Migration and Land Exploitation from Yuan to Qing Dynasties: Insights from 252 Traditional Villages in Hunan, China

Abstract

This paper investigates 252 traditional villages in Hunan, China, and uses ArcGIS and Geodetector to analyze village expansion and land exploitation from the Yuan dynasty to the Qing dynasty caused by factors such as migration during China’s middle and late imperial periods. This article demonstrates the development of land by the earlier settlers of ancient Chinese villages and shows the progression from easily exploited flatlands to more difficult-to-exploit mountainous areas. It also shows that early settlers relied more on natural factors when establishing their villages during the Yuan Dynasty but relied more on factors such as transportation due to the migration policies introduced during the Ming and Qing Dynasties. This paper will help us to determine the migration dynamics of ethnic groups and the distribution of settlements in the region (Hunan region) during the Yuan, Ming, and Qing eras.

1. Introduction

Villages are a type of settlement that began to form after the end of prehistoric hunter–gatherer societies and the introduction of farming. They are also where farmers live, work, and reproduce [1]. These traditional villages are a record of human activities and interactions with both natural and social environments and the contest between man, nature, and society in the imperial period of China. Traditional villages are the typical representatives of Chinese villages. As early as 2012, the Chinese government clearly defined “traditional villages”, pointing out that traditional villages have three characteristics, i.e., the architectural style is relatively well preserved, the site and layout maintain traditional features, and the cultural heritage is still active [2]. In November 2022, China’s Ministry of Housing and Construction announced that 6819 villages (five batches in total) had been listed as traditional Chinese villages, of which there are 658 in Hunan [3]. As essential carriers of vernacular culture, traditional villages play a crucial role in putting the rural revitalization approach into practice. Therefore, under the current policy requirements of beautiful countryside construction and the rural revitalization strategy, the study of traditional villages has been highly anticipated in China in recent years. Scholars have studied traditional villages in terms of their spatial characteristics, cultural landscape, value assessment, conservation, and tourism development, building a relatively comprehensive knowledge system of cultural heritage regarding traditional villages [4,5,6,7,8,9,10]. Furthermore, scholars have studied the spatial distribution of traditional villages in Hunan using kernel density analysis, the nearest neighbor index, and spatial autocorrelation analysis [11,12,13,14]. However, few researchers have studied the historical stages of the development of traditional villages in Hunan, and little is known about the historical background reflected by the evolutionary pattern and changes in traditional villages.

Responding to the above-mentioned shortcomings, this paper analyzes the evolution of the clustering area of traditional villages from the perspective of historical geography, focusing on three questions: How were traditional villages located and what were the patterns in Hunan during the Yuan, Ming, and Qing dynasties, respectively? Which factors of traditional villages play a decisive role in the location of Hunan Yuan, Ming, and Qing dynasty villages? What do these evolutionary patterns and influencing factors reveal? These research questions can find the reasons why villages were established and why populations migrated during the Yuan, Ming, and Qing dynasties across the traditional villages in Hunan, China, and can provide a new research perspective for studying population migration in the middle and lower reaches of the Yangtze River during the Ming and Qing dynasties. In addition, the analysis of the above issues can identify the reasons for land exploration by the early settlers of the Hunan region in the Ming and Qing dynasties. This helps to recover the forms of dwelling and labor of the inhabitants of the middle and lower reaches of the Yangtze River during the Middle and Late periods of Imperial Chinese history. It is helpful for us to understand the life of traditional village inhabitants and to better consider the conservation and developmental path of traditional villages from the perspective of their inhabitants.

2. Literature Review

2.1. Traditional Villages

Traditional Chinese villages are mostly located in rural areas, they were constructed from local materials and designed according to the local conditions, and they were often built by folk craftsmen or occupants [15]. This kind of folk, traditional village is a local manifestation of vernacular architecture. According to cultural historian Eugenia Leanren (2020), “vernacular” is a combination of local forms and homegrown features [16]. As a result, traditional villages are often a true and direct reflection of the needs of the people who built them. In the process of building, the user constantly adjusts the forms and principles of the design to the requirements of the surrounding environment (natural and social) [17]. Amos Rapoport (1969) points out that the location of vernacular settlements and dwellings can be limited by the social environment and even change the building methods and techniques as a result. The traditional villages inhabited by human beings are the product of cultural genes [18], and the manifestation of their forms contain much information. The traditional village space was created via the logical growth of the form of clustering, and it is both a form of cultural and material representation. For example, the traditional village can express the intervillage relationships, order, and rules through the layout of the clusters within the form [19]. Analyzing how village communities were established and the specific influential factors can reveal various associations, such as the influence of Feng Shui, culture, etc. [20,21,22]. Additionally, studying the distributional traits and landscape development of traditional villages can show the overall connections between the site of human civilization, its extension, and the physical geography [23,24].

