Someone said to me, "X is potentially faster than Y". Without any clarification at that point, I immediately assumed that the speaker thought that X was at least not slower than Y.

It was revealed in a further discussion, however, that the speaker did not know much about the current performance of either X or Y. What they meant was that, if both were given equal opportunities for optimisation, X would certainly end up faster (and, probably, that it might already be the case).

I agree that "potentially so and so" can certainly translate that way, i.e. something merely has the potential to be so and so. My question, however, is about an implication of "potentially" when it is used in a comparison.

So, when I hear, "X is potentially faster than Y", is it normal, without more information on the subject, for me to assume that X is at least not slower or should I wait for more information before jumping to a conclusion? Does it depend on the situation (e.g. heard in a conversation vs read in a technical article)?

  • 12
    In a word, no. All it says is that there is some conceivable set of circumstances where X might be faster than Y. "Potentially" is not a word of comparison. – Hot Licks Apr 16 '15 at 11:40
  • 2
    I am very familiar with arguments like "computer language X is faster than computer language Y", so I was writing from that viewpoint. – Hot Licks Apr 16 '15 at 12:07
  • 1
    In the case that you describe I would phrase that as "X has the potential to be faster than Y" – Martin Smith Apr 16 '15 at 12:17
  • 2
    Andriy M, it seems you feel the current answers have overlooked something? I don't think there is any difference in meaning between "X is potentially faster than Y" and "X has the potential to be faster than Y". No matter which way you phrase it, I don't see how you can interpret it to mean that slower is impossible or unlikely. That's simply not what it means, and it's logically a distinct concept. – herisson Apr 16 '15 at 13:43
  • 4
    If "slower" were impossible or unlikely, the person would probably say one of the following: "X is faster or at least the same speed as Y,""X is potentially faster than Y, and always at least as fast," or "X is always at least the same speed as Y, and potentially faster." – herisson Apr 16 '15 at 13:47

No. The implication I get from "X is potentially faster than Y" is either:

  1. In some circumstances we know about (for example, in the circumstance that the code is optimized for both X and Y), X is faster. In other circumstances we know about, it is the same speed, or slower.
  2. We don't know in which circumstances X is faster; for any particular case, it could be faster, or it could be the same speed or slower.

In neither case is there an implication that X is never slower than Y.

| improve this answer | |
  • 6
    I would say the actual claim is more like "We do not know that it is not possible for X to be faster than Y." – Sled Apr 16 '15 at 14:53
  • 10
    @ArtB There's an implication that we at least suspect X is faster in at least some situations. – cpast Apr 16 '15 at 20:44
  • Since the "someone" in the question is referring to me, I'd thought I'd chipped in: Yes, the first point would be accurate, re: "In some circumstances we know about (for example, in the circumstance that the code is optimized for both X and Y), X is faster. In other circumstances we know about, it is the same speed, or slower." – Pacerier Apr 18 '15 at 12:44

No. A potential state can be quite remote and independent from a current state.

| improve this answer | |

No. "is potentially faster" makes a restricting statement on the speed, namely that it "is [not [definitely [not faster]]]". In other words, it is possible that it is faster.

No further restrictions to the speed are made. Thus, it might or might not have the same or a lower speed.

| improve this answer | |

A lot of great answers but I like simple answers so I just wanted to say that "potentially faster" is not mutually exclusive with "potentially slower"

| improve this answer | |
  • This is just summing up what other answers are stating. Probably needs to be a comment under the question instead. – Nicole Apr 16 '15 at 20:48
  • @Nicole This answer uses a simple logical argument, providing a counter state and asserting that the two states do not exclude each other from existence. As such it is a distinct answer from the others currently posted. – Justin Ohms Apr 16 '15 at 21:13

No, not at all. A Ford Edsel is potentially faster than a Ferrari, if the Ferrari's gearbox is blown.

'potentially' is a weasel-word up there with 'theoretically' or 'hypothetically'. Especially beware when wielded by marketing people.

| improve this answer | |
  • 2
    It's not a weasel-word when the circumstances have been clearly stated. It's only a weasel-word when the circumstances are vague or not mentioned at all. – Pacerier Apr 18 '15 at 12:44
  • @Pacerier true, but I often see an unqualified 'potentially'. Especially as used by marketing people. – smci Apr 18 '15 at 20:06
  • 1
    Let's not let marketing people define our language. Its Latin roots mean power and capability not certainty. Words are not weasels speakers can be. – John May 13 '15 at 4:15

No. The words potential and potentially are used with overlapping but possibly contradictory meanings (bolding mine):

potential adj.

