This article investigates some functions of the determinative sum(e) in Old, Middle and Early Modern English. It traces, quantifies and models the diachronic development of sum(e) as a pre-head element from a usage-based, cognitive Construction Grammar perspective by postulating several semi-specified but also abstract constructional OE and ME NP-schemas and sketching the observable (changing) network (re)configurations. By analyzing texts from the Penn–Helsinki Parsed Corpus of Middle English (PPCME) and the Penn–Helsinki Parsed Corpus of Early Modern English (PPCEME), the article especially focuses on the demise of the so-called ‘individualizing’ usage with singular nouns and traces the incipient stages of sum(e) as an indefinite near-article with plural and mass nouns. R was used to calculate correlation coefficients and measures of statistical significance in univariate analyses, and for multivariate regression models to address questions involving more than one predictor variable. It is shown that the usage of sum(e) with singular nouns became marginalized because of constructional competition with the numeral ān. In Old English, the two forms were both occasionally used to mark indefiniteness before singular nouns, but ultimately ān became the default marker of indefiniteness ousting sum(e). We also show that that the usage of sum(e) as an indefiniteness marker for plural nouns increased drastically from the later ME period onwards, particularly in informal text genres. Moreover, from the earliest periods onwards, there is a strong preference for this function to occur with complex NPs with pre- and post-head modification, which seem to have acted as bridging contexts.

中文翻译：

---

构式竞争和网络重构：研究古英语、中古英语和早期现代英语中的 sum(e)

本文研究了行列式的一些功能总和（e）在古英语、中古英语和早期现代英语中。它追踪、量化和模拟总和（e）通过假设几个半指定但也抽象的构造 OE 和 ME NP 模式并勾勒出可观察（变化）的网络（重新）配置，作为基于使用的认知构造语法角度的前头元素。通过分析文本Penn-Helsinki Parsed 中古英语语料库(PPCME) 和Penn-Helsinki 解析早期现代英语语料库（PPCEME），文章特别关注单数名词的所谓“个性化”用法的消亡，并追溯了单数名词的初期阶段。总和（e）作为带有复数和质量名词的不定近冠词。R 用于计算单变量分析中的相关系数和统计显着性度量，并用于多变量回归模型来解决涉及多个预测变量的问题。可以看出，使用总和（e）单数名词由于与数词的构式竞争而被边缘化一个. 在古英语中，这两种形式偶尔都用于在单数名词前表示不定，但最终一个成为无限期驱逐的默认标志总和（e）. 我们还表明，使用总和（e）从 ME 后期开始，作为复数名词的不定性标记急剧增加，特别是在非正式文本体裁中。此外，从最早的时期开始，这种功能强烈倾向于发生在具有头部前后修饰的复杂 NP 上，这似乎起到了桥接环境的作用。