Most of the preserved traditional villages in Hunan were established mainly after the Yuan Dynasty (1279–1368 A.D.). The Yuan (1279–1368 A.D.), Ming (1368–1644 A.D.), and Qing Dynasty (1644–1911 A.D.) were periods of relatively rapid development in regard to traditional China villages. The scale of the villages gradually expanded during the Yuan Dynasty, and the village population increased. The village culture, organizational structure, and clan system matured during the Ming and Qing dynasties [25]. Hunan society changed significantly between the Yuan and Qing Dynasties (1644–1911 A.D.). The historical event of “Jiangxi filling Huguang” in the late Yuan and early Ming Dynasties led to a large number of people migrating to central and western Hunan from the east, which contributed to a significant population increase in western, central, and southern Hunan in the Ming and Qing Dynasties. Ge and Cao (2005) indicated that a large number of people migrated to the mountainous region of western Hunan from eastern and southern Hunan [26]. The migration policy of the early Qing Dynasty government guided the population of Hunan into Sichuan, contributing to the overall trend of Hunan’s population shifting to the west. This created the basis for the social environment for the development and evolution of the traditional villages in Hunan.

In addition, these traditional villages are situated within the natural ecological space of Hunan, where plains, mountains, and hills coexist, particularly in the Yangtze River’s middle reaches, where numerous river systems are located. The location and distribution rules of these traditional villages can be utilized to show how the settlers’ attitudes toward the environment, society, and nature evolved over time. This facilitates the natural conditions for the distribution and evolution of traditional villages in Hunan.

2.2. Land, Population Migration and Traditional Villages

Traditional villages reflect the relationship that the early settlers had with the land. As scattered humans concentrated in villages or settlements, villages became one of the ideal human habitats. Humans manage the land when they establish villages, for example, by centralizing and controlling the development of the land [27]. For communities that produce food on the land, the development of the land around the settlements affects the development of the community. Armstrong’s research (2022) indicates that the Nuchatlaht altered ancient cultural and historical landscapes in order to cultivate food, flora, and fauna in their territory [28]. The villages also demonstrate the relationship between the land, people, and policy. During the Ming and Qing dynasties the development of settlements in China’s mountainous regions was controlled by government policy, and the form of land exploration by the inhabitants was constantly changing [29]. This manifestation was not limited to China. Enclosures by government departments to preserve agriculture and forests as residents settled around land also occurred in England from 1861–1905 [30]. Therefore, the relationship between settlements and land is a mutual one. Researchers can use the combination of historical topographic maps and GIS to quantify and analyze the intensity level of ancient land use, traces of ancient routes, etc., to obtain the scope area of the heritage area and the pattern of settlements [31]. The characteristics of the land where the site is located can also be used to infer the lifestyle of the inhabitants, for example, by demonstrating the difference in lifestyle between the early settlers in the Yangtze and Yellow River basins by analyzing the chemical elements in the environment [32].

The type of land in an environment has a certain controlling effect on excessive population migration; this was especially true in ancient China, which was predominantly agricultural [33]. It has been argued that there is an increasingly clear relationship between traditional agriculture, which relies on land exploitation, and human migration and population movements [34]. Low migration rates have been a characteristic of China’s population for thousands of years, and this is largely related to the Chinese feudal land system [35]. Population change directly affects land use, and when rural populations migrate to the cities, they lose access to the development and use of the land [36,37]. As can be seen, the movement and migration of people brought about productivity and the establishment and development of numerous settlements [38]. However, there are also studies that point out that the change in human settlement, migration, and settlement choice is influenced by the climate, rivers, etc. This shows that there is a correlation between land and population movement and settlement [39].

2.3. GIS and Geodetector

Data management and analysis in GIS provides geographic information data for various types of research, and GIS is an intuitive and effective method for studying of cultural heritage. The data management and analysis tools of ArcGIS provide geographic information data for various types of research and are an intuitive and effective method for preserving cultural heritage [40,41,42,43]. Researchers such as Ryavec (2015), Cui (2021), and other related scholars used a combination of data and historical geographic information to restore ancient Feng Shui, villages, or urban scenes [44,45]. Sprague (2013) used ArcGIS to conduct multiple buffer analyses of rural Japanese landscapes depicted in historical maps surveyed in the 1880s to investigate the spatial patterns on the Kanto Plain near Tokyo during the period of traditional agriculture [46]. Moreno (2022) used ArcGIS to simulate ancient Roman roads and concluded that in the northern region of Rome, the road network used transportation routes to connect remote areas through villages and settlements [47]. The evolutionary characteristics of urban and rural settlements in Kangbashi New District were analyzed by Dong (2020), who conducted an analysis of the spatial evolution and applied kernel density estimation methods using a geographic information system [48]. It can be seen that GIS has matured as a research method for exploring traditional villages.