  1. Capable of being but not yet in existence; latent or undeveloped: a potential problem; a substance with many potential uses.



Adv. 1. potentially - with a possibility of becoming actual; "he is potentially dangerous"; "potentially useful"

[Based on WordNet 3.0, Farlex clipart collection. © 2003-2012 Princeton University, Farlex Inc.] [same link]

So 'we know it is possible to develop it' vs 'it may be possible to develop it'.

The speaker needs to be encouraged to estimate the probability X will be faster than Y by 2020 say.

The only logical thing to infer (assuming the speaker is not being totally disingenuous, and is not misinformed) is that, at the moment, X is not faster than Y. It could well be slower.

| improve this answer | |
  • "With a possibility of becoming actual" – well, that seems close, I'm just not sure what "becoming" means in this case. I would say that my interpretation was something like "with a possibility/probability to be confirmed as such". (That's how I would read "he is potentially dangerous" too, by the way.) – Andriy M Apr 16 '15 at 11:43

In an engineering context, "is potentially faster" would suggest that something is likely to be faster, but by an amount that may or may not be significant. It does not imply certainty that the thing will not be slower, but does imply a belief that it is unlikely to be significantly slower.

In things like computer science, there are many situations where performing an operation on a some number of objects will take a certain amount of time to set up the operation and then a certain amount of time per object. Often a willingness to accept a longer set-up time will allow one to reduce the per-object time [which may be helpful when processing many objects, and counter-productive when processing a small number]. To my ear, "Is potentially faster" might typically be used to describe a change that slightly increases setup time and reduces per-item time; the reduction in per-object time might or might not be sufficient to make up for the increased setup time, but even if it doesn't make up for the setup time the shortfall won't generally be significant.

| improve this answer | |
  • 1
    I agree with what you say about the implications of someone having said that. If there was significant chance of being significantly slower, whoever was talking would mention it, not just say "is potentially faster", unless it was already clear from context that the change had large potential downsides for some cases. I don't think most people would say "is potentially faster" without any other discussion of performance if they thought it would be noticeably slower in any common cases. – Peter Cordes Apr 16 '15 at 19:09
  • 1
    IDK why this got downvoted. It's the answer that says more about actual usage, and not just what the statement logically implies in a logic-puzzle sense. – Peter Cordes Apr 16 '15 at 19:10

I agree with all the posts here, but I also think it's reasonable to infer that the speaker intends you to believe that X will not be slower than Y.

Lawyers would argue that the statement is without meaning. While X has the potential to be faster than Y, it may also have potential to be a marshmallow. There is no reliable literal interpretation. The spirit of the statement however, is that, while X is not necessarily faster than Y, it certainly is not slower than Y.

| improve this answer | |
  • The accepted answer contradicts this one, and I think the accepted answer is right. – Ben Millwood Apr 20 '15 at 13:06
  • 1
    If you are making a comment, would you like to comment why you disagree? Not only do I not contradict the accepted answer, but I suggest that a speech act requires the participation of two people. In the absence of speaker clarity, the listener will make natural inferences. Moreover, if language is used in an idiomatic context, a meaning can and should be inferred, speaking would be too cumbersome otherwise. The accepted answer states merely that the absence of clarity devoids the statement of meaning. This is an oversimplification. Right and wrong are clear in maths, not in language. – John Apr 20 '15 at 19:16

"X is potentially faster than Y"

This statement, alone, is meaningless.

What logical inference can be drawn from such a barren premise?

The end. Why all the cud-chewimg over such a lightweight topic? Save your strength...

dbmllc a:a

| improve this answer | |
  • I've got to agree with you. – Hot Licks Apr 17 '15 at 3:28
  • I can't agree. In some domains, corner cases matter greatly (consider analysis of real-time performance guarantees, or correctness guarantees in domains where potential race conditions exist), and identifying whether such corner cases can exist is critical. – Charles Duffy Apr 17 '15 at 15:41
  • There are several good answers right up above this one that give a good and sensible meaning to the statement you provide, so this answer is wrong. – Ben Millwood Apr 20 '15 at 13:05
  • I can assure you that logical inferences can and have been drawn from premises of this form, and they are not at all barren. In formal logic, the entire sub-field of modal logic is concerned with statements containing quantifiers like possibly, probably, and necessarily. Consider, for example, that from X is potentially faster than Y we can infer It is not the case that X is necessarily faster than Y. These are distinctions with practical applications in many fields, as already mentioned. – dodgethesteamroller Apr 22 '15 at 23:55

Not the answer you're looking for? Browse other questions tagged or ask your own question.