The transmission of a text is the process by which it goes from the original writing of the author to the reader. In the case of a letter written today, usually the reader has before him the exact version written by the author. On an Internet forum or message board, a moderator may change something in the original text before it is presented to the reader, putting one more step between author and reader.
In the case of older texts, written in ancient or medieval times, things may be more complicated, many more steps may be involved. In other words: the transmission may be much more complicated.
Scholars have found some of the original copies of personal letters and shopping lists and written instructions from an employer to an employee, things like that, from the Middle Ages and some even from before. For example, this letter, one of thousands of pieces of ancient papyrus unearthed at the site of the ancient city of Oxyrhynchus in Egypt, has been dated to the 2nd century AD:
It has been translated as follows:
Thais to her own Tigrius, greeting.
I wrote to Apolinarius to come to Petne for the measuring. Apolinarius will tell you how the situation stands concerning the deposits and public dues. He will let you know the name of the person involved.
If you come, take out six measures of vegetable seed and seal them in the sacks, so that they may be ready. And if you can, please go up and find out about the donkey.
Sarapodora and Sabinus salute you. Do not sell the young pigs without consulting me. Good bye.
Not an earthshaking communication, perhaps, but if one examines a lot of such documents, together they may be very helpful in forming a mental picture of past places and times.
Then there are the texts referred to as "literary," intended for a larger audience: besides the genres we may think of as "literature" in a narrower sense, fiction and poetry and drama, these include works of history and philosophy and science. There's just one copy, from before the age of mechanical and electronic reproduction, of Thaius' letter to Tigrius, presumably either written in Thaius' own hand or dictated to a servant. There may be many copies of a given ancient literary text, but it's rare to find one made within even several centuries of its original composition. For example, as I recently found out, in the case of Sallust' histories of the Cataline conspiracy and the Jugurthine War, written in the 1st century BC, there appear to be over 500 manuscripts in publicly-accessible collections today. But apart from 4 fragments preserved in scraps of papyrus from the 4th and 5th centuries, none of these manuscripts is older than the 9th century. Here's the left edge of one of those papyri:
The fact that several of the manuscripts of Sallust's works are as old as the 9th century is very good. The fact that there are also some older ones known, even if they are just scraps, is exceptional. It's good from a point of view of the manuscripts as objects from the 9th century of historical interest in their own right, and it's also good on the general assumption that the older a manuscript is, the greater the chances are that it preserves the original text, that which Sallust actually wrote, with some sort of accuracy. That's a very general assumption. We don't generally know, for any given manuscripts of an ancient text, how many copies may lie between it and the original. A 4th-century papyrus may be a very sloppy copy of a very sloppy copy of a very sloppy copy... repeat many more times, of the original. On the other hand, a 15th century manuscript may be a very accurate copy of a very much older manuscript which was copied from Sallust's own personal copy. Manuscripts are judged on other criteria than age. But generally speaking, for those interested in accurately reconstructing the original text of an ancient author, when it comes to manuscripts, old is good and very old is very good.
And with very few exceptions, until the last couple of centuries, 9th century was about as old as any surviving manuscripts of pre-Christian Classical authors were. This is one of the main reasons why so many scholars point to the reign of Charlemagne as the end of the Dark Ages: because he instituted an educational program, including the study of those ancient pagans, and many of those 9th-century copies were made because of him. So why don't we have many of the copies from which the 9th-century copies were made? Because, before the Italian Renaissance in the 14th century, it rarely seems to have occurred to anyone in Western Europe that a manuscript -- or a building or anything else -- might be worth preserving simply because it was old. New copies were made, and the old, worn-out ones were thrown out. Some of those pre-9th-century exceptions include 8 4th and 5th century manuscripts of Vergil, and 4 5th century manuscripts of Livy -- at least 4. I know of 4, in addition to some of the papyri described below.
Many more older manuscripts have been found by archaeologists from the 19th century onward, written on papyrus and buried in the desert south and east of the Mediterranean, where, it turns out, papyrus can last for a very, very long time without decomposing. By far the most famous of these finds has been the Dead Sea Scrolls, but that discovery was just one of many. Most of the finds have just been scraps, like the papyrus of Sallust illustrated above, but still, because of their age, they're very exciting to students of ancient literature.
Here's a fragment of the Gospel of John, believed to have been copied out in the first half of the 2nd century, very close to the time that this text was originally written:
Speaking strictly as a layman, let that be perfectly clear, the general impression I get from the comparisons of these discoveries of old papyri with medieval manuscripts and with modern editions of ancient texts is that the medieval scribes tended to be very scrupulous and accurate and that the modern editors tend to be very good at their jobs. I know I could never do what they have done, and I'm very grateful for their efforts.