Geodetector is a new statistical method for detecting spatial heterogeneity and revealing its driving mechanism. Geodetector consists of four modules: risk detection, factor detection, ecological detection, and interaction detection [49]. More directly and accurately than other statistical analyses, geographical detectors demonstrate the interaction and causality of space. Geographical detectors are widely used in geological hazard assessments and resource analyses [50,51,52]. At present, there is much room for their application in exploring the aspects of traditional villages, such as the natural geography, social economy, and cultural life and for constructing an evaluation index system of the research area [53,54,55]. Through assessments via Geodetector, the driving mechanisms of the spatial differences and environmental factors affecting the development of villages can be obtained. Taking Qin (2022) and Tang (2022) as examples, their research involved the use of geographical detectors to undertake a quantitative evaluation of the suitability of the spatial distribution and the developmental laws of the village space [23,56].

In summary, the spatial and influencing factors of traditional villages identified using GIS and Geodetector are practical and can help us understand the historical development of villages at the level of quantitative analysis and grasp the development logic of traditional villages comprehensively.

3. Materials and Methods

3.1. Materials

In this article, the base map was created using the original 1:1,000,000 vector data of Hunan Province from the base map database supplied by the Ministry of Natural Resources, and the latitude and longitude data for the traditional Chinese villages were taken from the Global Change Data Repository [57,58]. The 30 m digital elevation model (DEM) and the river systems of Hunan were gathered from the National Geomatics Center of China [59]; the data on ancient post roads and ancient cities in the Hunan region were taken from CHGIS (The China Historical Geographic Information System) [60,61]. Due to the uncertainty and lack of information about the dynasties in which some traditional villages were built, we could only obtain 252 traditional villages with specific dynasties from the Digital Museum of Traditional Villages of China, local chronicles, and statistics from the Hunan Provincial Department of Housing and Urban–Rural Development. Among them, 82 were from the Yuan Dynasty, 124 from the Ming Dynasty, and 46 were from the Qing Dynasty [62,63]. These traditional villages have been evaluated and assessed by the Ministry of Housing and Construction in terms of architecture, site selection and village pattern, cultural heritage, etc., and have been selected for the list of traditional Chinese villages that have specific architectural, cultural, and historical values [64]. These 252 traditional villages cover nearly all of the prefecture-level cities in Hunan Province and occupy a certain proportion of each region, which is representative of the area. Therefore, these 252 national traditional villages in Hunan Province were selected as the research objects (Figure 1 and Figure 2).

3.2. Research Methods

Firstly, the above data (villages, river systems, ancient post roads, ancient cities, and elevations) were collected and imported into ArcGIS to form a database. Secondly, the spatial autocorrelation in ArcGIS using the Getis-Ord Gi* index to analyze the spatial distribution hotspot areas of traditional villages from the Yuan, Ming, and Qing dynasties and the areas where traditional villages were mainly concentrated and where they were less frequent in the three historical periods can be obtained. This phase of the analysis is the basis of this study, because the clustering of villages represents the agglomeration of a large residential population and labor force. The main distribution areas and directions of the geographical layout of the traditional settlements in Hunan at each historical stage are analyzed through standard deviation ellipse comparison; the proportion of villages that were influenced by rivers and ancient post roads is analyzed using buffer zones. Furthermore, the distance between ancient post roads, ancient cities and villages is analyzed using Euclidean distance. Finally, the data obtained in ArcGIS were exported, and the main influencing factors (slope direction, river system, ancient cities, elevation, ancient post roads) that affected the location of traditional villages in the Yuan, Ming, and Qing dynasties were quantified and analyzed using Geodetector to further obtain the influence values of each influencing factor on the development of villages.

3.2.1. Spatial Autocorrelation and Getis–Ord Gi* Index

Moran’s global index and the Getis-Ord Gi* index are two indexes that can be used to measure the clustering of the spatial units, which is measured using spatial autocorrelation. This study speculates whether the traditional villages in Hunan show an aggregation trend by observing the Moran global index because the aggregation of villages represents the aggregation of a large residential population. The value of Moran’s index is less than 0 for negative correlations, 0 for no correlation, and greater than 0 for positive correlations; the possible spatial clustering patterns are clustered, random, and dispersed. The calculation formula is from [65].

$$\begin{array}{c}\mathrm{M}\mathrm{o}\mathrm{r}\mathrm{a}\mathrm{n}’\mathrm{s} I=n\frac{{{\displaystyle \sum }}_{i=1}^n{{\displaystyle \sum }}_{j=1}^n\left(x_i-\overline{x}\right)\left(x_j-\overline{x}\right)}{{{\displaystyle \sum }}_{i=1}^n{{\displaystyle \sum }}_{j=1}^n{w}_{ij}{{\displaystyle \sum }}_{i=1}^n{\left(x_j-\overline{x}\right)}^2}\end{array}$$

