Empirical Study on Grow th of Evil Forces in Land Requisition and Relocation in City G of Hubei Province Based on Social Network Analysis

Hongxia ZHANG，Xia ZHOU，Yanan LI

1．School of Management，Hubei University of Education，Wuhan 430205，China；2．Jingshan County Bureau of Land and Natural Resources of Hubei Province，Jingmen 431800，China；3．School of Public Administration，Huazhong University of Science and Technology 430074，China

1 Introduction

It has become an indisputable fact that evil forces always intervene in land requisition and relocation field．In this study，evil forces mainly refer to underworld organizations forcing farmers and relocation households to remove their land by brute and violent forces，such as beating，smashing and looting．For example，Han Enping underworld organization consists of 29 members and does evil things in Hohhot City of Inner Mongolia，and they accumulate money mainly through helping developers to relocate by violent forces［1］．The head of sinister gang，Fu Qiang，had been involved in land requisition and relocation works for a long time．In the end of2008，he called together more than 100 hooligans to pull down houses of relocation households in a resettlement project，and houses of four households became ruins just in a morning［2］．Such method of＂accumulating wealth by brutal forces＂and＂protecting developers by violent means＂is strongly condemned by the society．Tan Shukui，professor of Huazhong University of Science and Technology，stated that land premium is still a big pillar of financial revenue of some areas，and urban construction progress is the political performance pursued by some cadres；in consequence，it is inevitable that some cadres get blinded by the lust of gain，collude with developers and evil forces behind the scenes，leading to violent land requisition and relocation and harming benefits of the masses［2］．Land requisition and relocation go in accompany with China's urbanization．In the process of land requisition and relocation，it involves institutional adjustment and benefit allocation．Brutal and evil people obtain survival opportunity in areas where rural grass-roots political power is weak and social management is relatively insufficient，making them tainted with the color of evil forces of hooligan style requisition and relocation．

Schneideret alrevealed causes of growth of evil forces through reviewing marginalized global enemies［3］．Cheloukhine stated that the root cause of Russianmafia is shadow economy after privatization［4］．Scalia analyzed transmutation of Sicilian Mafia in Fordism and Post Fordism from economic and cultural perspective，and pointed out globalization brings crime opportunities and creates new illegal market［5］．Patrimonial found existence of regional hereditary politics provides possibility for development of underworld organizations［6］．Bovenkerk believed that some strategies of government have incentive influence on growth of evil forces［7］．He Xuefeng considered that evil forces and official institutions gradually become two foundations of village governance［8］．Ou Sanrenet alfound that evil forces develop in some rural areas in an abnormal way and become background pusher of many rural group accidents［9］．Liu Yufenget alstated that public power of the state is gradually removed from rural areas after reform and opening up，then rural evil forces obtained growth and development［10］．Geng Yu further stressed that grass-roots political power and evil forces have the tendency of collaboration［11］．In sum，growth and development of evil forces have certain researches，but there are no definite answers about how they growth in China，how they are focused，and what about their growth environment and characteristics．In view of these，we made an empirical study to provide reference for land consolidation．

2 Social network of evil forces

In order to reveal interested parties of evil forces and answer questions of whom evil forces will exert influence on，from whom they will seek protection，and in which way the counterpart will react，we adopted social network analysis method（SNA）．Social network analysis is the mapping and measuring of relationships and flows between people，groups，organizations，and other connected information/knowledge entities．The nodes in the network are the people and groups while the links show relationships or flows between the nodes．SNA provides both a visual and a mathematical analysis of human relationships．Social network analysis stresses mutual influence and dependence of actors，consequently leading to overall emerging behavior［12］．

2.1 Building of social networkIn December of 2011 to March of 2012，using Expert Consultation Method，we surveyed 266 experts in all over the country，including professors，famous scholars，and government officials．Questions mainly involve＂how evil forces intervene in land requisition and relocation＂，＂which parties have related interests with evil forces＂，and＂what about the type of their relationship＂．We received 100%copies of questionnaire．According to their answers of questions，we obtained social network table（Table 1）of evil forces．Inputting corresponding survey and statistical data with the aid of Net draw software，we established adjacent matrix of interested parties（Fig．1）．

Table 1 Interested parties of evil forces in land requisition and relocation

2.2 Analysis of related indicatorsNodes represent actors．They form social network through connection of information and contract，while individual actions are influenced by their position in the network．Use arcs to denote the relationship between interested parties，and assign weight to arcs using binary method．If there exits a relationship，the weight is 1，if there is no relationship，the weight is0．Use the centrality to measure the central position of interested parties．The centrality includes degree centralityCD（ni），closeness centralityCC（ni），and betweenness centralityCB（ni）.Calculation formula is as follows：

