本文接受Alexander Wendt對結構現實主義的部分批評，並從一種修正後的現實主義立場對Wendt的理論進行批判與補充，達到理論對話的效果。本文與傳統結構現實主義的顯著不同，在於主張一個體系結構從形成、穩定到崩解或轉型的不同時期中，結構與單位的影響力會有此消彼長的演變：另一方面，本文也承認體系中的「文化」會對行為者產生影響，但是現實主義強調的「能力分布」不能予以忽略，在探究體系結構的要素時，應兼顧體系的文化與能力分布。本文從現實主義角度釐清在Wendt的理論中「文化」與結構、體系的關係，並指出權力平衡體系與洛克文化的高度重疊。由於在Wendt的理論中霍布斯文化與洛克文化互斥，且廣義的文化可以當作體系的同義詞，而洛克體系又與傳統的權力平衡體系高度重疊，則根據邏輯推論應存在一個與權力平衡體系低度重疊，甚至是彼此互斥的多極體系，筆者遵循Wendt的用語，稱之為「霍布斯體系」。本文一方面援用Wendt的文化變項做為結構的要素之一，證明霍布斯體系與權力平衡體系是兩個不同的體系，並指出Wendt的理論有助於現實主義釐清霍布斯體系與權力平衡體系之區別；另一方面，本文也站在現實主義的角度反思Wendt的文化觀點在解讀現實主義傳統的霸權體系與兩極體系時會產生矛盾與侷限。最終，本文思考將現實主義的體系觀點與Wendt文化觀點全面結合的可能，透過Wendt對霍布斯文化與洛克文化的區分，嘗試將現實主義傳統的多極體系劃分為權力平衡體系與霍布斯體系，並在修正Kaplan的理論後，建立霸權體系、兩極體系、權力平衡體系、霍布斯體系四種國際體系理想型的論述。

This article re-examines Wendt's constructivism from the perspective of revised realism. The author's perspective is different from traditional structural realists in that I argue the effects of structure to units or units to structure will change during the periods of system formation, stability, transformation and collapse. The author accepts that the ＂culture＂ of a system does affect actors, but at the same time also insists that the ＂distribution of capabilities＂ as a variable cannot be ignored. We should consider both ＂culture＂ and ＂distribution of capabilities＂ when we examine factors related to system structure. The author identifies the relations among structure, system, and culture in Wendt's constructivism from the perspective of realism, and finds that the Lockean system largely overlaps with the balance-of-power (BOP) system. Given that Hobbesian and Lockean culture are distinct from each other, and that the Lockean system is largely overlapping with the BOP system, if Wendt regards ＂culture＂ as ＂system＂ in a general way, than we can deduce that there could be a multi-polar system that is largely independent from the BOP system. Following the language of Wendt, the author names this system as the ＂Hobbesian System＂. By using Wendt's constructivism to deduce the distinctiveness of the ＂Hobbesian System＂ and the BOP system, the author analyzes the structural differences between these two systems, while also pointing out that Wendt's constructivism has its limits in interpreting the realist tradition's hierarchical system and bipolar system. Finally, author examines the possibility of integrating the system approach of realism and the cultural approach of Wendt. In distinguishing between Hobbesian culture and Lockean culture, the author also distinguishes the Hobbesian system from the BOP tradition of realism. Revising the theory of Morton Kaplan, author argues that there are four ideal types of international systems: the hierarchical system, the bipolar system, the BOP system, and the Hobbesian system.