(1)

A local spatial autocorrelation index based on the distance weight matrix, the Getis-Ord Gi* index, was developed by Ord and Getis and can identify high- and low-value clustering. This is a further analysis of the specific areas where villages gathered in the three historical periods, and the forms of clustering. The traditional villages with high-density clustering in a certain area and the traditional villages in the surrounding area also show high-density clustering are considered to have high–high clustering (hot spot), and vice versa as having low–low clustering (cold spot).We can obtain the main clustering locations of traditional villages in different historical periods based on the analysis of the results. Compared with the local Moran’s index, the Getis–Ord Gi* index can display the clustering area more accurately, thus we chose to use the local Getis–Ord Gi* index for the calculation. The formula is as follows [66,67],

$$\begin{array}{c}{G}_i^{*}=\frac{{{\displaystyle \sum }}_j^n{W}_{ij}{x}_j}{{{\displaystyle \sum }}_j^n{x}_j}\end{array}$$

(2)

3.2.2. Standard Deviation Ellipse

The standard deviation ellipse (SDE) can precisely explain the centrality, directionality, spatial morphology, and other characteristics of the spatial distribution of geographic elements from the perspective of the global space. It reveals the spatial distribution and the process of the spatiotemporal evolution of traditional villages in the Yuan, Ming, and Qing dynasties. The parameters of the standard deviation ellipse include the center of gravity of the ellipse, the azimuthal angle, the semi-minor axis, and the semi-major axis of the standard deviation ellipse. The formula is as follows [68]:

$$\begin{array}{c}{\overline{X}}_w=\frac{{{\displaystyle \sum }}_{i=1}^n{w}_i{x}_i}{{{\displaystyle \sum }}_{i=1}^n{w}_i},{\overline{Y}}_w=\frac{{{\displaystyle \sum }}_{i=1}^n{w}_i{y}_i}{{{\displaystyle \sum }}_{i=1}^n{w}_i}\end{array}$$

(3)

3.2.3. Buffer Zone and Euclidean Distance

The buffer zone is the range of influence of the geospatial target. In this study, based on the geographic distribution of traditional villages in Hunan Province, we establish buffer zones of lines (rivers and ancient post roads). The number of villages in each period under the limited distance (30 km and 10 km) of roads and rivers can be obtained by establishing buffer zones. The distance of 30 km is based on the distance between ancient post stations. A distance of 10 km is the farthest distance considered walkable for the settlers. Its calculation formula is as follows:

$$\begin{array}{c}{D}_{\left(x,y\right)}=\sqrt{{{\displaystyle \sum }}_{i=1}^n{\left[w_{i\left(x_i-y_i\right)}\right]}^2}\end{array}$$

(4)

The Euclidean distance is the actual distance between two points in space. The Euclidean distance can be calculated more accurately to obtain the exact distance between two points than the calculation of the buffer zone, which can only obtain the number in the area of influence. We need this value for further analysis of geographic detectors. In this analysis, the Euclidean distance is the straight-line distance from the location of the traditional village to the ancient city. Its calculation formula is as follows [69]:

$$\begin{array}{c}\mathrm{d}=\mathrm{s}\mathrm{q}\mathrm{r}\mathrm{t} \left[{\left(x_1-x_2\right)}^2+{\left(y_1-y_2\right)}^2\right]\end{array}$$

(5)

It should be noted that due to the lack of comprehensive information on ancient transportation, we can only rely on the Euclidean distance to perform an analysis of the real distance between two locations.

3.2.4. Geodetector

The spatial heterogeneity is measured, the explanatory factors are found, and the interactions between the variables are examined using the value-q of Geodetector [49]. Using Geodetector, it is possible to analyze the factors influencing the emergence of differences in the distribution of traditional villages during the Yuan, Ming, and Qing dynasties and to examine the extent to which the influencing factors have the ability to explain the differences. In this paper, influencing factors, including slope aspect, river systems, ancient cities, DEM, and ancient post roads were selected for quantification. The closer the value is to 1, the stronger the explanatory value of the influential factors in explaining the differences in the spatial distribution of the occurrence of the traditional villages.

$$\begin{array}{c}q=1-\frac{{{\displaystyle \sum }}_{h=1}^L{N}_h{\sigma }_h^2}{N{\sigma }^2}\end{array}$$

(6)

4. Results

4.1. The Changing Direction and Clustering of Villages in the Yuan, Ming and Qing Dynasties

4.1.1. Changes in the Clustering Area

To depict the dispersion of Hunan’s traditional villages, this study geographically analyzed the traditional villages in Hunan Province using Moran’s global index. Moran’s global index was found to be 0.513 (p = 0.000, Z = 9.679), suggesting that the geographical distribution of traditional villages in Hunan Province had a significantly clustered distribution pattern.

