid red, surrounded by elaborate walls with white parapets. Other major courier stations and nearby fortresses are painted in light pinkish red in the shape of an enclosed wall. Beacon towers are depicted as tall white buildings with flags on the top of hills, arranged in a line parallel to a narrow defensive trench right in front, called “double-lined trenches” (erdao gou 二道溝). ${}^{225}\mathrm{A}$${}^{225}\mathrm{A}$^(225)A{ }^{225} \mathrm{~A} series of mountains and river streams are painted in blue and green, except for the Yellow River, which is colored yellow in a distinct wavy pattern.
地图集对军事特征的描绘与甘肃地方图相似，表明官僚对制图符号的标准化。图 49 中，"红城子要塞图说 "页面上的地图突出了红城子要塞、附近的主要城市兰州以及金城关的存在。 ${}^{224}$${}^{224}$^(224){ }^{224} "金城关 "是兰州的门户，横跨黄河上的镇远桥。 ${}^{224}$${}^{224}$^(224){ }^{224} 与其他防御设施和景观相比，这三个地方的规模都有所扩大，均以鲜艳的红色为主色调，四周筑有精致的城墙和白色护栏。其他主要信使站和附近的堡垒则被涂成浅粉红色，呈围墙状。烽火台被描绘成山顶上插着旗帜的高大白色建筑，与正前方的狭窄防御壕沟平行排列，称为 "二道沟"。 ${}^{225}\mathrm{A}$${}^{225}\mathrm{A}$^(225)A{ }^{225} \mathrm{~A} 除了黄河被涂成明显的波浪形黄色外，其他山脉和河流都被涂成蓝色和绿色。

However, mountains and rivers are miniaturized in the Shaanxi four atlases in contrast to the relatively abundant and spatially more dominant mountains and river streams in the Gansu difang $tu$$tu$tut u. In figure 49, rather than appearing all over the place, mountains are concentrated on the lower part of the map on both sides of the Yellow River and the Zhuanglang River mouth, on the upper-middle part parallel to the trench and the beacon towers, and on the top right with the name of Mount Xiaosong. Such a selective use of mountain images provides a general sense of the changing altitudes of local terrain, which can be further specified when combined with other
然而，与甘肃地方志 $tu$$tu$tut u 中相对丰富且在空间上更占优势的山川相比，陕西四图中的山川却显得微不足道。在图 49 中，山峦并没有出现在地图的所有地方，而是集中在地图下部的黄河和庄浪河口两侧、与壕沟和烽火台平行的中上部以及右上方的小松山。这样有选择性地使用山地图像，可以大致了解当地地形的海拔高度变化，如果结合其他

place names, such as the Zhuanglang River mouth or fortresses. Beyond the trench three bodies of water are identified respectively as Fujia 傅家水, Lujing’er 蘆井兒 and Xia’er 硤兒水, all of which are mentioned in the textual essay explaining the enemies’ invasion routes. ${}^{226}$${}^{226}$^(226){ }^{226} $$$$qquad\qquad
地名，如庄浪河口或堡寨。壕沟外有三条水域，分别是傅家水、芦井儿和厦儿水，在说明敌人入侵路线的文章中都有提及。 ${}^{226}$${}^{226}$^(226){ }^{226} $$$$qquad\qquad 。
Illustrated landscape features attained further meaning when being read side by side with the accompanied text. The textual essay is structured in a similar way throughout the atlases, always beginning with a general explanation, such as the character of the local terrain, the difficulty of defending the region, the official(s) in charge, and the jurisdictional boundaries with neighboring units. Then the text moves on to detailed descriptions of enemy invasion routes and the role of each military unit in defense under each invasion scenario. For instance, in figure 49, the textual essay attached to the Hongchengzi Fortress map reads as follows:
图文并茂的地貌特征在与附带的文字一起阅读时更有意义。整本地图集的文字结构大同小异，总是以概括性说明开始，如当地地形的特点、防守该地区的难度、负责官员以及与邻近单位的管辖边界。然后再详细描述敌人的入侵路线，以及每个军事单位在每种入侵情况下的防御作用。例如，在图 49 中，洪城子要塞地图所附的文字说明如下：

