Another important point to note is that we might expect IDyOM to do rather better at modelling morphemes rather than syllables; and indeed, inspection of the results in the figure and the wider corpus do support that: false positives in the syllabisation task are often (though not always) at morpheme boundaries.[…]This leads me to the unastonishing prediction that the STM alone would work very poorly for syllabisation, because it will not learn from the corpus, but only from the prefix of each utterance as it proceeds.

Thy substance, valued at the highest rate, / Cannot amount unto a hundred marks.

And there have been Common-wealths that having no more Territory, than hath served them for habitation, have neverthelesse, not onely maintained, but also encreased their Power, partly by the labour of trading from one place to another, and partly by selling the Manifactures, whereof the Materials were brought in from other places.

I do not deem it necessary, while dancing is so universally practiced, and conceded to be an agreeable and innocent diversion from the ordinary duties of life, to quote extracts in its favor from the Bible, ancient Greek authors, historical dictionaries, philosophers, poets, and celebrated men and women, nor even from the numerous works on dancing— consisting of essays, histories, choregraphies and orchesographies—now before me, many of them out of print, dating variously from the year 1700 to the present time.