Based on the results above, we selected a total of 252 traditional villages in Hunan across three historical periods (Yuan Dynasty, Ming Dynasty, and Qing Dynasty), analyzing the cold spots and hot spots for the clustering of villages in each of the three historical periods, i.e., whether the distribution of villages in each county of Hunan Province in different historical periods showed a high–high density agglomeration (hot spots) and a low–low density (cold spots) distribution agglomeration. This helps us to find the main areas of aggregation. It was found that traditional villages in Hunan Province in the three historical periods of the Yuan, Ming, and Qing Dynasties showed hot spot clustering and cold spot clustering areas. First, traditional villages from the Yuan Dynasty showed three hot spot clustering areas and one cold spot clustering area. The hot spots were mainly in Yongzhou and Chenzhou in the south of Huna; Chengbu Miao Autonomous County, Suining County, and Hongjiang County in Shaoyang; and Yongshun County, Baoding County, and Longshan County in the north. Second, the traditional villages of the Ming Dynasty showed two hot spots and one cold spot. The hot spots were mainly concentrated in western Xiangxi Tujia Autonomous Prefecture and Huaihua city in the southwest; the cold spots were in Changsha, Xiangtan, and other surrounding areas. Third, the traditional villages of the Qing Dynasty are mainly concentrated in Xiangxi Tujia Autonomous Prefecture, while the cold spots are still in Changsha, Xiangtan, and the surrounding areas (Figure 3).

It is obvious that the hotspot clustering areas of the traditional villages from the Yuan Dynasty to the Qing Dynasty gradually decreased from the original three areas of clustering (southern Hunan, southwestern Hunan, and western Hunan) to one area in western Hunan, while the hotspots’ aspects gradually shifted from the south to the west of Hunan. In addition, no significant changes occurred in the cold spot clustering areas. This indicates that the area of high-density village clustering shifted over time, starting from the southern region to the west as the early settlers shifted away from primarily establishing villages in the southern, southwestern, and western regions (Yuan Dynasty) to mainly establishing villages in the western region (Qing Dynasty), revealing that the village population was constantly moving westward.

4.1.2. Change in the Direction of Village Distribution

Regarding the aspect in which Hunan’s traditional villages evolved, we analyzed the developmental tendencies of traditional villages in Hunan across different eras at the time of their establishment using standard deviation ellipses. The results of the study showed that the point coordinates of the center of gravity of the standard deviation ellipse of traditional villages from the Yuan Dynasty to the Qing Dynasty moved from (26.96° N, 111.23° E) to (27.43° N, 110.97° E), spanning about 0.3° from east to west and 0.5° from north to south, showing a trend of the center shifting from the southeasterly to the northwesterly aspect (Figure 4). The semi-minor axis of the standard deviation ellipse can be used to express the range of the data’s distribution. The longer the semi-minor axis, the more pronounced the centripetal force; conversely, the greater the dispersion, the longer the semi-major axis. It can be seen in Figure 4 that the semi-minor axis length of the standard deviation ellipse shortened between the Yuan and Qing empires, indicating that the centripetal force of the distribution of traditional villages increased, with the changes in the spatial distribution through the different eras showing a concentration trend in western and central Hunan.

In terms of the azimuth, the rotation value was 133.32° for the Yuan Dynasty, 130.11° for the Ming Dynasty, and 127.19° for the Qing Dynasty, with an average decrease of about 3° in the azimuth value per era (

Table 1. Data of the standard deviation ellipse and the center of gravity’s point coordinates.

	Shape Area	CenterX	CenterY	XStdDist	YStdDist	Rotation
Yuan Dynasty	9.14	111.23	26.96	2.41	1.21	133.32
Ming Dynasty	9.27	110.71	27.23	2.20	1.34	130.11
Qing Dynasty	7.49	110.97	27.43	2.37	1.01	127.19

). From the Yuan Dynasty to the Qing Dynasty, the geographic variation in the ancient villages in Hunan Province shifted from south to west.

4.2. Analysis of Natural and Social Factors in Traditional Villages of the Yuan, Ming and Qing Dynasties

4.2.1. Analysis of Elevation, Slope Aspect, and River System

Elevation is a geographical factor that influences the spatial patterns of towns and villages since it restricts the size of growth and settlement dynamics. In order to depict the geographical distribution of traditional villages in Hunan throughout the Yuan, Ming, and Qing Dynasties across diverse terrains, this study categorized the terrain according to the distinct height ranges as plains (200 m), hills (200–500 m), and mountains (>500 m). Overall, 388 traditional villages (representing 58.97% of the total number) are located in hilly areas in Hunan Province, according to the data; 153 traditional villages are located in mountainous areas (representing 23.25%), and 117 traditional villages are located in the plains area (17.78%). On this basis, a statistical study of the topography of the traditional villages with precise establishment dates (a total of 252 communities) was undertaken (

Table 2. Elevation percentages for different historical periods.