The area within the jurisdiction of Hongchengzi Fortress is extremely difficult to defend since it is wide and has many routes from the mountains. [The court] established a commandant (shoubei) as the official [in charge]. Toward the east, its jurisdiction borders Jinglu 靖虜 Garrison at Lake Lutang 蘆塘湖. Enemies come either from (1) the direction of Lake Lutang, Benhong Gorge 奔紅溝, Fujia and Xia’er waters; or (2) the direction of Sunjia Gorge 孫家溝, Huoshao Gorge 火燒溝, and Cao’er Village槽兒冲. Once they pass the boundary (bian) and reach Nandatong Station 南大通 and Hongchengzi Fortress, through a wide passage they would run toward the west. Passing by the areas of Benfei 奔費 and Jiasha Gorges,家沙溝, Keke Stream 可可川, and Honggu Fort 紅古城，they would cross the Datong River 大通河 and invade the areas of Nuanzhou 暖州 and Honghua Temple 弘化寺. [If then,] soldiers and horses in Xining [Army A] should set up defense at the mouth of the Xiachuan River 下川口 to block enemies, while those in Zhuanglang, Hongchengzi, and Zhenkang 鎮羌 [Army B], should react according to the enemies’ movements. (1) If enemies pass Qingyangtou
洪城子寨辖地广，山路多，极难防守。[朝廷]设指挥使（守备）为[主管]官员。其辖区向东与蘆塘湖的京虜驻军接壤。敌人或来自（1）泸塘湖、奔红峡、付家水、峡儿水方向；或来自（2）孙家峡、火烧峡、草儿村槽儿冲方向。穿过界（边），到达南大通站和红城子炮台后，穿过一条宽阔的通道，向西奔去。经过本飞奔费、贾沙峡、家沙沟、可可溪、可可川、红古堡等地，他们将渡过大通河，攻入暖州、弘化寺等地。[若是如此，]西宁[甲军]兵马应在下川口设防阻敌，庄浪、红城子、镇羌[乙军]兵马则应根据敌军动向做出反应。(1) 如果敌人通过青羊头

Ridge 青羊頭嶺 and Shiban Gorge 石板溝, Army B must attack from the rear, along with Army A from the front side. (2) Once the enemies raid the Keke stream, Army B must wait to attack until the enemies spread out in the rugged terrain…(3) One possible route of invasion is from Xia’er Water, passing through Lujing’er Water, Xianshui(he) Tower 鹹水河, and the relay station of Kushuiwan 苦水灣. Then the enemies would follow a big road at the river mouth of Zhuanglang. In this case, they would raid the towers of Zhangjia 張家臺 and Tuojia 拓家臺. Or if they cross the Yellow River, they would raid the areas of Lanzhou, Jijitan Fortress 積積灘堡 and Aganxian Fortress阿干峴. In this case, the troops at Hongchengzi should cooperate with those at Zhuanglang garrison to chase after the enemies to the river mouth [of Zhuanglang], taking advantage of the rugged terrain and intercepting the enemies. If then, there is no chance that all of the enemies can retreat, and so our soldiers can expect a major victory. ${}^{227}$${}^{227}$^(227){ }^{227}
青羊头岭和石板峡，乙军必须从后方进攻，甲军从正面进攻。(2)一旦敌人突袭柯柯溪，乙军必须等待敌人在崎岖的地形中分散开来再进攻......(3)一条可能的进攻路线是从霞儿水出发，途经泸井儿水、咸水河、苦水湾中继站。然后，敌人会沿着庄浪河口的一条大路前进。在这种情况下，他们会袭击张家台和拓家台的碉堡。如果他们渡过黄河，则会突袭兰州、积石滩堡和阿干县城峴。在这种情况下，红城子的部队应该配合庄浪守军追击敌人到[庄浪]河口，利用崎岖的地形截击敌人。如果这样，敌人就不可能全部撤退，我军就可望取得重大胜利。 ${}^{227}$${}^{227}$^(227){ }^{227}

Although not included in the above quotation, other essays include additional information about local topography that is useful for military actions, such as possible passages, river-crossings, and drinking-water sources for soldiers and animals used by either the Ming or the enemies’ side.
虽然上述引文中没有提及，但其他文章中也包含了对军事行动有用的当地地形的其他信息，如可能的通道、渡河口、明朝或敌方士兵和牲畜使用的饮用水源等。

