The reduction of the preposition para (to, for) in Brazilian Portuguese according to the previous word

ABSTRACT This article analyzes the reduction of the preposition para (to, for) to pra in Brazilian Portuguese. Sequences of (previous word + para or pra) selected from a corpus of spoken language were statistically analyzed to understand which factors could be playing a role in this reduction. A statistical analysis of the 30 most frequent sequences of (previous word + para or pra) was carried out to identify the factors that seem to influence this reduction. The concepts of string frequency (Krug 1998, 2003, Bush 2001), relative frequency (Torres Cacoullos 2006), and token frequency (Bybee 2002a, 2002b, 2006) were adopted to verify whether frequency of usage in any of these forms would influence the reduction. Results show that relative frequency favors the reduction of para to pra, alongside word class, number of syllables, and stressed syllable. Based on this analysis, the lexical storage of multiword sequences is discussed, along with its effects for processes of linguistic variation and change.

摘要 本文分析了巴西葡萄牙语中介词para（词义为“至、为”）弱化为pra的现象。本文从口语语料库中提取「前接词+para或pra」的序列，通过统计分析探究影响该弱化过程的相关因素；并针对其中30个最常见的此类序列开展专项统计分析，以明确潜在的影响因子。本研究采用串频度（string frequency，Krug 1998、2003；Bush 2001）、相对频度（relative frequency，Torres Cacoullos 2006）以及词例频度（token frequency，Bybee 2002a、2002b、2006）的相关概念，验证各类使用频度是否会对该弱化现象产生影响。研究结果显示，相对频度、词类、音节数量以及重读音节位置均会促进para向pra的弱化。基于上述分析，本文进一步探讨了多词序列的词汇存储机制，及其对语言变异与语言演变进程的影响。