Reading “The growth of the transitivising Reaction Object Construction” by Tamara Bouso Rivas.

1. Words I do not know

construe acquiesce squeal concomitantly plash collostructional waning bray neigh snuffle warble grawl whoop hoot coo guffaw pshaw

2. Terms I do not know

2.6. compositionality

Is it a capability to be inserted into other constructions? Or ability to accept subconstructions?

2.7. transitivisation

Is it the capability to accept an object? As in “greet someone”, but not “sit a chair”.

2.23. echoic verbs

3. Notes

3.2. I do not like the linguists’ citation style. [Author, year] is much better! UPD: Ah, fuck, it’s not a citation, it’s a forward-reference to an example! Fuck forward references!

3.3. Page 240, line 14, “means and manner” – “means” supposedly means “a method by which”

3.4. Page 240, line 14, “means and manner” – “manner” probably means “style of doing”

3.5. Page 240, line 20, Why is “she smiled” called a “monovalent” argument structure? Shan’t it be zero-valent?

3.6. Page 240, line 21, “thanks” and “adoration” are called “non-prototypical” objects. What would be the prototypical ones?

3.7. Page 240, Modern English has much more verbs that can be both transitive and intransitive.

3.9. Can the reference list at the end of the paper include full names of the scientists?

It is super annoying because many people have the same name.

3.11. Page 241 line 5, “, or verbs that”, is it a clarification or an option?

3.14. Page 241, line 22, Why is “for-dwīnan” the word that means “sleep”?

3.15. Page 241, last line. Is “discourse” the same thing as “context”?

3.17. “the remainder of the article” -> “the rest of the article”

3.18. Page 243, line -2, what is this formal notation for the construction structure? (SUBJi [VintrOBJi])?

3.19. Page 243, line -1. Lack compositionality as meaning? What does this mean?

3.20. Page 244, line 6. The “definition” of Goldberg is not understandable.

3.22. Page 247, line 21. This syntactic notation probably can be reformulated in a Backus-Naur form to be more sound.

3.26. Page 247, line 26 “The proposed schema …” sentence is super long and confusing.

3.27. By the page 248, the way-construction has been mentioned many times, but no example has been given.

3.28. Page 248, paragraph 1, speaks about polysemy. But spotting polysemy begs for an algorithm of lifting polysemy for the analysing agent, after the parsing process has let us indentify the ROC construction.

3.29. Page 248, line 11, the “manner interpretation with while” has no example.

3.30. Page 249, line 7, yet another type of English, Early Modern English, I’m completely lost already. Can there be made a table?

3.31. Page 249, same ambiguity: “Similarly to the way-construction, in the ROC in the means subschema the

verb describes the means whereby a reaction or an emotion is expressed.“ What is a subitem of what? As it is written, it seems that the ROC is a subitem of the ”means subschema“, not the other way round. I would have written it as ”Similarly to the way-construction, in the ROC, in the means subschema, the verb describes the means whereby a reaction or an emotion is expressed.“ As ”in the means subschema“ is a detalisation of the previous entity (that is the ROC).

3.32. Page 250, line 19-20. Is there a full list of verbs used in the ROC? Having Google and corpora, it shouldn’t be too hard to make an extensive list.

3.33. Page 252, line 18, the regular expressions for search would look better in a floating figure.

3.34. Page 256,Figure 1. It would be better if the “English Periods” be marked at the X-axis.

3.35. Page 257. I am still confused at how exactly the analysis was performed. There should be some code doing it? Nobody can realistically read through all that text and spot ROCs.

4. Wrap-up

It took me about 6 hours to read the full text. These 6 hours spanned 2 long and 1 short session.

The main thing that is worth mentioning is the almost total absence of the research protocols. (There is just one regular expression presented for querying some not very well specified database.) This research is therefore not reproducible. Neither the databases queried are well specified (except OED), nor the analytical procedures in the form of a code, or at least a natural language numbered list of actions.

The second thing that is probably not that strictly required, but would be almost obvious to include is the application part. Nothing is said on the application part of this study. Clearly, the most obvious application would be a software subroutine for identifying ROCs in a text to aid the readers and even more importantly, the translators in spotting those expressions in a text. Secondly, a software tool could help in identifying which subschema of the ROC is employed in each particular case, and thus aid human or AI translators in finding better expressions in the target language.

Thirdly (although way beyond the scope of this paper), a tool could be written to identify non-ROC patterns in the source non-English languages that would admit a solid translation into English with the help of a ROC.

Fourthly, it would be interesting to forecast which verbs, which are non-transitive yet, are the most likely to undergo transitivisation in the nearest future. Such a forecast, if successful, can make present day software more future-proof and more robust. If not successful, the forecast would serve as a falsifying tool (in the Popperian sense), and would indicate that some deeper process may be actually taking place, and suggest reassessing the described phenomena in a more abstract framework.

In any case, it was a fascinating reading. I am happy to be introduced into the world of language analysis.