Words about . . . Words

Two of the readings for this week seemed to bridge the gap between morphology and semantics.  In other words, they described words in ways that they wouldn't be described in a traditional dictionary.  The Atkins-Levin reading suggested that dictionaries for lexicographers should address the concept of internal vs. external causation.  Internally caused verbs like quake (for which you can quake but someone/something else can't quake you) are often listed as synonyms for externally caused verbs like shake (for which you can shake someone or someone can shake you), which doesn't address the nuance between these two verb types.  The Slobin reading discussed how "frog stories" are often told in different ways across languages.  It can be interesting to study which parts of the story speakers of different languages are left to infer.  To do this, Slobin focused specifically on motion words like pop up and climb in oral narratives.

The other two seemed to be more technical texts designed to acquaint readers with various tools of morphology, all of which could be applicable to lexicographers.  The first Haspelmath reading defined many morphological terms, such as lexeme (like read) and morpheme (like read or s).  Haspelmath then zeroed in on word parts like affixes, bases, roots, and allomorphs.  The second Haspelmath reading discussed compounds, such as light year (endocentric since light describes year) and pickpocket (exocentric since pick describes pocket).  Linguists often try to formulate rules to describe how compounding works, much like they do for how sounds go together in phonetics or how kinds of words go together in syntax.  They represent these morphological rules in tree structures.

These four readings made me think about what information should be represented in dictionaries.  Dictionaries should certainly list the morphemes within each word, since that information is useful for understanding where words come from.  I also think it's necessary to describe which word relies on which for compound words like light year.  As for internal vs. external causation, I don't necessarily agree with Atkins-Levin that this information should be excluded from dictionaries.  I think it could help non-native English speakers understand the nuances between different verbs, and even native English speakers might be able to use this info when writing.  At the same time, including all this extra information would increase the size of dictionaries and perhaps make them less user-friendly.  With all this in mind, my verdict is that the causation of verbs should be excluded from print dictionaries but included in online dictionaries via a link that says something like More Information.  That way, everyone with an Internet connection would have access to this data, but the information wouldn't interfere with readability.  I also think the info about how a word is used in oral narratives could be nice to include in the More Information tab.  Although humans might not use it much, such information would be invaluable to the translation programs we worked with (like Siri and Google Translate).