Catching Single-Letter Spelling Errors

Isobel. While is would be possible to exclude lower case a and upper case I from the search, these should be included because they might be errors, for example Buddh a instead of Buddha.

On a large document the quickest way to identify errors is to highlight single letter words using the second of my suggested Find and Replace commands then eyeball the document. It is surprising how quick and effective it is.

Another difficult error to find is where a proper noun is spelt in two different ways with only one character different. The technique is to use a macro to count and sort all the spelling errors then eyeball the results. Two different spellings will be adjacent to one another, for example:

Wungirilwil 2
Wungurilwil 6

After using all the tools in the armory to detect errors then comes the hard part. Reading every word in the document.

Kim was "disappointed" because using my tip still requires a human proofread. It had never occurred to me before that anyone would even dream of sending a document out before a careful proofread. Such an awakening!

Ken: I like your idea (and thank you for pointing out that Wildcards must be checked first) but a tiny correction: the search string should be <{b-z,A-Z]> to exclude the frequently occurring lower-case "a".

It's possible to find single characters by using Wildcards. With Wildcards turned on, type a space followed by [b-z] followed by another space. This will find all the lower-case single characters in the alphabet except for "a". It will not find capital I. All this is what we want it to do.

If the single character is immediately followed by a punctuation mark, however, it will escape the search. There may be "Wildcardese" that can be written for that, but I don't know how and it's probably not worth the trouble.

I'm sorry to say that I found this a rather disappointing tip.
If you are conscientious, you are probably going to proofread your document anyway. But no amount of proofreading can guarantee that all mistakes will be caught.
I was hoping you would discuss using wildcards for search and replace instead. Like typing ^w^?^w in the search field to find all single letters, for instance. Obviously, it would catch a's and I's unnecessarily, but it's still less time-consuming and more reliable than proofreading the entire document.

Find & Replace can find single letter words and you can step through and correct those that need correction. The Find string is:
<{a-z,A-Z]>
"Use Wildcards" must be ticked for this to work.

Another approach is to highlight all single letter words with:
Find string:
(<[a-z,A-Z]>)
Replace string:
\1
and select Format > Highlight. Before running you need to select a highlight color in the Font group in the Home banner.
After the corrections have been made remove highlighting by selecting the whole document and clicking No Color.