In Formula 1 through Formula 3，N signifies network size，mm（ni）denotes number of lines connecting with nodeni，d（ni，nj）refers to short line distance between nodesniandnj，gjksignifies number of short lines between nodesniandnj，g（jk ni）stands for number of short lines between two actors including actorni.Degree centrality refers to the number of ties a node has to other nodes．Actors who have more ties may have multiple alternative ways and resources to reach goals-and thus be relatively advantaged．Degree centrality is used to measure if there are intensive transactions in the position，in which specific interested parties exist．Closeness is a measure of the degree to which an individual is near all other individuals in a network．It is the inverse of the sum of the shortest distances between each node and every other node in the network．Closeness centrality is used to represent ability of interested parties of sharing resources and information in the network．Betweenness is a measure of the extent to which a node is connected to other nodes that are not connected to each other．It is a measure of the degree to which a node serves as a bridge．Between centrality is used to measure the ability of interested parties of controlling resources and information in the network．The enhancement density and central potential network density represent closeness degree between actors．Generally speaking，the higher the network density，the more significant restraint the actor suffers from network structure，and the weaker independence ability of individuals．The network density is calculated as per Density＝L/［N（N－1）］，whereLis number of arcs．Central potential depicts difference degree of actors in the whole network．The higher the central potential value indicates that network has tendency of centralization and the distribution of powers will be more uneven．The central potential of network is the ratio of sum of difference between centrality of the most core point（corresponding maximum centrality）and centrality of other points to the sum of largest possible difference．The formula is as follows：

whereCmaxis the maximum centrality andCiis centrality ofni.

Table 2 tructural indicators of network

The density of entire network is 0．3469（Table 2），i.e．34．69%possible relationship exists．Outward degree centrality（31．22%）is obviously smaller than inward degree centrality（44．315%），indicating that the distribution of interested parties using network resources is more unbalanced than interested parties outputting resources．The betweenness centrality indicates that there exist interested parties having ability to effectively control resources in the network（35．18%）．It is impossible to calculate closeness centrality because the built social network graph is not strong connected graph，the overall closeness between interested parties is relatively low and network structure has decentralized power center．Through calculation of degree centrality，betweenness centrality and closeness centrality，we found that public security organs have the maximum outward degree centrality（60．50%），while developers have the maximum inward degree centrality（73．00%），showing their influence in the network．Besides，developers and public security organs have higher betweenness centrality，40．40%and 25．18%separately，indicating that they are situated in core position of the network and are likely to control other parties．Inward closeness centrality of evil forces is higher（52．33%），showing that they are not controlled by other interested parties in the resource input，or in other words，other interested parties are influenced by evil forces in the resource input．Public security organs，developers and village collective have higher outward closeness centrality（65．86%），indicating that all these three parties highly depend on other parties in the resource input．

The above results show that evil forces intervene in land requisition and relocation field on the condition that there is close social relationship with developers and public security organs．They two parties promote growth of evil forces．On the one hand，developers cooperate with evil forces to jointly participate in forced demolition and relocation and project development，seek gray income．In other words，evil forces are employed by developers or evil forces are developers themselves．On the other hand，public security organs are weak in law enforcement．Even，some public officials become umbrella of evil forces．Asa result，public security organs provide survival space for evil forces．In addition，public security organs are major functional departments of local government．This further explains growth of evil forces is inseparable from inefficient management of local government．

3 Spatial relation network of evil forces

In order to reveal spatial relation network of growth of evil forces，we made an empirical survey of City G in Hubei Province in April to September of 2012．We found that City G is at the rapid development stage and has rapidly increase of land demands and the task of land requisition and relocation is heavier and heavier．Since 2000，many violent demolition and relocation accidents occurred in City G，as shown in Fig．2．Life and property safety of some relocation households get seriously threatened．Here，violent demolition and relocation of evil forces specially refer to violent demolition of homestead and houses of relocation households by evil forces consisted of local tyrants and hooligans by illegal means such as threatening，sneak attach，disturbances，siege，destruction，forcing，and internment．

Radilet alimbedded criminal gangs into competitive network of certain geographical position and derived possibility of their actions through setting of spatial restraint［13］．Titaet alanalyzed spatial distribution of incidents of violence through establishing spatial weight matrix［14］．According to this research idea，we made auto-correlation analysis，cluster analysis，and regression analysis on growth space of evil forces in combination with popula-tion census data and crime characteristic data of City G．

3.1 Moran index of crimes of evil forcesFirstly，calculate Moran index of crimes of evil forces as shown in Fig．2 as per the formula 5．

