You can find the occasional author who wasn't too happy about the prospect of their text being divided up. None other than the Book of Revelation ends:

I warn everyone who hears the words of the prophecy of this book: if anyone adds to them, God will add to that person the plagues described in this book; if anyone takes away from the words of the book of this prophecy, God will take away that person’s share in the tree of life and in the holy city, which are described in this book.

In medieval Europe and West Africa, however, it's quite clear that regardless of any authorial intention or hope, people were transmitting and receiving texts in abridged, modified, serial, remixed, rewritten, and partial form. I want to talk about both areas because we associate medieval Mali more with oral tradition (though there was quite a thriving Islamic literary culture!) and later medieval Europe more with written, manuscript culture (though reading aloud and song/poem performance were still vital, given literacy rates around or below 10%).

You can actually see the Mande (West Africa) practice for yourself in the published versions of Sundiata, the famous epic of medieval Malian kingship. Yes, versions, plural. Most of us familiar with it through school know D.T. Niane's Sundiata: An Epic of Medieval Mali. As you find out in the book, he consolidated multiple performances by djeli Mamadou Kouyate into one unified tale and translated it to French and prose. It's a nice, full story. However, you can also buy the Penguin Classics edition Sunjata, which contains two different verse versions of the epic performed by Bamba Suso and Banna Kanute. They are strikingly different from each other, and from Niane's streamline. (And Daniel Biebuyck briefly mentions a version where Sundiata is eventually transformed into a hippopotamus...)

This, as they say, is a feature not a bug! It is the responsibility of a djeli to respond to the composition of their audience, or perhaps the social situation of their sponsors, by selecting, adjusting, and modifying episodes in a given performance. Sometimes one episode in an epic serves as the starting point for an appropriate spin-off tale. And where a djeli has the ear of a king or other powerful figure, the strategic abridgement or excerpting of an epic serves as a form of emotional/narrative advising--placing certain values and types of actions in the mindset of someone who, in the djeli's opinion, ought to be taking them (or perhaps offering an anti-model). The fluidity of Mande oral storytelling here serves a purpose--it's hardly an accident of "oops, that's a long poem and we're tired now, good night."

From medieval Europe, unfortunately, we don't have a storytelling tradition strong and vibrant enough to survive modernity. However, surviving manuscripts show a deep commitment to a similar culture of excerpts, mash-ups, spin-offs, and alternate universes in narrative and instructional-religious texts alike.

You can identify practical reasons for this, too, of course. In addition to the popularity of reading texts aloud--thus creating the time problem, given the difficulty of assembling the same group every night (barring a setup such as Boccaccio gives his Decameronians)--there is the expense and unwieldiness of the physical manuscript! There's a reason there are very few full parchment Bibles. Instead, we have Gospelbooks, one or the other Testaments, individual books, epistolaries, Psalters (all the Psalters), etc. But like in West Africa, the material situation was integrated into literary culture rather than being the end of the story.

The most famous example here is probably Chaucer's Canterbury Tales. The manuscript situations of the Tales is...kind of insane. No complete version survives--if there ever was a complete version. Different manuscripts have different groupings. Some tales tend to travel together (if a ms has one tale, it will also have a partner), especially if there are internal textual clues linking one tale and its closing bit to another tale's prologue.

Langland's Piers Plowman, also from the late 14th century, offers a different case of a "no platonic ideal" text. Scholars have identified three--count 'em three--distinct versions of the poem (creatively called A, B, and C). Oh, yeah, and some scholars posit the existence of a fourth, called Z, for which we have no copies. A is short, breaking off partway through the narrative it sets out to tell. B goes the distance (like, rise of Antichrist/end of the world distance). C veers off a different path from B. Which did Langland "intend" to be read?

And what was true in "fiction" was also true in "non-fiction." (Those are anachronistic and ill-fitting labels for medieval texts, but it's a good enough approximation for present purposes). A boring way to describe what I actually study is "late medieval German didactic literature," that is, texts meant to instruct readers about various religious topics. Morality, behavior, the Eucharist, how to do confession best, how to pray, what to pray about, &c &c. Well let me tell you--even the tip-top scholars who publish the major overviews of medieval literature get to the 15th century and have to pull out adjectives like "repetitive" and "monotonous". They're not wrong. If you sit down with Otto von Passau's 24 Elders or Johannes Nider's 24 Golden Harps or the sermon collection Dormi secure (sermons were popular texts for individual reading as well), you'll learn the same things over and over and over--both among different elders' tales, and within the same one.