Category	DEM (Meters)	Yuan Dynasty	Ming Dynasty	Qing Dynasty
Plains	<200	17.07%	17.74%	28.26%
Hills	200–500	62.20%	67.74%	43.48%
Mountains	>500	20.73%	14.52%	28.26%

).

In the table above, it can be seen that in all three historical stages (the Yuan to Qing eras), traditional villages in Hunan mainly developed in hilly areas. Ancient villages of the Yuan and Ming eras were predominantly constructed in hilly regions, accounting for 62.20% and 67.74%, respectively. However, the number of villages located in hilly areas began to decrease after the Ming Dynasty, and the proportion of traditional villages established in hilly areas shrank to 43.48% in the Qing Dynasty. The villages established in plain and mountainous areas showed a rising trend, from 17.07% and 20.73% to 28.26%, respectively; in particular, the villages in mountainous areas showed an upward zigzag trend in the Ming Dynasty, although they decreased, they showed an upward zigzag trend overall. Therefore, the sites where the villages were built in the Hunan region during the Yuan, Ming, and Qing periods were mostly concentrated in the hills; development in the plains started in the Ming Dynasty; and the development of the hilly areas by settlers slowed down in the Qing Dynasty as their focus turned to the plains and mountains.

The amount of solar radiation a village receives is influenced by the aspect of the slope on which it is located. Generally speaking, a sunny slope receives more solar radiation than other slopes and has better temperature conditions, so the aspect of the slope affects the establishment of a village from the residential perspective [70]. In ArcGIS, the slope aspect is represented by the angle of clockwise rotation, with the starting point facing north (0 or 360°), and is divided into eight aspects (north, northeast, southeast, south, southwest, west, and northwest) with 45° intervals. The results of extracting the slope aspect were reclassified and overlaid with traditional villages according to these eight aspects, and the number of each slope aspect of traditional villages during the Yuan, Ming, and Qing Dynasties was counted (Figure 5).

According to the statistics, the establishment of traditional villages in the Yuan Dynasty did not have too much bias in the choice of slope aspect, and each aspect was relatively balanced, except for the fact that more traditional villages were established in areas with a northwesterly slope aspect, accounting for 17%. Throughout the Ming and Qing Dynasties, the traditional villages in Hunan showed a specific preference regarding the slope aspect. The traditional villages in the Ming Dynasty had predominantly southerly and southeasterly slope aspects, accounting for 20.9% and 15.1%, respectively, while the other slope aspects were relatively balanced. During the Qing Dynasty, the main preference was to establish villages with a southward slope aspect; the proportion reached 30.7%, followed by the southwesterly aspect, accounting for 15.3%; villages with an eastward slope aspect were relatively few, accounting for only 0.4%. The slope aspect, in a broad sense, can be divided into two types of slope: sunny slopes and shady slopes. This analysis showed that traditional villages in Hunan after the Ming and Qing Dynasties were established on sunny slopes, which differed greatly from the slope distribution of villages in Shanxi province. It is important to note here that slope-aspect preference can, to some extent, reflect the influence of the Feng Shui ideology that influenced the early Chinese settlers when they built their villages in the Ming and Qing dynasties. However, since the slope aspect is only one component of Feng Shui, we objectively elaborated on the results of the direct influence of the slope aspect on the establishment of villages.

Regarding river systems, a buffer of 10 km was used to analyze the river systems in Hunan, and we found that 73.23% of the villages were distributed within the buffer zones of the river system. Villages across different eras were equally distributed around the river systems. In particular, 84.15% of the villages in the Yuan Dynasty and before the Yuan Dynasty were distributed within 10 km of the river system, and the average proportion of villages within 10 km in all other dynasties exceeded 50% (Figure 6).

4.2.2. Analysis of Transport and Ancient Cities’ Distances

The transportation network is not only an important means of transporting resources and people but is also a unique medium for the spread of human civilization and culture. Historically, people established village sites within the constraints of the transportation network. The ancient transportation network was mainly divided into two types: riverways and land-based roads. In order to fully reflect the influence of ancient transportation on the villages’ location, this study selected ancient post roads and river systems in Hunan Province for the study and analyzed both land and river buffer zones according to the ancient post road recorded in the CHGIS.

According to the statistics, 40.88% of traditional villages in Hunan Province are located within the 30 km buffer zones of the ancient post roads, while the remaining 59.12% of villages are distributed beyond 30 km (Figure 6). In line with these findings, this study further analyzed the traditional villages and the 30 km buffer zones of the ancient post roads across different periods. According to the results, 30.49% and 23.39% of the villages in the Yuan and Ming Dynasty, respectively, were within the 30 km buffer zone of the ancient post road, while the rest were outside the 30 km buffer zone of the ancient post road. The distribution of traditional villages in the Qing Dynasty is closely related to the ancient post roads, with 47.83% of the villages established within 30 km of the ancient post roads.

