Working with Arthurian Legend Source Material
I spent more years than I'd like to admit trying to figure out what medieval authors were actually saying versus what later retellings added. The Lady of the Lake comes up constantly in Arthurian stuff, and most people walking into this don't realize how much the source material contradicts itself depending on which century you pull from. Let's get the basics straight first. The Lady of the Lake, known as Vivien in some traditions and Nimue in others, is a liminal figure who appears across medieval romance and chronicle. She gives Arthur the sword Excalibur in many versions, raises Galahad, and in the later Welsh-influenced texts she's the one who traps Merlin. The name changes alone should tell you this isn't one coherent story. It's a tradition that accumulated variants over roughly four hundred years of writing. The earliest solid references show up in Chrétien de Troyes's "Lancelot, the Knight of the Cart" from around 1177. He calls her Nimue and has her raise Lancelot. Wace brings her into his "Roman de Brut" around 1155. Then Marie de France mentions her in "Lanval." By the time you get to Thomas Malory's "Le Morte d'Arthur" in 1485, the character has merged elements from multiple earlier traditions into one somewhat consistent figure. That's the version most modern retellings use, and that's also where the biggest problems start for anyone doing serious research.
I ran into this exact issue when I was trying to map the different versions for a paper on Arthurian transmission. If you cite Malory as if he's the single authoritative source, you're going to frustrate anyone who's actually read the French and Welsh material that came before him. The trick is deciding which tradition you're working in and staying consistent about it. Don't blend the Vulgate Cycle's version with Malory's without acknowledging the divergence.
Common Texts and Where to Find Them
There isn't really a single definitive edition of these texts. The manuscripts vary, scribes made changes, and early printed editions sometimes introduced errors that later editors just copied along. Here's what I actually use when I need to check something: The Loeb Classical Library edition of Malory is decent for the English but you'll want something better for the French sources. The "Lancelot-Grimaud" cycle is available in the five-volume edition edited by Michel Zink and published by Gallimard, but that's only in French. For English translations, the Oxford World's Classics version of Malory edited by Stephen H. A. Henderson is serviceable, though P.J. Granskou's translation of the "Prose Lancelot" fragments is better if you can find it. The "Post-Vulgate Cycle" has no complete English translation, which is annoying because that's where a lot of the Lady of the Lake material diverges from earlier versions. The Welsh angle matters more than most people give it credit for. The "Mabinogion" tales, especially "Peredur" and "Gereint en Enid," contain figures that scholars associate with the Lady of the Lake tradition, though the connections are interpretive rather than explicit. The "Triads of the Island of Britain" reference her indirectly. If you're only reading French and English sources, you're missing a significant strand of the tradition.
Get the Full Details

What Most People Get Wrong
Two issues come up constantly, and both are worth pointing out early because they affect how you interpret everything else. First, the sword in her lake is not always Excalibur. In the earlier French romance tradition, the sword she gives is just a magical weapon. The name Excalibur gets attached later, and even then it's inconsistent. In Malory, Arthur gets one sword from the stone and another from the lake. The lake sword is the one with the scabbard that keeps him alive. Confusing these two swords is the most common mistake I see in casual discussions of the legend. They're separate objects in the primary text. Second, the enchantress who tricks Merlin is not the same character across all versions. In Robert de Boron's "Joseph d'Arimathie," the enchantress is simply unnamed. In the later prose "Merlin," she becomes Vivien, a named antagonist. In the Welsh tradition, the figure trapping the magician has different characteristics entirely. The conflation of these figures into one character happened gradually through the medieval period, and modern popular culture just locked it into place without that history.
I learned this the hard way when someone pointed out an error I'd made in citing a source. I had treated Vivien as the standard name for the Lady of the Lake throughout the entire tradition. She isn't. Using the wrong name in the wrong context signals that you haven't actually looked at the primary texts.
Practical Research Approach
Start with Malory if you're new to this, but treat it as an entry point, not the end of the investigation. Read the "Book of the Sword" section carefully. Then move backward. The "Lancelot Proper" in the Vulgate Cycle gives you the earlier French treatment. After that, check the Welsh material for the pre-French substrate. When you hit a detail you're unsure about, don't trust a secondary summary. Look at the original manuscript description or a critical edition with apparatus. The differences between manuscripts are where the interesting scholarship lives. A single variant in a Middle English manuscript can change the entire interpretation of a character's role. If you need quick reference points, the "Arthurian Dictionary" by Linda Malcor and the "Companion to Arthurian Literature" edited by Helen Fulton are both reliable. They won't replace the texts, but they'll save you from making elementary errors about which version contains what. I keep both on my desk.

The tradition is messy because it's a tradition, not a single authored work. Accept that upfront and you'll spend less time confused about contradictions and more time reading the actual material.