Separated elements vs Sword as a whole

This is a question that's been nagging at me for a while, and I don't think it has a clean answer — but I want to lay it out and hear what you think.

For decades we've used Oakeshott's typology, which classifies blades, crossguards, and pommels largely as separate elements. It's a useful system, up to a point, and I don't want to be misunderstood as dismissing it. But I think it's worth asking: has it quietly trained us into a kind of selective perception?

Here's what I mean. The overwhelming majority of swords we actually have access to — in archaeological finds, in graves, and above all in art, sculpture, and manuscript illumination — survive as complete objects, not as isolated components. Separated crossguards on their own? I can think of less than 20 well-documented and significant examples. Isolated pommels, a bit more, but still a small number. Against a dataset of well over a thousand swords we have catalogued (medieval material from Europe), these isolated-element finds are a real minority — I'd put it somewhere around 4-5%, though I want to go back and pin that number down more precisely before I lean on it too hard.

So here's my question: why has so much of the analytical framework been built around classifying elements separately, when the evidence we're actually swimming in is complete, whole swords?

In fact, this method's usefulness comes when we find a separate element (a pommel, for example) and can determine its type and general characteristics. But can we theoretically link it to a specific type / form of sword? Not entirely, at the moment. We only have a period and some clues in Oakeshott's description. Meanwhile, most pommel types can be associated with specific sword groups, thus easily and precisely determining the most likely appearance of the entire sword.

The alternative-supplementary classification system I am working on since a few years combines Oakeshott's system with a grouping approach based on shared characteristics of the whole object — because I think a sword's style, proportions, weight distribution, and purpose were conceived together, as one coherent design, shaped by the period and type. Pulling it apart into three independently-classified pieces risks losing exactly that coherence, and I think it introduces ambiguity that a whole-object view often avoids.

In my opinion, the classification of separated elements should be a supplement to the main classification and not its basis.

Now, there's a real counterargument I want to take seriously rather than wave away: swords weren't always assembled once and left untouched. Blades were expensive, often outlived their original hilts, and were frequently rehilted — sometimes generations later, sometimes by a completely different workshop, possibly in a different region entirely. Blade-making and hilt-making were often distinct crafts. So a lot of the 'complete' swords we study aren't really single, unified design acts — they're composites, assembled across time. That's an argument for keeping element-level classification around, at least as a tool: it's how you'd even notice that a blade and its mounts don't belong to the same generation in the first place.

I don't think that fact overturns the case for a whole-object view — if anything, I think it makes the historical and stylistic context around a sword more important, not less. But still, it does mean I'm not arguing element typology should disappear. I still think it earns its place — for exactly this kind of detective work, and for precise scientific description. What I'm questioning is whether it should be the default lens, when the objects in front of us are, overwhelmingly, whole.

So — genuinely asking: when you've used element-by-element classification, what has it given you that a whole-sword, contextual approach couldn't? I've mostly heard 'it gives us precise terminology for publications,' and I don't think that's nothing, but I'm not convinced it's enough to justify treating three parts of one designed object as three unrelated puzzles.

This is not my whim, but rather a result of my many years of experience. Almost every day I look at tons of swords across catalogs, photos, data, exhibitions, and museum publications. This dilemma comes straight out of that daily work — when I'm simply unable to give a proper, full and clear description, because the current tools don't make it easier...