The map’s other function was to document evolving defensive walls and trenches, especially in the areas proximate to the Ordos plateau. ${}^{228}$${}^{228}$^(228){ }^{228} Most maps in Ningxia and Shaanxi album B illustrate walls, trenches, and beacon towers attached to walls as in figure 50. The entrances of waterway, gates, and secret gates are marked on the walls. Combined with the visual emphasis on linear structures on the maps, the essays attached to these maps testify to the increasing importance of linear structures in devising defense strategies in these areas. Places are
地图的另一个功能是记录不断演变的防御墙和壕沟，尤其是在靠近鄂尔多斯高原的地区。 ${}^{228}$${}^{228}$^(228){ }^{228} 宁夏和陕西专辑 B 中的大多数地图都绘有城墙、壕沟和附在城墙上的烽火台，如图 50 所示。城墙上标有水道、城门和暗门的入口。结合地图上对线性结构的直观强调，这些地图所附的文章证明了线性结构在这些地区制定防御战略中日益重要的作用。地点是
${}^{227}$${}^{227}$^(227){ }^{227} Gansu zhen zhanshou tulüe, “Hongchengzi tushuo.”
${}^{227}$${}^{227}$^(227){ }^{227} 甘肃镇守使图录，"洪城子图录"。
${}^{228}$${}^{228}$^(228){ }^{228} For the building of extended walls and trenches and their cartographic representations, see chapters three and four.
${}^{228}$${}^{228}$^(228){ }^{228} 关于扩建城墙和壕沟的建造及其制图，请参见第三章和第四章。
categorized as those located outside of the outer wall (邊外地) or identified in terms of their distance from the walls (離邊…里). ${}^{229}$${}^{229}$^(229){ }^{229}
这些建筑被归类为位于外墙外（边外地）的建筑，或根据其与外墙的距离（离边......里）来确定的建筑。 ${}^{229}$${}^{229}$^(229){ }^{229}

These defensive structures also function as markers of jurisdictional boundaries between the territories inside and outside of the Ming, and between neighboring military units within the Ming territory. The enemies are considered invaders when they pass through the outer wall, and most placenames for soldiers’ actions are located south of the inner wall. This may reflect the reality that there was usually a time lag between when invasion alerts were signaled at the beacon walls located to the north and the mobilization of soldiers for military actions while enemies were heading south. Accordingly, the area between the outer and inner walls served as a buffer zone, which was created and maintained by the Ming for defense purposes. The area beyond the outer wall was treated as the enemy space beyond the Ming frontline. Ming defense efforts concentrated on the area below the inner wall. Combined with the repeated presence of walls and trenches throughout the album, the images of stone stelae inserted on the west and east side of each map to indicate the names of neighboring defense units such as forts and garrisons give a sense of spatial continuity, even though individual maps are discrete from each other, divided by pages and following a different directionality depending on the location of each fortress, which is placed at the center of the map. ${}^{230}$${}^{230}$^(230){ }^{230} The walls also provided a reference to
这些防御建筑也是明朝内外领土之间以及明朝领土内相邻军事单位之间管辖边界的标志。敌人通过外墙时被视为入侵者，而士兵行动的地名大多位于内墙以南。这可能反映了这样一个现实，即从位于北面的烽火台发出入侵警报到动员士兵采取军事行动之间通常会有一个时间差，而此时敌人正在南下。因此，外墙和内墙之间的区域成为缓冲区，由明朝出于防御目的而设立和维护。外墙以外的区域被视为明朝前线以外的敌方空间。明朝的防御工作主要集中在内墙以下的区域。整本画册反复出现城墙和壕沟，每幅地图的西侧和东侧都插入了石碑图像，以标明邻近防御单位（如堡垒和驻军）的名称，这给人一种空间连续性的感觉，尽管每幅地图都是相互独立的，被页面分割开来，并根据位于地图中心的每个堡垒的位置而遵循不同的方向性。 ${}^{230}$${}^{230}$^(230){ }^{230} 城墙还提供了以下参照物

measure the width of the area that each defense unit covers. In figure 50, the textual annotation included on the map records the length of both the Outer Wall (Dabian 大邊) and the Inner Wall (Fuli bian 服裏邊) within the depicted area of each map. ${}^{231}$${}^{231}$^(231){ }^{231}
测量每个防御单元覆盖区域的宽度。在图 50 中，地图上的文字注释记录了每幅地图所描绘区域内的外墙（大边）和内墙（服里边）的长度。 ${}^{231}$${}^{231}$^(231){ }^{231}

