One of the great names in 18th century Bible study was Benjamin Kennicott (Britain; 1718-1783). He made an amazing, unparalleled contribution to textual criticism of the Hebrew Bible, yet not one without flaws.
He undertook a massive project of collating as many Hebrew Bible manuscripts as he could find, sparing no expense, with the purpose of recording its variae lectiones (variant readings). His collation was huge (over 600 manuscripts and more than 50 printed editions), but by no means comprehensive. Following him was Giovanni de Rossi (an Italian Hebraist) who greatly increased the number of manuscripts used in this project). Together this collation formed the foundation of modern textual criticism of the Bible.
Kennicott had the right idea, but was not very critical as far as assigning equal weight to bad and good texts.
As M.H. Goshen-Gottstein (Biblical Philology and the Concordance, Journal of Jewish Studies 8, 1-2 (1957) (1). 5-12) wrote:
"When Kennicott and de Rossi set out to collect "extra-massoretic" readings, they faithfully noted down every mistake, which some scribe or monk who taught himself to write Hebrew cared to make, and we who stand on their shoulders, must not accuse them of being ignorant of the problems involved in their undertaking. Their collections, however, became an unfailing source for the modern scholar, who could "prove" the correctness of the reading in LXX (and other versions) from Hebrew manuscripts, without bothering to figure out in what way ancient aberrant readings could have crept into the post-massoretic manuscripts used by Kennicott and de Rossi."
In any case, Kennicott was not, of course, merely a collector. He was also a scholar.
I found this interesting passage concerning the earlier and later Biblical orthography of the name David (דוד and דויד) in a work of his from 1753 on comparisons between I Chron. XI and II Sam. V and XXIII. He is trying to date these books.
The discussion begins like this, with a comparison of the two texts:
As you can see, he observes that the plene spelling (דויד) is later than the defective spelling (דוד).
Nothing much to add, but a thought which has been on my mind recently. It is important to remember that written languages don't intrinsically have a uniform orthography (i.e., one correct way to spell words). Rather, uniformity is usually imposed eventually at a certain stage in the history of the written language. At an earlier stage languages basically are written according to how the words sound, and lacking dictionaries or some kind of authoritative text to dictate spelling, the spelling is a choice made by the one writing the text. What tends to happen (or tended, since nowadays most written languages have already settled upon uniformity of sorts) is that the influence of widely disseminated, very important texts begins to dictate uniformity in spelling. This is compounded by printing, when massive quantities of texts that are exactly alike can appear. Finally, prescriptive authorities like dictionaries and lexicons can easily finish the job. Indeed, these ingredients were mostly missing from English until the 16th century, and that is why English spelling did not begin to stabilize until then.
In the case of Hebrew, the existence of supremely authoritative texts compounded by the nature of the language itself accelerated this process at an earlier stage. What I mean by the nature of the language (and this applies to other so-called Semitic languages; better described as Triliteral languages) is that many of the peculiarities of English are missing from it and it lends itself to more uniformity, as the consonants are more stable and consistent in their pronunciation. English, however, is highly influenced not only by its Anglo-Saxon roots, but also several stages of Latin and Greek. This accounts for most words that use the letter F or PH for the same sound. Without knowing the etymology, how is one to know how to spell such words? Only through memorization.
If I want to spell the name "Moshe" in Hebrew, without having ever seen what the word looks like, I could probably come up with משה. Perhaps I'd have written מושה, but that's a minor variant. In English, I might choose to spell it Mosche or Mosheh or Mose or any of several other ways. In English different consonants often have exactly the same value. Thus, if I had to spell the name of the Biblical Caleb I might have guessed that I could use a K. In fact, without ever having seen the name before I probably would have--it so happens that C used to be fashionable for the hard K sound, while today K is usually chosen. Although in Hebrew there appears to be little or no difference between כ and ק, which should mean that if I have never seen the word I might choose to spell Caleb either as כלב or קלב, in fact you can tell which consonant is meant without seeing the word quite easily. Just add a לכלב : ל, and the hard K become the guttural /kh/.
Thus it was that Hebrew (sort of) had more uniformity in spelling much, much earlier than English. Hebrew spelling was highly influenced by the Bible. But the Bible itself, particularly the earlier books, could hardly have been influenced by itself. So you find more variation then you do in later stages of Hebrew. One particular kind of variation is whether or not words are written plene (that is , using letters like ו, ה and י). It seems that at its earliest stage Hebrew was always written defective. A noteworthy example is the Mesha Stele, which begins with the word אנכ for the word אנכי, which we are familiar with. What apparently happened was that eventually it was realized that adding little touches like a י could guide the reader in how to pronounce the word.
Thus the name דוד eventually began to be spelled דויד. (Although the former can be pronounced Dod, while the latter cannot be confused for it, it is doubtful that this was the specific reason for the change. It seems hard to believe that later writers really felt that דוד without a י was ambiguous. Rather, the tendency at that point was to fill out defective words to aid in pronunciation overall. Thus, דוד became דויד in those later books of the Bible. Eventually the plene spelling mania tipped back a little bit and went out of style. Once vowel points were invented, they became fairly unnecessary to begin with.)
Now, according to the rules of Tiberian vocalization דוד and דויד are not pronounced exactly alike. The former is a short /i/ vowel and the latter a long /ee/, so we're talking about something like David and Daveed.
In effect, the nekkudot force an artificial pronunciation. This is not to say that the Masoretes invented this dual pronunciation. By the time they devised the nekkudot to record the traditional liturgical chant in all its details, this dual pronunciation could have already existed, hence the Tiberian rule that chirik-yud is /ee/ while chirik-without-yud is /i/.
But the net effect is the same: the nekkudot influenced the pronunciation of the name, thus to this day, the name David is pronounced Daveed in Israeli Hebrew, while there are no grounds for choosing /ee/ over /i/ in trying to reconstruct how the historical name was truly pronounced.