首先，本文問題意識的提出與過去不同，不再單純找尋在臺客家與閩南的泉州、漳州建築之差異為目的，反而是思考為何兩者之間如此相似的原因。之所以有這樣的發問，是因為前往所謂的客家「原鄉」進行田野調查時，這個「原鄉」也只不過是渡台先民們，在中國長期遷徙過程中的一個「轉繼站」，而不是客家族群的起源地。這雖是簡單的事實，但卻是非到訪原鄉地無法輕易理解的事情。因此本文不再以找出臺灣客家建築的「靜態特徵」為目標，而是著重於原有漢族文化的合院建築共通性的討論。如今無論是存在於臺灣客家原鄉的粵東、閩西或是贛南的建築形態，甚至台灣的客家建築所表現的特徵，都是經過多次移民遷徙、定居，因應各時代的社會情境，融合各地風土民情一時性的「動態發展」之結果。因為田野調查改變了研究問題的發問，同樣是田野調查帶來的觀點，亦即在20年前即已在台灣發現，但又苦於提出合理解釋的客家建築「雙棟」作法。因為田野調查，知道這種作法在中國是不限定客家建築才有的特徵。更且有不少案例不僅有「雙棟」，甚至「桁檁」與「椽木」（桷木）也都採上下獨立的雙層作法，亦即完全的「雙重屋頂」結構。因此，幾乎可以確定認為臺灣「雙棟」是「建築結構部材」經過歷史發展變遷後的結果。而這種「雙棟」或是「雙層屋頂」作法，甚至可以上推至中國上古時期即已存在。既然客家建築或是偏遠地區的建築，仍保有古代的作法，同理可推，當今存在於中國中心文化偏遠地區的臺灣，其傳統客家建築是否仍保有古代中國建築特徵之片段痕跡？這是今後客家建築研究，或是整體的閩粵建築研究可以期待的研究方向。

Based on a field survey of Hakkanese architecture in eastern Guangdong and western Fujian in 2008 and other studies, it is clear that Hakkanese architecture exhibits different styles in western Fujian, eastern Guangdong, and southern Jiangxi. There is no one style that is representative of Hakkanese architecture in China. However, an analysis of the architecture from these three provinces shows that they share a common type of building, namely the heyuan jianzhu (合院建築) or courtyard compound. Yet, because of climate influence and social conditions, as well as localities and development trends, this courtyard compound evolved into many different types of buildings. Thus far, researchers have believed that certain characteristics were unchanged in their migration from China to Taiwan, but this is not always true. Consequently, Hakkanese architecture is not static but always changing depending on environmental conditions. This study shows that double ridge poles were put on the top of gable walls under the roof, which is a simplified version of the two-layer roof structure. Double roofs have existed from ancient times in China.