The illustrated topography was no longer the awe-inspiring landscape seen in the Gansu difang $tu$$tu$tut u. Instead, it was transformed into one that awaits observation, planning, and alteration as needed, to block the enemies’ invasion routes or to establish defensive structures. The main purpose of the atlases was to enable court officials to become familiar with the distant frontier geography and to devise strategies that can make best use of the topographic advantages. This information was not likely produced all at once, but instead must have been gradually accumulated with the participation of soldiers and military officers native to the region. What was new in the mid-sixteenth century atlases was the court’s ambition to collect such information, which previously had been available only locally, and to reproduce it according to the standard bureaucratic language to make it legible for court officials. Such knowledge production enabled the court’s (sense of) micromanagement of military actions in distant frontiers. Due to the great distance and lack of modern communication technology, it was still impossible for the court officials to make battlefield decisions in frontiers under attack. However, based on the topographic information it collected, the court compiled and confirmed a set of strategies according to each invasion scenario, which imparted a sense of centralized
图中的地形不再是《甘肃地芳》 $tu$$tu$tut u 中令人敬畏的景观。取而代之的是有待观察、规划和根据需要修改的地形，以阻断敌人的入侵路线或建立防御结构。绘制地图的主要目的是让朝廷官员熟悉遥远的边疆地理，并制定能充分利用地形优势的战略。这些信息不可能一下子就制作出来，而一定是在当地士兵和军官的参与下逐渐积累起来的。16 世纪中叶地图集的新颖之处在于，宫廷雄心勃勃地收集这些以前只能在当地获得的信息，并按照标准的官僚语言将其复制，使宫廷官员能够清晰地阅读。这种知识生产使朝廷能够对遥远边疆的军事行动进行（意义上的）微观管理。由于距离遥远且缺乏现代通信技术，朝廷官员仍无法在受到攻击的边境地区做出战场决策。然而，朝廷会根据收集到的地形信息，按照每种入侵情况编制并确认一套战略，这给人一种中央集权的感觉。

control. Maps became a major vehicle for the court to plan, guide, and supervise military actions in areas located far from the center.
控制。地图成为朝廷在远离中央的地区规划、指导和监督军事行动的主要工具。

In a sense, the court’s frustration about inaccurate and outdated knowledge, mentioned earlier in this section, was another side of the same coin with its growing ambition for thorough territorial knowledge and control based on such knowledge. More precisely, the frustration was not simply a by-product generated during the process of the court’s knowledge production, but also the impetus for the court’s active archiving of detailed frontier geographical information. Some examples reveal that submitted maps were actually utilized in the court’s policymaking. In the preface of Maps and Essays of Nine Defense Commands（Jiubian tushuo 九邊圖說）, the late sixteenth century atlas of northern frontier garrisons and fortresses compiled by the Ministry of War in 1568, the minister Huo Ji 霍冀 (1516-75) mentioned that the maps were used by officials to assess difficulties throughout the northern frontier defense zone and to provide new criteria for evaluating the performance of the officers in charge of military units. ${}^{232}$${}^{232}$^(232){ }^{232}
从某种意义上说，本节前面提到的法院对不准确和过时知识的失望，与法院日益增长的对领土的全面了解和基于这种知识的控制的雄心是同一枚硬币的另一面。更确切地说，挫折感不仅仅是法院知识生产过程中产生的副产品，也是法院积极将详细的边疆地理信息存档的动力。一些例子表明，提交的地图在法院决策中得到了实际利用。在《九边图说》（兵部于 1568 年编纂的 16 世纪晚期北方边疆驻军和要塞地图集）的序言中，兵部尚书霍冀（Huo Ji Ho冀）说："《九边图说》是中国历史上最早的边疆地图集、兵部尚书霍冀（1516-75 年）提到，官员们用这些地图来评估整个北部边疆防区的困难，并为评价负责军事单位的官员的表现提供新的标准。 ${}^{232}$${}^{232}$^(232){ }^{232}

