This article aims to provide a concise overview of the most relevant topics concerning the implementation of acceptability judgments in generative research. The first part focuses on theoretical issues, including the reasons underlying the prevalence of acceptability judgments, the skepticism towards the wide use of informal judgments, the arguments for and against the continued use of informal data points and the challenges related to experimental data collection and the gradient nature of judgments. The second part further explores the concept of gradient acceptability and its different sources. We suggest that violations of soft constraints correspond with partial acceptability, unlike violations of hard constraints, which cause strong unacceptability. Based on our investigation of selected partially acceptable syntactic phenomena in Polish, we show that their syntactic accounts can also benefit from the inclusion of non-syntactic factors, which can be more reliably identified via experimental methodology. More specifically, we propose that (i) the lack of subject orientation of pronominal possessives could be attributed to lexical ambiguity, (ii) coreference of possessive cataphora is facilitated by antecedent backgrounding and (iii) WCO is improved by focus marking of the possessive pronoun, (iv) while processing of the last two structures could be ameliorated with more informative fillers.

中文翻译：

---

波兰生成语言学可接受性判断的全景

本文旨在简要概述与生成研究中可接受性判断的实施最相关的主题。第一部分侧重于理论问题，包括可接受性判断盛行的原因、对广泛使用非正式判断的怀疑、支持和反对继续使用非正式数据点的论点以及与实验数据收集和判断的梯度性质。第二部分进一步探讨了梯度可接受性的概念及其不同来源。我们认为，违反软约束与部分可接受性相对应，而违反硬约束则导致强烈的不可接受性。根据我们对波兰语中部分可接受的句法现象的调查，我们表明他们的句法解释也可以受益于包含非句法因素，这些因素可以通过实验方法更可靠地识别。更具体地说，我们建议（i）代词所有格缺乏主语定向可能归因于词汇歧义，（ii）先行语背景促进了所有格隐喻的共指，（iii）WCO通过所有格代词的焦点标记得到改善，（iv）而最后两个结构的处理可以通过更多信息填充来改进。