Ancient cities were the main political and economic centers of ancient China. Compared with traditional villages, ancient prefectural seats were the main cities in China’s historical periods. In order to analyze the distances between traditional villages in Hunan and ancient cities in the Yuan, Ming, and Qing empires, this study determined the coordinates of ancient cities based on the prefectural seats established in Hunan during the period of the Yuan to Qing given by Tan Qixiang’s Historical Atlas of China and used the Euclidean distance in ArcGIS to determine the distance from these prefectural seats to the traditional villages (Figure 7).

It was found that the average distance of the traditional villages from seven ancient cities in the Yuan Dynasty period was 11.253 km, the average distance of traditional villages from eight ancient cities in the Ming Dynasty was 8.655 km, and the average distance in the Qing Dynasty was 7.111 km (

Table 3. Distance between traditional villages and ancient cities.

Dynasty	Euclidean Distance (km)	Number of Ancient Cities	Major Ancient Cities
			Northeast Regions	Central Region	Southern Region	Western Region
Yuan dynasty	11.253	7	Changde, Yuezhou, Tianlin	Baoqing	Hengzhou, Yongzhou	Yuanzhou
Ming dynasty	8.655	8	Changde, Yuezhou, Changsha	Baoqing	Hengzhou, Yongzhou	Chenzhou, Yongshun
Qing dynasty	7.111	10	Changde, Yuezhou, Changsha	Baoqing	Hengzhou, Yongzhou	Shouzhou, Yuanzhou, Chenzhou, Yongshun

). Among these, the average distance from traditional villages in the Qing Dynasty was the closest. This shows that the average distance between traditional villages and ancient cities in Hunan from the Yuan Dynasty to the Qing Dynasty decreased. From Figure 2, it can be seen that there were relatively more villages in the Ming Dynasty, reaching 124, while there were only 46 traditional villages in the Qing Dynasty. Additionally, combined with Table 3, it can be seen that the Euclidean distance obtained is the average distance, and the average distance between the villages and ancient cities was the closest in the Qing Dynasty. This can exclude the effect of the increase in traditional villages on the distance. As a result, the average distance between the traditional villages and ancient cities in Hunan decreased from the Yuan Dynasty to the Qing Dynasty, and the administrative and governmental control and influence of the ancient cities on the traditional villages also increased.

4.3. Impact Analysis of Natural and Social Factors

In order to further understand the main driving factors of the geographical dispersion of traditional settlements in Hunan during the Yuan, Ming, and Qing Dynasties, this study used the Geodetector to determine the differences in these driving factors by selecting the elevation (X1), slope aspect (X2), river systems (X3), ancient post roads (X4), and ancient cities (X5) within both the natural and social environments [71].

The results (

Table 4. Values of the drivers of the spatial distribution of traditional villages.

Category	Yuan Dynasty		Ming Dynasty		Qing Dynasty
	q-Value	p-Value	q-Value	p-Value	q-Value	p-Value
Ancient cities (X1)	0.315	0.00	0.033	0.41	0.018	0.66
DEM (X2)	0.301	0.00	0.112	0.01	0.205	0.10
Slope aspect (X3)	0.629	0.30	0.063	0.05	0.211	0.06
River (X4)	0.125	0.05	0.058	0.03	0.081	0.54
Ancient post road (X5)	0.692	0.00	0.291	0.00	0.389	0.00

) showed that there were differences in the influential factors driving the spatial distribution of traditional villages in Hunan in the three historical stages (the Yuan, Ming, and Qing Dynasties). During the Yuan Dynasty, ancient post roads, ancient cities, elevation, and the river system significantly influenced the spatial distribution of villages, and the q-values of these factors were ranked by the power of their influence: ancient post roads (0.692) > ancient cities (0.315) > elevation (0.300) > river system (0.125); the influence of slope aspect on the distribution of traditional villages was not significant. During the Ming dynasty, elevation and ancient post roads became the factors that significantly drove the spatial distribution of traditional villages, where the q-values were ranked as follows: ancient post roads (0.148) > elevation (0.11), while the other factors were not significant for the distribution of traditional villages. During the Qing Dynasty, ancient post roads (p = 0.389) became the most important driving factor for the spatial distribution of traditional villages.

5. Discussion

Compared to previous studies of traditional villages, this paper analyses the establishment points of traditional villages in Hunan and further advances the study of the rule of traditional village development from the Yuan to the Qing dynasties.