In contrast to the modern scholarly understanding that knowledge was expected to increase power, we need to be cautious in thinking of the relation between knowledge and power
与现代学者认为知识可望增加权力的理解不同，我们在思考知识与权力之间的关系时需要谨慎行事

in this case. ${}^{233}$${}^{233}$^(233){ }^{233} We had little evidence to believe that such knowledge actually improved the military power of the Ming state in terms of better supervision of its forces or enhanced defense capabilities in the northern frontiers. I argue instead that these maps created an illusion of control for the Ming court. Illusive control should not be misconstrued as no control, or the failure of control, in contrast to actual control. Instead, this illusion of control is no less important in conceptualizing power in the sense that the Ming court developed an emerging territorial notion of power in frontiers using the media of maps. As seen in the earlier Gansu difang tu, military power had been previously imagined as a bureaucratic network of military officers in charge of local fortresses, strongholds and garrisons. Sovereign power was still imagined as condensed into a set of fortified towns, although their numbers were increasing. The court lacked interest in knowing the geography of their surrounding two-dimensional spaces. In the new atlases, a fortified town still symbolized imperial sovereignty, serving as the central focal point of a broader regional defense network. While the spatial visualization of the frontiers still relied largely on such military networks, each fortress was now expected to actualize control over broader areas outside of its confining wall. The intensified mapping and archiving of local topographic information changed how the Ming court imagined its control over frontier spaces. It facilitated the court’s increasing familiarity with frontier topography and its ability to represent space uniformly according to standardized pictorial elements. The regularized accumulation and archiving of spatial information shaped a new territorial discourse of power.
在这种情况下。 ${}^{233}$${}^{233}$^(233){ }^{233} 我们几乎没有证据可以相信，这些知识实际上提高了明朝的军事实力，更好地监督了明朝的军队，或增强了北方边疆的防御能力。相反，我认为这些地图为明朝朝廷制造了一种控制的假象。与实际控制相比，幻觉控制不应被误解为没有控制或控制失败。相反，这种控制幻觉在权力概念化方面的重要性丝毫不亚于明朝朝廷利用地图媒介发展出的新兴边疆领土权力概念。正如早先的《甘肃地方图》所显示的，军事权力之前被想象为由负责地方要塞、据点和驻军的军官组成的官僚网络。主权仍被想象为凝结在一系列坚固的城镇中，尽管这些城镇的数量在不断增加。宫廷对了解其周围二维空间的地理情况缺乏兴趣。在新的地图集中，坚固的城镇仍然是帝国主权的象征，是更广泛的地区防御网络的中心点。虽然边境的空间可视化在很大程度上仍依赖于此类军事网络，但现在人们期望每个要塞都能实际控制其围墙外的更广阔区域。加强对当地地形信息的测绘和存档改变了明朝朝廷对边疆空间控制的想象。它促进了朝廷对边疆地形的日益熟悉，以及朝廷根据标准化图像元素统一表现空间的能力。 空间信息的规范化积累和存档形成了一种新的领土权力话语。

The mid-fifteenth century Tumu debacle is considered a major turning point from aggressive toward defensive strategic thinking among Ming historians. However, the current scholarly emphasis on crises, both foreign threats and the deterioration of domestic military institutions, focuses on the narrative of the Ming military failure, which explains the eventual conquest by another northern foreign enemy, the Jurchens. Instead, this chapter focuses on the neglected issue of how this military crisis brought about changes in envisioning the relationship between the imperial center and frontiers.
十五世纪中叶的土木之变被明代史学家认为是明朝战略思想从侵略性转向防御性的一个重要转折点。然而，目前学术界对危机（包括外国威胁和国内军事机构的衰败）的强调，主要集中在明朝军事失败的叙事上，这也解释了明朝最终被另一个北方外敌女真征服的原因。相反，本章关注的是一个被忽视的问题，即这场军事危机如何改变了对帝国中心与边疆之间关系的设想。

