Non-canonical coordination occurs when non-constituents are conjoined. Two major types of non-canonical coordination are gapping and non-constituent coordination (NCC). Non-canonical coordination has received interest for its relevance to core issues in syntax, such as constituency and phrase structure. The aim of this article is twofold. First, it provides empirical evidence for non-canonical coordination from Jordanian Arabic (JA) via an experimental investigation. Two acceptability judgment tasks (rating on a Likert scale) were designed to investigate the status of gapping and NCC in JA. A generalized linear model (GLM) was fit to the data, using the (glm) function from the mlogit package in R. There was a main effect of gapping (χ 2 = 13.203, p < 0.001) and NCC (χ 2 = 15.371, p < 0.001) on the acceptability of the constructions in the positive direction. Second, the article provides an analysis of the facts that is couched in terms of left-to-right syntax via a hybrid analysis that assumes that non-canonical coordination can have two sources: a complementizer phrase source and a vP source.

中文翻译：

---

约旦阿拉伯语中非规范协调的语法：实验研究

当非成分被连接时，就会发生非规范协调。非规范协调的两种主要类型是间隙协调和非成分协调（NCC）。非规范协调因其与语法核心问题（例如选区和短语结构）的相关性而受到关注。本文的目的是双重的。首先，它通过实验调查为约旦阿拉伯语（JA）的非规范协调提供了经验证据。设计了两项可接受性判断任务（李克特量表评级）来调查 JA 中间隙和 NCC 的状况。广义线性模型 (GLM) 对数据进行拟合，使用 (格鲁姆）函数来自姆洛吉特R 中的包。间隙的主要影响（χ 2= 13.203，p< 0.001) 和 NCC (χ 2= 15.371，p< 0.001) 正方向结构的可接受性。其次，本文通过混合分析对从左到右语法表达的事实进行了分析，该分析假设非规范协调可以有两个源：补足词短语源和 vP 源。