Replicable. Simple enough word meaning "able to be replicated". Now, the other day I needed an antonym for it - I honestly can't remember why - and rather than using "not replicable" I felt compelled to find something that more succinctly summed up the concept.
This, of course, was not the first time I'd been derailed by a language non-issue.
I don't know whether the internet is responsible for my malady or whether it just exacerbated it, but I've noticed a tendency to get distracted by such irrelevant word issues. Sure it might only be a few minutes, and the internet might be
just there, but I think I get irritated by the fact that I'm irritated about the issue as much
as the issue itself. And I don't even want to think about how I feel about that.
For example, whilst writing up some code comments describing some interactions between two parties communicating, I referred to the information that went to and fro between client and server. Except that due to the phrasing, I needed to use "toing and froing". Problem was, as I sat there at the keyboard I couldn't convince myself it was even a valid phrase, let alone settle on a spelling for it. So off to the internet I dived and came back with some reasonable level of confidence that it was a valid phrase (generally speaking) but torn between the inclusion of hyphens or the hyphen free version.
So you can guess what I did.
I consulted a friend.
I began by describing to him the general situation that the code covered, namely how we had some initial data going to and fro.
"So it's handshaking?" he interrupted.
Pause. Well, yes. And whilst at that point I realised that handshaking was indeed the term I should be using, it nevertheless felt like if I used it then I was merely going to be avoiding the issue. Nothing like missing out on a definite resolution to suck the joy out of writing code comments.
The toing and froing situation is reminiscent of the Periodic dilemma. In this case, what I had was behaviour that was decidedly *not* periodic. I searched around through various definitions - of random, of irregular, of non-repeating, of variable and vacillation - sure that somewhere there was a word in the English language that perfectly captured the kind of timing behaviour I was witnessing.
Lo and behold, the word "aperiodic" was hiding in plain sight. The twist in the tale is that as I wrote up a description of the situation I realised it wasn't all that aperiodic after all.
Whilst performing a hell of a lot of mathematical manipulation on some data once, I realised that there were some interdependencies involved that meant I couldn't just update values on the data set as I went along, or I'd skew the rest of the set. Working on a temporary copy was easy enough, but once again I was sidetracked by the description of the problem. How to best explain that I couldn't work on the data in-situ? I know what you're thinking - why not say you have to work on a temporary copy because of the data interdependencies? Rather as I did above?
Well, once the Latin expression of "in-situ" for "in place" had popped into my head, I became convinced that the solution was to find another Latin phrase meaning something roughly the opposite.
Do I know Latin? Of course not!
Cue crazy search through pages of 'common' Latin phrases. Only to find, once again, the obvious answer: "ex-situ". On a side note, I came across a number of very interesting Latin expressions that I thought might add a richness to my expression. Naturally, none have survived even in memory.
I won't even delve into the sad story of how splitting a hard drive's partition into three separate partitions resulted in my joyous discovery of the word "trifurcation". It gets brought up enough whenever people have to read the resulting document.
But to return to the opening point. I was after an antonym - yes, another, I swear they're the bane of my existence - for replicable and once again looked to the internet for a solution.
Turns out there's not really any such definitive word in English.
First, a breakdown in stats for several google searches:
- nonreplicable: 75 K
- unreplicable: 11 K
- irreplicable: 3 K
Of the three antonyms list above, none are present in merriam webster's dictionary.
However, there's some interesting discussion over at wordreference.
For a start, someone else did a google search and recorded the hits, so we have growth data. Here's the (alleged) results from 5 years ago (2006):
- unreplicable: 10.2 K
- irreplicable: 0.6 K
So there's been some minor increase in unreplicable (< 10%) but a vast increase in the use of irreplicable (~400%). However, both are still a long way behind nonreplicable.
In favour of irreplicable, one user suggests it follows the pattern of similar words, such as irreparable, irreplaceable, irrepressible.
Of course the counterargument would point to words such as unreproduceable, unrepealable and unrepeated. The last is rather interesting, as merriam webster recognises "unrepeated" but not "unrepeatable". How positively bizarre.
As to the general question of when to negate with "un-" and when to negate with "ir-", alas there seem to be no formal rules: that the current words have come into existence purely from usage, and the usage arises from the phonetics of the forms.
Finally, a note on the origin of several English prefixes:
un- : Old English
ne- : Old English
in-, im-, ir-, il- : Latin
di-, dis- : Latin
a-, an- : Greek
Addendum: I'm not bothered by the fact that when trying to find a title for a piece that's about looking for the perfect word for something in English, it was a French phrase that best captured the essence.
