I'll have to tackle the Shakespeare question later, but I agree with fugu. There's a good chance that he either coined or popularized many of those phrases, though of course there's always the chance that someone else used them first and we just don't have a record of it.
Here's the deal with
says and
said. Very common words tend to change their pronunciation faster than other words, so a word like
say is more likely to change than a phonologically similar but less frequent word like
pay. One of the main ways in which frequent words change is in reduction of vowels, which can mean shortening, laxing, and destressing.
Said has a short, lax vowel, while
say has a long, tense vowel (well, technically a diphthong, but that's irrelevant).
So why would
said and
says get reduced vowels while
say and
saying don't? Because they have coda consonants, and this makes them phonologically heavier. Originally they had the long, tense vowel plus the consonant at the end, which gives them the structure CVVC (consonant–long vowel–consonant) (because long vowels count for two). Normally this is fine, but super-frequent words don't like being so heavy, so they drop the extra vowel weight and become CVC. But
say is just CVV, which isn't heavy enough to trigger reduction.
Does that make sense? More importantly, would that make sense to a six-year-old and a four-year-old?