In order to do so, this chapter traces back to the early Ming imperial practices of collecting and managing the information of frontiers and compares them with the later sixteenth century practices. Section Two investigates the efforts of the central bureaucracy to gather and update the information of strategic resources in frontiers during the late fourteenth and early fifteenth century. Under the circumstance that Ming state power was restrained by budget constraints and overwhelmed by local and indigenous contenders, the court concentrated on maintaining its power in specific strategic places by managing the key resources, such as registered soldiers, horses, and agricultural colonies (that generated revenue and provisions). In addition, the court dedicated efforts to secure constant and reliable communication and the circulation of goods and people between the imperial center and frontiers. Section Three examines courier route maps that enabled the court to supervise the official transportation networks. These maps enabled viewers to visualize emperor’s power of controlling movements in the depicted space through their physical engagement with the handscroll, as the actions of
为此，本章追溯了明朝早期收集和管理边疆信息的做法，并将其与 16 世纪后期的做法进行了比较。第二节探讨了十四世纪末十五世纪初中央官僚机构收集和更新边疆战略资源信息的努力。在明朝国家权力受到预算限制、地方和土著竞争者压得喘不过气的情况下，朝廷通过管理关键资源，如注册士兵、马匹和农业殖民地（产生收入和供给），集中精力维持其在特定战略要地的权力。此外，朝廷还致力于确保帝国中心与边境之间持续可靠的通信以及货物和人员的流通。第三部分研究了使朝廷能够监督官方交通网络的信使路线地图。这些地图使观众能够通过与手卷的实际接触，直观地感受到皇帝在所描绘的空间内控制行动的权力，因为手卷上的行动都是由皇帝控制的。
unrolling and rolling the scroll imitate the imagined official travel from the center to the periphery, and vice versa. These courier route maps show how the imperial power was conceptualized based on the court’s monopolized control of strategic resources, which ensured the empire-wide circulation of goods, people, and information through official transportation networks.
卷轴的展开和滚动模仿了想象中从中心到外围的官方旅行，反之亦然。这些信使线路图显示了皇权是如何被概念化的，它基于宫廷对战略资源的垄断控制，通过官方交通网络确保整个帝国的货物、人员和信息流通。

From the mid-sixteenth century, constant military crises instigated and intensified new types of mapping projects. Maps not only enabled the court to supervise the ever-increasing number of defensive structures, but also to newly imagine the imperial power over the depicted frontier spaces-which I call the territorialization of the power. Through repeated map submission and ever-refined bureaucratic regulations, the court gathered more and more detailed information about local territories, and based on this information planned and directed the way it engaged with local landscapes and people. The map helped the court to envision the frontier space based on the imposed administrative and military networks, and to claim its domination over the depicted territories, when it was disputed by multiple contenders.
从 16 世纪中叶开始，持续不断的军事危机激发并强化了新型制图项目。地图不仅使朝廷能够监督数量不断增加的防御建筑，还能对所描绘的边境空间进行新的帝国权力想象--我称之为权力的领土化。通过反复提交地图和不断完善的官僚规章，朝廷收集到了越来越多关于当地领土的详细信息，并根据这些信息规划和指导朝廷与当地景观和人民打交道的方式。地图帮助朝廷在强加的行政和军事网络基础上设想边境空间，并在多个竞争者提出争议时宣称其对所描绘领土的统治权。

I have a mixed answer to the question of whether the court’s such mapping projects led to a change in imagining frontier spaces among the people outside the court. Although it was increasingly obsolete, the very ideal of secrecy prohibited the circulation of above-mentioned maps, and therefore their direct impact was limited. However, the very officials in charge of producing and submitting these maps took the lead in the unauthorized circulation and printing of these maps. Eventually, these court-originated maps provided a shared image and vocabulary for a broader scholar-official readership interested in frontier affairs, which will be discussed in chapters three and four of this dissertation.
对于宫廷的此类地图绘制项目是否改变了宫廷之外的人们对边疆空间的想象，我的答案不一。虽然这种做法已日渐过时，但保密的理想本身禁止了上述地图的流通，因此它们的直接影响是有限的。然而，负责绘制和提交这些地图的官员却带头未经授权分发和印刷这些地图。最终，这些源自宫廷的地图为更广泛的对边疆事务感兴趣的学者-官员读者群提供了共享的图像和词汇，本论文的第三章和第四章将对此进行讨论。

