I liked that she used the term "insight," because a word like "said" is often considered to be a "sight word" - a term which implies that one must recognize the word by sight, and use one's visual memory to spell it. Gina Cooke (Linguist-Educator Exchange) uses the term "insight words" instead of "sight words" because, with a little investigation of a word's morphology, the spelling may not seem so random.
Word study (investigating a word's history and its meaningful parts) gives us insight into a word's spelling, giving us deeper understanding beyond visual memorization of letters. This typically results in better, more accurate spelling, especially for kids who struggle with word-form memory (as is often the case with children with dyslexia and/or dysgraphia).
So, I wrote to our client:
I get my understanding of the word "said" from Pete Bowers (Word Works Kingston). Here is what I understand from his tutelage:
We know "said" is the past tense form of "say" - so these words are definitely related. We also know that we cannot seem to assume that "said" is merely <say> + ed, because even if we change <y> to an <i> it would be spelled "saied." We do not drop letters from suffixes, so we cannot conclude that the <e> was dropped from the -ed suffix.
Therefore, the word "said," while clearly a past-tense word, does not use the productive suffix of -ed to mark its past-tense status. But that's ok! While -ed is a common way to mark past tense, we have many words whose past-tense forms are just irregular. (i.e. ran, slept, paid).
Let's first think of other related words to "said" that we know we CAN break into word sums based on our morphology knowledge so far:
says --> say + s
saying --> say + ing
saith (a very old form, as in "saith the Lord") --> sai + th (-th is an Old English Suffix, no longer in use today because it has been replaced by -s).
We also know that <ay> and <ai> spellings are related vowel teams that say the long /a/ sound. And we note that we seem to have several irregular past-tense words that follow the <ay> to <ai> conversion pattern:
Looking back at "saith" (sai + th), we now know that <-th> is an old suffix that we no longer use, so we call it a non-productive suffix. Past-tense suffixes like <-ed> are productive, because it is still used to create new past-tense versions of words. Pete comments that <-d> is also a non-productive suffix, one that is no longer in use today. Both "said" and "saith" are very old words, and thus may both have non-productive suffixes that we no longer recognize widely today.
Thus, we have:
sai + d --> said
sai + th --> saith
say + s --> says
say + ing --> saying
SO, what have we learned? Well, in our initial word sum we tried to connect the words "said" and "say" as if they shared a common base of <say> (i.e.: say + ed). However, once our word sum proved this to be incorrect, we have to pause and consider that these two words (while clearly related in meaning) may NOT share a base, but rather only share a root. If you look up "said" on Etymoline, it will point you to its relative "say," and give the Old English root: secgan - "to utter, inform, speak, tell, relate."
We now know that "said" and "say" do not share a base, but they are both derived from the Old English root secgan.
If you'd like to read Pete's blog on "said" for a slightly more in-depth discussion/explanation, here is the link:
This is, indeed, a very tricky word! But hopefully after all this discussion and investigation, the spelling will stick in your child's head, which is our goal!