Why? Because people were rarely reading or even having access to whole texts until the age of print. (You can find 15th and 16th century versions of Otto on Google Books/archive.org, for example). Even a shorter and INSANELY popular text like Henry Suso's Little Book of Wisdom had individual chapters circulating independently that were even more popular than the full book. The popularity of the short versions was even more true for books that taught proper procedure for the sacrament of confession!

My favorite example here, though, is the German vernacular translation of the Speculum peccatoris, called the Sundenspiegel or "Mirror for Sinners." It's a pretty generic didactic text, mostly moral teaching using the vices and virtues in various forms (seven deadly, nine alien, four that cry to heaven, &c &c). It's so generic, in fact, that one manuscript family is only half Sundenspiegel. Right in the middle it switches to a different text entirely--and the transition wouldn't be noticeable unless one was familiar with the S.p. or the other Sundenspiegel families. It didn't matter for the integrity of those copies of the text, and it apparently didn't matter for the integrity of the OG Sundenspiegel to have only part of it passed down.

So--collections of tales, longer narrative works, and nonfiction/instructional literature. If medieval manuscripts can be a guide to how readers experienced texts, we can see that western Europeans were no more faithful to the One True Version of a story than their West African counterparts.

There are some cases where completion might have been valued. In monastic tradition, monks and nuns were supposed to receive a book on Ash Wednesday and spend Lent going through it, for example. However, the much more typical experience of medieval literary culture--written, oral, European, African--can be summed up by the many prophecies added onto and interpretations sucked material from poor John of Patmos, whose thunderclap threat at the end of Revelation went massively unheeded.

The earliest written Greek hexameter epics, the Homeric Poems and the Hesiodic Poems, were the result of a centuries-old oral tradition. Epic, after all, was the product of ἀοιδοί, singers (from ἀείδω, "to sing," c.f. ἀοιδή, "song") just as lyric was by nature melodic and musical. In this oral tradition, as in so many other oral poetic traditions across the world throughout time, oral poets composed verse consisting of line fragments, formulas, and memorized chunks, that were combined and recombined in ways that fit the meter (or were stretched in order to make it fit) and plot. Certain themes and general plots made up the Epic Cycle--the νόστοι (homecomings), the Ἰλίου πέρσις (sack of Troy), the Theban Cycle (both the original Seven and the Ἐπίγονοι), etc.--but the precise details were not always the same, even though the bards themselves might insist they were, as Alfred Lord discovered when he recorded the Yugoslavian poets in the twentieth century. Indeed, Homer presents plenty of stories that appear to deviate even from what Hesiod says or are contradicted by other parts of Homer. In one famous case, the duel between Menelaus and Paris in Iliad 3 seems totally out of place in the ninth year of the war, and is clearly a part of the story of Troy that chronologically would have taken place at the beginning of the war but that the poet inserted in "anachronistically." This sort of interweaving of bits and pieces of oral tradition must be understood as the foundation of the early Greek epic tradition before anything else can be said.

Each of the two major Homeric epic poems, the Iliad and the Odyssey, are too long to be recited in a single sitting as they are written down for us, taking up over twenty hours to recite all at once. How they were initially recited when they were written down sometime during the Dark Age or early Archaic Period is something of a mystery, but they certainly were not recited all at once during the Classical Period. Although by this time the poems had been committed to text (albeit in various versions, before the Hellenistic and later recensions) and any educated Greek or Roman could recite by heart large chunks, or even the entirety, of Homer, in the classical world poetry remained a predominantly performative form, one that was subject to recitation. By the Archaic and Classical Periods we see the existence of Rhapsodes. The name (ῥαψῳδός) derives from ῥάπτω (to sew together) and ᾠδή (hence, "ode"), which gives an idea of how the early Greek epics worked--the Greeks at least thought of them as woven together, and the modern scholarly mind from there argues in favor of the oral composition of the poems from known fragments. The history and early identity of the Rhapsodes is controversial, but whatever they were early on by the Classical Period proper they had become pretty much exclusively public reciters of Homeric verse. How precisely the Rhapsodes did this is, again, not totally clear. They apparently memorized some fixed text, although until the later recensions dozens if not hundreds of local variants of the Homeric text were floating around the Greek-speaking world, and one rhapsodic school's text was not necessarily the same as another's. They performed at public festivals (we know specifically that they performed by law at the Panathenaea) and also in competition--how competition would have worked with what was, at least in theory, supposed to be a fixed text is disputed. Plato (or Pseudo-Plato?) tells us in the Hipparchus that Pisistratus established that the Rhapsodes were to recite the Homeric Poems in order, picking up where the previous one had left off, and Diogenes Laertius attributes a similar law to Solon. But how, and for how long? Homeric scholars debate this to sort of ridiculous lengths, and it's not even entirely clear whether what the Rhapsodes were reciting was in fact "Homer," or whether it was "Homer"+improvisation. It's certainly not at all clear that the Homeric Poems were recited by Rhapsodes reciting one after another in their entirety at the Panathenaea.

