The Lexical Integrity Hypothesis

Plag 2003 establishes that all compounds, even extremely long ones, can be analysed as essentially binary structures. This is due to the attribute of compounds known as recursivity, wherein compounds, especially noun-noun compounds, can be continuously stacked syntactically onto one another to form longer compounds.

However, after the creation of this principle, the nature of the units that can be combined to create compounds then come into question. The most important piece of controversy is the incorporation of syntactic phrases as the dependent members of compounds.

Syntactic phrases, if incorporated into compounds, are far more likely to be dependents than semantic heads (Plag 2003). Lieber 1992 further hypothesised that only lexicalised

The Lexical Integrity Hypothesis as stated by Booji (2005) and Spencer (2005) is outlined as follows:

Syntax cannot manipulate internal structure of words. Syntax cannot enter into the internal structure of words.

Phrasal compounds pose a large problem for the Lexical Integrity Hypothesis because phrasal compounds entail syntactic phrases being incorporated into a compound, which itself arguably acts like a linguistic unit. The claim that the phrasal component of phrasal compounds have to be set phrases, if true, would allow phrasal compounds to be reconciled with the Lexical Integrity Hypothesis. This is because if true, the set phrases could then be considered to be lexical entries and would not have to be considered under the rules of syntax.

This paper will examine the validity of the theory that phrasal compounds inherently require the incorporation of a set phrase, using data collected from various sources. This paper will then examine the consequences our findings have on the lexical integrity hypothesis.

Selection Criteria for Compounds and Brief

An example can be seen in the term “Post-industrial rock” as seen in our data. Going by the “American History Teacher” example shown above, the term can be read as either “rock music under the post-industrial genre” or “after industrial rock music”. Notably, the second reading would not even be considered a phrase but would be considered semantically and syntactically incomplete (ie. “Post industrial rock the following occurred…”). So the term can only make sense if read as a compound with the head “rock” and the set-phrase dependent “post-industrial”, thus showing that Spencer’s theory cannot be treated as a catch-all solution to the problem of phrasal