Iis between－1．0 and 1．0．If it is negative，it means negative correlation；if it is positive，it means positive correlation．xis number of crimes．Wijis binary matrix of adjacent spatial weight as per adjacent standard or distance standard，and the purpose is to define mutual adjacent relationship of spatial objects．Generally，Wijof adjacent standard is：

wherei＝1，2，…，n；j＝1，2，…m；m＝norm≠n.Rook adjacency refers to adjacent relationship defined by common boundary length．Queen adjacency refers to adjacent relationship defined by common boundary points．From Table 3，we know that in the first order adjacency，I＝0．0907 and the significance level is0．047，while in the second order adjacency，I＝ －0．0865，indicating weak negative auto-correlation．These show that there exists spatial relationship between evil forces in City G，but the relationship is relatively complicated and needs further study．

3.2 Spatial distribution network of evil forcesWith the aid of Net draw software，we built spatial distribution network of activities of evil forces in City G．According to data provided by public security organs of City G，since2000，there have been 29 gangs of evil forces in land requisition and relocation field in City G．For convenience of study，we took 120 population census areas of City G as geographical setting range．Among these areas，23 areas had no intervention of evil forces，while 95 areas had different degrees of intervention of evil forces（2 areas have no statistics because of no cooperation with local public security and family planning authorities）．Code 8 and Code18 gangs of evil forces have the most activity areas，respectively 15 and 14 population census areas．Code 18，20，21，and 29 gangs of evil forces have fewer activity areas（two areas），and Code 7，10，11，14，16，17，25 and 27 gangs of evil forces have fewest activity areas（one area）．According to this，we depicted new topological graph of spatial network（Fig．3）．

Table3 Moran index for spatial auto-correlation of crimes of evil forces

Then，using cluster analysis method，we classified spatial relationship for evil forces intervening land requisition and relocation in City G of Hubei Province．In the freedom degree df＝1，df＝3，and df＝7，significance level of three clusters is relatively high．Fig．4 is the visual geographical illustration of three clusters．We found that activities of evil forces in City G have complicated spatial correlation，and their geographical distribution is uneven and takes on irregular characteristic．Some gangs intervene in wider areas（up to 10 areas），while some gangs intervene in narrow areas（only two or three areas）．Generally speaking，asymmetry in spatial structure reflects strength of activity ability，organization size，and development direction of evil forces．Gangs with wider control areas have higher activity ability and large organization size．This reflects relative consolidation of brutal people governance，or they promote land requisition and relocation in high efficiency and they tie together with some lawless officers and developers．Collaboration of benefits leads to penetration of evil forces into land requisition and relocation；gangs with narrow control areas have weak activity ability and small organization size，so they have to intervene in small areas．This reflects gangs of evil forces have not got firm foothold，they remain in the fighting stage of simple violence．Although not cooperating with lawless officers，the value of being used is not well recognized．

3.3 Spatial dependence model of evil forcesFrom the above discussion，network matrix of evil forces in City G is not a binary matrix completely connected．Each unit at least has an adjacent relationship，indicating that several gangs share the common geographical blocks（population census areas），so there may be some spatial dependence．Therefore，we established spatial dependence model of evil forces．

wheretis a scalar denoting spatial auto-regression parameters，w denotes weight matrix of relationship between geographical positions，Xsignifies exogenous independent variable matrix，and β is vector of regression coefficient．ε is error term．Assuming it is normal distribution，there will be：

Since wy is endogenous，it is not appropriate to adopt OLS regression，but should adopt two stage least square or MLE to estimate parameters．We used potential crimes（i.e.crime prediction）to measure spatial lag．

The first model，we obtained crime prediction value of each block of evil forces，the formula is as follows：

Table 4 Descriptive statistics of variables

The second model，in formula8，add ρafter β0，in other words，taking prediction value of crime×Wgas variable to measure whether evil forces are influenced by adjacent units．The third model，in formula 8，add ρWaafter β0，to inspect spatial status of distribution of evil force gangs．The fourth model，in formula8，add ρWgand ρWnafter β0，to calculate overall crime prediction value of spatial lagand network lag of evil forces．Formula will not be listed one by one．

In the setting of variables，we referred to opinions of Olate［15］，we focused on searching micro-data of criminals（members of evil forces）and macro-data of place of crime（activity area of evil forces）．Micro-data were provided by public security organs of City G and macro-data were provided by statistical department of City G．There are 12 independent variables and the dependent variable is number of crimes of evil forces，as listed in Table 4．

Table 5 Regression results of spatial dependence model of evil forces

Regression results（Table 5）indicate that in all 4 models，independent variables＂extreme poverty situation of criminals＂，＂criminal record or not＂，and＂percentage of non local population＂are significant．These reflect that evil forces intervene in land requisition and relocation field mainly because land requisition and relocation can provide opportunities for them to earn money and improve their living situation．Besides，many members of evil forces have criminal records or most of them are not local people，indicating there exists some social discrimination and psychological unbalance which promotes to earn black money．From Model 3 and Model 4，we know that network lag of crimes of evil forces is 0．086（the significance level is 5%），and the overall value of adjacency lag and network lag is 0．084（the significance level is 5%）．These reflect that activities of evil forces in land requisition and relocation of City G have no complete regularity in time and space．Growth of evil forces has certain incubation period or inhibition period，which is mainly related to effort and intensity of cracking down the evil forces．