At the final section of this chapter, I briefly mention how the extensive wall (in Chinese Changcheng, often translated as the Great Wall) in the Shaanxi Four Atlases was depicted with the emphasis on its roles in defining jurisdictional boundaries and planning military actions (but not so much as effective deterrence or a defensive structure against approaching enemies). The next two chapters investigate how the materiality and roles of the extensive wall had been changed throughout the fifteenth and sixteenth centuries, and how the cartographic transformation in the way of depicting this wall brought about a fundamental shift in imagining the northern frontier.
在本章的最后一节，我简要地提到了《陕西四图》中如何描绘广袤的城墙（汉语为 "长城"，常被译为 "长城"），重点强调了它在界定管辖边界和规划军事行动方面的作用（而不是有效的威慑或抵御敌人逼近的防御结构）。接下来的两章将探讨在整个十五世纪和十六世纪，广袤的城墙的物质性和作用是如何发生变化的，以及描绘城墙的地图学方式的转变是如何给人们对北方边疆的想象带来根本性转变的。

PART TWO. DRAWING AND CREATING THE GREAT WALL
第二部分绘制和创建长城

The Great Wall of China (wanli changcheng 萬里長城) is one of the most well-known monumental landmarks in human history. Its impressive physicality is advertised in modern visual media including photographs and videos and testified over and over by tourists’ visit to actual sites. The Badaling stretch is one of the most visited sections of the Great Wall. Walking on top of the wall, which is approximately 7.8 meter high, tourists can enjoy a panoramic view from the wall that stretches far out across the mountain ridges beyond the horizon (figure 52). ${}^{234}$${}^{234}$^(234){ }^{234} Despite its recent history of restoration, its imposing presence both physically and as a conceptual landmark testifies the timeless division between Sinic agrarian-sedentary society and nomadic people of the steppe. ${}^{235}$${}^{235}$^(235){ }^{235}
中国万里长城是人类历史上最著名的地标性建筑之一。现代视觉媒体，包括照片和视频，都在宣传它令人印象深刻的实物，而游客在实际参观中也一再证明了这一点。八达岭长城是游人最多的一段长城。站在高约 7.8 米的城墙顶上，游客可以欣赏到城墙上的全景，远处的山脊延伸到地平线之外（图 52）。 ${}^{234}$${}^{234}$^(234){ }^{234} 尽管最近进行了修复，但它作为实体和概念地标的宏伟存在，证明了中原农牧社会与草原游牧民族之间永恒的分界线。 ${}^{235}$${}^{235}$^(235){ }^{235}

保護工程） in 2005. Under the name of preservation, each provincial government investigated,
保护工程）2005 年。各省政府以保护为名进行调查、