Firstly, the paper quantifies village sites to show the migration of Hunan people and their aggregation from the south and east to the west of Hunan during the Yuan, Ming, and Qing dynasties. This result is mirrored by studies focusing on the historical development of Chinese society [72,73]. The historicists used literature and historical materials to analyze the population migration in Hunan during the Ming or Qing dynasties, demonstrating that the population moved from the east to the west. Our study, using village sites to compare the dynamics of population migration from the Yuan Dynasty to the Qing Dynasty, specifies the main points of village clustering and the angle of offset and uses data and visual comparisons to more clearly point out the impact of socio-historical events (Jiangxi filling Huguang and Huguang filling Sichuan) on traditional villages during the Yuan, Ming, and Qing Dynasties.

Secondly, our findings point to the sequence of land exploitation. During the Yuan and Ming dynasties, traditional villages were mostly established in hilly areas, with the number of villages in the plains increasing. From the Ming Dynasty onwards, however, the development of the hills began to slow down, and by the Qing Dynasty, it began to decline. During the Qing Dynasty, the number of villages established in the mountainous areas was on the rise. This shows a shift in land exploitation from lower to higher elevations and from the more easily developed plains and hilly areas to the more difficult mountainous areas. The reason is that the wars and ruler-induced migrations during the Ming and Qing dynasties brought more people to Hunan, requiring more land to ensure survival. The first areas that immigrants arrived in were in the east, and the eastern areas were the plains and hills, so the first areas to be exploited were the hills and plains. In a study undertaken on the wetland area of Dongting Lake in the Hunan region, some scholars point out that earthworks and embankments were constructed in the Dongting Lake plain area in the eastern part of Hunan during the Ming Dynasty, forming a plain and hilly area with superior cultivation conditions, so the settlers began to exploit it extensively [74,75]. However, from the early Ming to the end of the Qing dynasty, the immigration policy of “Huguang filled Sichuan” forced people to move from the plains to the less developed mountainous regions [26]. Thus, with the exploitation of the plains and hills, there was a significant increase in the exploitation of the mountains by the inhabitants. In addition, after the establishment of the Ming and Qing dynasties, a series of policies were promulgated to encourage land exploration, which also stimulated the exploitation of mountainous areas to a certain extent. During the same period, the surrounding southern regions, such as Jiangxi and Sichuan, accelerated the trend of exploitation of mountainous areas. Therefore, the results of the land exploration undertaken in this study are consistent with the labor productivity of the settlers during the Yuan, Ming, and Qing dynasties [76].

Finally, our results also reveal that the influence of the slope aspect, river, and elevation on traditional villages weakens with time, with the most significant influence being that of ancient post roads. This, combined with the reduction in the distance between traditional villages and major cities from the Yuan to the Qing dynasties, reveals that communication between rural settlements and urban settlements increased and that people moved by road. During the Yuan and early Ming dynasties, when China was at war, the settlers moved away from the cities to avoid war. During the Ming and Qing dynasties, however, China entered a highly developed centralized imperial period, where society was relatively stable, and the cities became the center of the economy, so people gradually established villages near the cities, promoting and driving progress in the villages.

6. Conclusions

Traditional villages are nodes in spatial history, carrying and condensing the spatial information of different historical stages. Similar to cities, the spatial process of villages can also reveal the temporal sequence of regional development. Through a specific analysis of 252 traditional villages in Hunan Province in terms of their founding time and geographic location, their spatial and temporal characteristics, and typological evolution, the following conclusions were obtained:

(1) Traditional villages established before the Yuan Dynasty were mainly clustered in the east, south, and west of Hunan Province, and traditional villages in the Ming and Qing Dynasties were mainly distributed in the west of Hunan Province, which indicates that the direction of regional development in the Ming and Qing Dynasties in Hunan Province shifted from the east and south to the west.

(2) The average elevation of the villages in Hunan Province increased, with development mainly seen in the hilly areas during the Ming Dynasty, while development in both the mountains and the plains began in the Ming and Qing Dynasties. This reveals the change in the establishment of traditional villages in Hunan during the Yuan, Ming, and Qing dynasties from easily exploited flatlands to more difficult-to-exploit mountainous areas.

In terms of the driving factors of traditional villages in Hunan, although these driving factors were different in each period, ancient post roads determined the spatial distribution of traditional villages to a much greater extent than other influential factors and became the main driving factor. The factors in each historical period ranked from strong to weak, were ancient post roads (0.692) > ancient cities (0.315) > elevation (0.300) > river system (0.125) in the Yuan Dynasty; in the Ming Dynasty, the factors were ancient post roads (0.148) > elevation (0.110); and in the Qing Dynasty, ancient post roads (0.389) were the main driving factor.

It should be noted that the conclusion we reached is based on an analysis of only 252 traditional villages in the Hunan area due to data limitations; at the same time, since relatively few traditional village sites remain in Hunan before the Yuan Dynasty, further in-depth explorations and analyses are needed in order to form a more comprehensive understanding of the development of Hunan’s traditional villages across other historical stages.