4 Conclusions

In line with survey of City G in Hubei Province，according to social network analysis，in certain sense，developers（lawless people）and public security organs（weak attackers）constitute interested parties of evil forces in land requisition and relocation field．They determine resource input and output of evil forces and they have established certain interdependent relationship．Besides，the inward closeness centrality of evil forces is high，manifesting that evil forces independently possess decentralized power of network and have unscrupulous behavior in land requisition and relocation to a certain extent and play the role of fist theory and one-off business．Therefore，growth of evil forces comes from premeditation and collaboration of lawless developers，lack of functions and weak attack of public security organs．According to spatial relation network analysis，activities of evil forces have complicated spatial correlation and their geographical distribution is uneven，taking on irregular characteristics．In the field of land requisition and relocation，some evil forces are expanding and spreading，while other forces are relatively weak．In addition，evil forces have adjacency lag and network lag in space，reflecting lawless actions and brutal means will not be supported and will be punished by social justice at last．Thus，growth of evil forces has periodic changes，when in power，they will take opportunity to expand，while losing power，they will hide or even disappear．We revealed growth logic of evil forces in land requisition and relocation field at empirical level．Although 226 experts are few and City G of Hubei Province is just an individual case，we must take serious treatment in the face of objective data and scientific demonstration．

［1］LI JF．How to cleave“the hand of evil”in the demolition［N］．China Youth Daily，2011．（in Chinese）

［2］LI JF．How to restraint vicious power involving in land requisition and demolishing［EB/OL］．http：//news．xinhuanet．com/legal/2011-07/05/c＿121626832．htm，2011-07-05．（in Chinese）

［3］Jane Schneider，Peter Schneider．Civil society versus organized crime：Local and global perspectives［J］．Critique of Anthropology，2001（21）：427－446．

［4］Serguei Cheloukhine．The roots of Russian organized crime：From old-fashioned professionals to the organized criminal groups of today［J］．Crime Law＆Social Change，2008（50）：353－374．

［5］Vincenzo Scalia．From the octopus to the spider's web：The transformations of the Sicilian mafia under Postfordism［J］．Trends in Organized Crime，2010，（13）：283－298．

［6］Randall Collins．Patrimonial alliances and failures of state penetration：A historical dynamic of crime，corruption，gangs，and mafias［J］．The Annals of the American Academy of Political and Social Science，2011（636）：16－31．

［7］Frank Bovenkerk．On leaving criminal organizations［J］．Crime Law＆Social Change，2011（55）：261－276．

［8］HE XF．Study on private life and rural governance［J］．Readings，2006（11）：130－133．（in Chinese）

［9］OU SR，ZHANGWJ．Gangland influence and its effects on rural group events［J］．Journal of Hunan Agricultural University（Social Science）2010，11（6）：34－42．（in Chinese）

［10］LIU YF，XIE GX．The survival logic of the grey forces in the rural and countermeasures［J］．Journal of Jiangxi Institute of Education，2011，32（4）：43－46．（in Chinese）

［11］GENG Y．Gangland and rural governance involution［J］．Journal of China Agricultural University（Social Sciences Edition），2011，28（2）：71－77．（in Chinese）

［12］CHEN L．The measurement of employees' knowledge stock and empirical study based on social network analysis［J］．Journal of Industrial Engineering and Engineering Management，2009（4）：49－53．（in Chinese）

［13］SM Radil，C Flint，G E Tita．Spatializing social networks：Using social network analysis to investigate geographies of gang rivalry，territoriality，and violence in Los Angeles［J］．Annals of the Association of American Geographers，2010，100（2）：307－326．

［14］GE Tita，SM Radil．Spatializing the social networks of gangs to explore patterns of violence［J］．Journal of Quantitative Criminology，2011（27）：521－545．

［15］ROlate，C Salas-Wright，MG Vaughn．Predictors of violence and delinquency among high risk youth and youth gang members in SanSalvador，El Salvado［J］．International SocialWork，2012，55（3）：383－401．

［16］HUANG T．The study of problems and solutions about the land expropriation system in China［J］．Journal of Anhui Agricultural Sciences，2015，43（27）：272－273．（in Chinese）．

［17］WAN HL，LIU DY．Problems of land use during urbanization of western China：a case study in Baoji City，Shaanxi Province［J］．Agricultural Science＆Technology，2015，16（9）：2011－2013，2019．