殷卜辭出土之後，大乙、唐很快地被考釋出即〈殷本紀〉的天乙與湯，卜辭做人名與地名的成字與金文合，但沒有人注意人名的成即成湯之成。由於卜辭有咸戊，即文獻中的巫咸，又有人名的「咸」，因此早期學者誤以為「咸」為咸戊的省稱。自陳夢家將卜辭從戌從丁的「□」與從戌從口的「□」一分為二，一指咸，一指成，甲骨學者從之者眾，因此有人主張「□」與「□」俱為成湯之成，甚至有人力主成湯之成為「□」非「□」。本文從字形分析，剖析卜辭之「□」為成湯之成，「□」與「□」俱為咸字，亦指成湯，而前賢謂《尚書．酒誥》及《禮記．緇衣》的「湯咸」，即卜辭中指成湯之「咸」，是可信的。

After the Shang oracle bones were unearthed, it was not long before the characters “Dayi” 大乙 and “Tang” 唐 were taken to be the characters “Tianyi” 天乙 and “Tang” 湯 from the “Yinbenji” 殷本紀. The character “Cheng” 成, used in people and place names within oracle bone inscriptions, is exactly the same as the graph found on bronze inscriptions. Yet no one has noticed that the graph “Cheng” used in people’s names is in fact the character “Cheng” as in “Cheng Tang” 成湯. As the name “Xianwu” 咸戊 (巫咸 in later texts) appears on the oracle bones, as does the name “Xian” 咸, early scholars mistakenly assumed that “Xian” was an abbreviation for the regional toponym “Xianwu.” Chen Mengjia treated and as two separate characters that had evolved from a single more primitive version. He thought the former had developed from the characters 戌 and 丁, and the latter from the characters 戌 and 口. thus refers to “Xian,” while refers to “Cheng.” Many scholars of the oracle bone inscriptions followed in this line of thinking, and thus some argued that and were both representations of the character “Cheng” as in “Cheng Tang.” Others even insisted that , and not , should be read as the character “Cheng” as in “Cheng Tang.” The present paper, in contrast to previous studies, relies on an analysis of the forms of the characters themselves. This article proposes that the oracle bone graph should be read “Cheng” as in “Cheng Tang,” and that and should both be interpreted as “Xian” and that both refer to “Cheng Tang.” Furthermore it is plausible that the oracle bone graph “Xian” that refers to “Cheng Tang” is none other than “Tangxian” 湯咸 from “Jiugao” 酒誥 of the Shangshu 尚書 and from “Ziyi” 緇衣 of the Liji 禮記, as previous scholars have suggested.