Opposed to this political propaganda, several scholars indicate that such a monumental structure as we see in the Badaling stretch did not exist before the sixteenth century. In The Great Wall of China: From History to Myth, Waldron calls the prevalent idea of an ancient origin and two-millennia existence of the Great Wall a myth. ${}^{236}$${}^{236}$^(236){ }^{236} Instead, most of the Great Wall that we see now is the Ming or more modern (re)construction. Before the mid-sixteenth century, the building of linear defensive structures including walls was of a smaller scale and limited to local defensive practices, without being considered as a part of the empire-wide frontline. Recent archaeological survey in China revises certain aspects of Waldron’s arguments by showing that some physical remains of older walls survived into the Song, and some of these pre-Song walls
与这种政治宣传相反，一些学者指出，像我们在八达岭看到的这种纪念性建筑在 16 世纪之前并不存在。在《中国长城：在《中国长城：从历史到神话》一书中，Waldron 将长城起源古老、存在两千年的普遍观点称为神话。 ${}^{236}$${}^{236}$^(236){ }^{236} 相反，我们现在看到的长城大多是明长城或更现代的（再）建筑。16 世纪中叶以前，包括城墙在内的线性防御结构的建造规模较小，仅限于地方防御实践，并不被视为帝国全境前线的一部分。最近在中国进行的考古调查修正了华德隆论点的某些方面，表明一些较早城墙的实物遗迹一直保留到宋代，其中一些宋代以前的城墙
discovered, and reconstructed what was considered as remains of the Great Wall. This project brought about controversy, as overseas scholars criticized that under the name of cultural preservation, the government often imposed the category of the Great Wall over architectural remains of technological and social diversity. For the details of this criticism, see Carlos Rojas, The Great Wall (Harvard University Press, 2010).; Senghyun Hong, “Chunggugŭi changsŏng boho gongjŏng gwa changsŏng yŏn’gu ŭi saeroun kyŏnghyang,” Tongbuga yŏksa nonch’ong 45, no. 45 (2014): 339-401.
发现并重建了被认为是长城的遗迹。这一项目引发了争议，海外学者批评说，在文化保护的名义下，政府往往将长城的范畴强加于技术和社会多样性的建筑遗迹之上。有关这一批评的详细内容，请参见卡洛斯-罗哈斯，《长城》（哈佛大学出版社，2010 年）；洪森贤，《中国长城》，第 45 期（2014 年）："中国长城"。45 (2014):339-401.
${}^{236}$${}^{236}$^(236){ }^{236} Arthur Waldron, The Great Wall of China. While there is still a debate on the definition of the Great Wall, one of the leading Chinese architectural historians, Luo Zhewen, defines the Great Wall as (1) long, (2) not circular, and (3) defensive in nature. Spindler presents similar, low-threshold criteria of “a linear manned border defense system, consisting, in key points, of unenclosed walls.” Luo Zhewen, “Fang Xiangxi Fenghuang Ming Changcheng,” Guangming Ribao, February 6, 2001.; David Spindler, “A Twice-Scorned Mongol Woman, the Raid of 1576, and the Building of the Brick Great Wall,” Ming Studies 60 (2009): 69-70.
${}^{236}$${}^{236}$^(236){ }^{236} Arthur Waldron，《中国长城》。虽然关于长城的定义仍有争议，但中国著名建筑史学家罗哲文将长城定义为：(1) 长，(2) 非环形，(3) 防御性。斯平德勒提出了类似的低门槛标准，即 "一个线性的有人驻守的边防系统，在关键部位由非封闭的城墙组成"。罗哲文，《方湘西凤凰明长城》，《光明日报》，2001 年 2 月 6 日；戴维-斯平德勒，《被两次蔑视的蒙古妇女、1576 年的突袭和砖砌长城的建造》，《明代研究》第 60 期（2009 年）：69-70.
were not limited to local efforts but substantial both in geographical scope and size. ${}^{237}$${}^{237}$^(237){ }^{237} However, such trans-local wall-building efforts seem to be halted between Tang and mid-Ming periods. The Tang adopted a defense-in-depth military strategy, stationing mounted troops in fortified cities of forward garrisons without building linear walls. ${}^{238}$${}^{238}$^(238){ }^{238} While the Song attempted to construct lengthy and continuous borderlines with the northern polities, wall construction was avoided due to the negative historical memory of the ancient Great Wall, which was often associated with the state’s overtaxing of the populace that led to the early demise of the Qin dynasty. ${}^{239}$${}^{239}$^(239){ }^{239} Having strong cavalry forces from nomadic origins and as a unifier of the steppe and agricultural worlds, the Mongol Yuan court did not need to build such structures. During the earlier Ming period, there were some attempts to build the wall, while most of wall sections were made of earth and easily decayed within a few decades. ${}^{240}$${}^{240}$^(240){ }^{240} A more monumental defensive structure, made of bricks or stone with crenellations and fighting towers, started to be built in response to intensified military crises in the mid-sixteenth century.
${}^{237}$${}^{237}$^(237){ }^{237} 然而，这种跨地方的筑墙活动似乎在唐朝至明朝中期停止了。 ${}^{237}$${}^{237}$^(237){ }^{237} 然而，这种跨地方的城墙修建工作似乎在唐朝至明朝中期停止了。唐朝采取了纵深防御的军事战略，在前沿驻军的坚城中驻扎骑兵，而不修建线性城墙。 ${}^{238}$${}^{238}$^(238){ }^{238} 虽然宋朝试图与北方诸国构筑漫长而连续的边界线，但由于古代长城的负面历史记忆往往与国家对民众的过度征税联系在一起，导致秦朝早亡，因此宋朝避免了修建城墙。 ${}^{239}$${}^{239}$^(239){ }^{239} 蒙古元朝出身游牧民族，拥有强大的骑兵部队，又是草原和农业世界的统一者，因此不需要修建此类建筑。在明朝早期，曾有过一些修建城墙的尝试，但大部分城墙都是用土筑成的，几十年后就容易腐烂。 ${}^{240}$${}^{240}$^(240){ }^{240} 16世纪中叶，为了应对日益加剧的军事危机，人们开始建造用砖或石头砌成的、带有垛口和战斗塔的更不朽的防御建筑。