Apparata - A Really Interesting "Neo-Classical Plural"

|

If you're familiar with the alumnus/alumni plural rule and the symoposium/symposia rule, then you know about what I will call the "classical" plural or Latin or Greek words in which we match the Latin/Greek singular with the Latin/Greek plural.

You may also be familiar with the "neo-classical" plurals in which English speakers create Latinate plurals which never actually existed in Latin (these are probably a class of hypercorrection). A famous case octopi (many octopuses). As Ask Oxford will explain, octopus (ὀκτώπους) is actually a Greek word with the Greek plural octopodes.

I like to collect these - not to make fun of people who don't know enough Latin or Greek, but because they show interesting insights into how grammars work with foreign borrowings. The one thing speakers DON'T do is replicate the grammar of the original language (although they may be trying).

As you may imagine, English speakers who encounter any vaguely technical term ending with -us will wonder if the plural is -i or not. Hence we see other new inventions like opus~opi (original Latin plural was opera) and omnibus~omnibi (techincally omnibus is already plural in Latin). In fact one speaker asked if an -is word had an -i plural (warning: taboo word ahead in link). This is not an odd question because both -us and -is are pronounced /əs/.

But one which was unique was someone said "apparatuses...apparata"? First I thought, shouldn't that be apparati? I know I never hear apparati, but that is what I would expect on a naive level, and it may even be the original Latin plural.

Instead, the speaker is using the -um ~ -a plural similar to symposium/symposia, but in English this is normally restricted to -um or some -on words. Is this speaker actually conflating the two rules to one Classical plural -a?

Or...is the final -at syllabus in the root throwing him off. Other famous -a plurals include automata, errata, strata and data. It's possibly that this speaker has a rule that any techincal term with a final -at in the root will have an -a plural. An interesting kink in the long thread of "neo-classic" plurals.

And a final message to my anonymous data source - please don't be embarrassed. We're ALL confused by this technical jargon.