But in any case the nature of Homeric performance appears to be one of discrete units, whether or not the entire Homeric corpus was recited in its entirety in sequence at the Panathenaea, which remains only the best known of the countless Rhapsodic competitions and festivals. Likewise, aside from the constant allusion to or quotation of Homer in our texts we find references all the time to individuals who could recite or allude to large sections of Homer independently of each other. Socrates, in the Ion, gives a list particular scenes that the titular Rhapsode Ion might perform, for example. These sort of details lead to lots of speculation about the composition of the poem, including some (generally fruitless) attempts to break Homer up into its constituent "original" mini-epics or to argue that Rhapsodes must have recited passages in pairs or as a group. This sort of speculation is beside the point, and usually doesn't go anywhere--I'd be lying if I didn't admit that I think that Homericists tend to go around in circles a lot. What's clear is that Homer could be and was chopped up into pieces regularly, and though one might read the text all at once (presumably not in one go, people need to sleep) it's not even clear whether one could have heard the text recited in its entirety by the Classical Period over the course of several days, regardless of how the poems were "originally" performed. It must also be mentioned that this sort of recitation, which is not necessarily the same as the oral composition (which might pick up or end at any particular episode the poet pleases, the way one alternate ending to the text of the Iliad ends with a lead-in to the Aethiopis), requires some degree of knowledge from the audience. If Rhapsodes recited particular scenes or, even if they didn't, with respect to the way that we know people could quote or allude to large sections of the Homeric Poems then to follow the audience must have known basically the outline of the Epic Cycle. Which is not hard to imagine, any more than it is hard to imagine that a modern American or British schoolchild would know the basic stories about Prometheus bringing fire or whatever, even if they haven't read Hesiod, the Prometheus Bound, or Plato's dialog Protagoras from which those stories come. These were the fundamental cultural tales of the Greek-speaking world, and Homer was their original formulation in text. Anyone in a Rhapsode's audience could recognize the scene he recited, even individual lines--perhaps even recite the entire thing himself!--as he recited, because he himself had read the same text, or an alternate local version of it. That was not why he listened to the recitations of Homer.

I've tried to stay away from saying something too controversial. I hope I've succeeded. I'm not a Homerist and the consensus (if there even is one) on Homer and the Homeric Poems shifts rather rapidly in its details, so if any Homerist wants to jump in and tell me everything is awful and life is suffering by all means go for it, I won't be especially offended.

Virgil's another story, though, although the answer is still basically the same. The Aeneid is also too long to be recited comfortably in one sitting, and like Greek hexameter epic Latin hexameter was also performed. Virgil, of course, composed his poem in a decidedly literate, post-oral society, and it shows in the composition of the work. The Aeneid has clear breaks where scenes end and change, where the action stops. By contrast the division of Homer into books was made by the Alexandrian scholars in the Hellenistic Period, and the Homeric Poems flow more or less seamlessly from one scene to the other, with few relatively few sharp breaks. Poetry in the Roman world was frequently recited, though it was also read privately or in small groups. We know that Virgil recited parts of the Aeneid on several occasions, such as the incident when Donatus tells us Octavia fainted during Virgil's recitation of Aeneid 6 when he reached the half-line, "tu Marcellus eris." Virgil never recited the poem in its entirety, however, in any complete state, simply because the poem was not quite completed at the time of his death. He had, however, recited at least three books of it to Augustus while working on it, and Augustus of course published the poem after Virgil's death despite the poet's supposed desire that it be burned. This, and the actual content and organization of the poem, gives an indication of how the poem's reception worked. Unlike with the Homeric Poems, in which scenes and verse are truly woven into each other seamlessly, the Aeneid falls quite conveniently into a series of episodic books, which can be recited easily (even a few at a time) in a single sitting and then picked up later on. This sort of division of hexameter epic into discrete episodes corresponding usually to books was not an innovation of Virgil, but was a result of the gradual spread of a truly literary society, working from written texts. Already in the third century Apollonius Rhodius' Argonautica divided neatly into four books (albeit quite long ones) sorted more or less by connected scenes. Ennius' Latin Annales appear to be similar, to the point that one scholar whose talk I recently attended suggested that Ennius may well have been known to literate Romans more for particular episodes rather than as a complete text. In any case, no matter how well the text of Ennius, Homer, or anybody else was known in its entirety, generally speaking epic poems were read and performed in chunks by the Roman Period, since they were too long to read all at once, and the structure of epic adapted in conjunction with the development of a literary society (also to fit onto book scrolls).