What symbol would I use if I wanted to express that, in the context of some binary relation $P$ implied from context, that $\exists (a,b)\in P: a\ne b$, but not to the extent that $\forall (a,b) \in P: a\ne b$.

The use of this would be if one were discussing a more restricted system, but then move to discussing a less restricted one. Like, "if we know for sure that $a\cdot b=b\cdot a$, then .... However, if $a\cdot b \mathrel{\rlap{=}\,?} b\cdot a$, then the previous reasoning doesn't apply, so ...". ("$\mathrel{\rlap{=}\,?}$" instead replaced with the real symbol)

  • $\begingroup$ Perhaps you mean $A \ne B$ $\endgroup$ – DanielV Nov 30 '13 at 0:23
  • $\begingroup$ I don't understand the question. The notation "$\exists a \in A\,\exists b \in B\, (a \ne b)$" does not imply $\neg(\exists a \in A\,\exists b \in B\, (a = b))$. So what is wrong with this notation for what you are trying to express? $\endgroup$ – Trevor Wilson Nov 30 '13 at 0:24
  • $\begingroup$ @DanielV what I was after would be something that wouldn't necessarily require explicitly stating what sets $a$ and $b$ are in. Edited to make it clearer. $\endgroup$ – AJMansfield Dec 1 '13 at 18:01
  • $\begingroup$ @TrevorWilson I was looking to see if there was some kind of shorthand for the statement. $\endgroup$ – AJMansfield Dec 1 '13 at 18:02
  • $\begingroup$ I believe you are trying to introduce something like modal logic. I've never seen a use for it, but if this subject interests you then you should look it up. I believe they also have a (compound?) symbol for what you are trying to describe. $\endgroup$ – DanielV Dec 1 '13 at 18:28

There is a lecture series on Digital Signal Processing available on Youtube, in which a symbol appears which quite elegantly states "not necessarily equal to" by subscripting the "not equals" sign with the letter n.

Symbol for "not necessarily equal to"

Screenshot source: Prof. S.C Dutta Roy, Department of Electrical Engineering, IIT Delhi


tl;dr: the formal notation for this is:$~~~~\neg\square(a=b)$


Modal logic formally defines the following dual operators:

  • Operator "$\square$" meaning "it is necessary", and
  • Operator "$\lozenge$" meaning "it is possible".

For any proposition P, the following are true:

  • $\square P \leftrightarrow \neg \lozenge \neg P~~~~~~~~$, i.e. : "P is necessarily true" is equivalent to "P cannot possibly be false"
  • $\lozenge P \leftrightarrow \neg \square \neg P~~~~~~~~$, i.e. : "P may be true" is equivalent to saying "P is not necessarily false"

Therefore, if you're happy to concede that your 'equality' is a logical statement, then you can express such statements formally as follows: $$ \lozenge(A = B) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\text{(i.e. $A$ can be a $B$)}$$ or $$ \lozenge(A \neq B) ~~~~~~~~~\text{(i.e. $A$ can be something other than $B$)}$$ depending where you want to place the emphasis.
Or if you really want to express it in terms of necessity: $$ \neg\square(a = b) ~~~~~~~~~\text{(i.e. it is not necessary that a = b)}$$ etc.

PS. I suppose, if you preferred a "one-symbol-only" binary operator, like your $\overset?=$, you could define in your article the symbols $\overset{\square}=$, $\overset{\lozenge}=$, $\overset{\square}\neq$, and $\overset{\lozenge}\neq$ respectively in terms of the modal operator syntax stated above, and I'm sure these would be straightforward to follow in your text.


You may find $\equiv$, used to denote $\forall x (f(x)=g(x))$, to be useful, eg.:

$$a\cdot b \not\equiv b\cdot a$$

For reference, see Identity (mathematics) on Wikipedia.


It might be easiest to write out "however, if $a$ does not necessarily equal $b$" or "however, if $a$ doesn't have to equal $b$" ... upon a quick Google search, there doesn't seem to be a clear symbol for what you need, and it doesn't take that long to write it out.

Given that you won't be using this phrase as often as you would "there exists" or "for all" or "if and only if", it seems unnecessary to have a separate symbol for it.


I think this is a good point, as this kind of sentence is quite common in mathematics. However, one such symbol does not exist yet. So you should propose a new symbol for it, with a bit of creativity. Your equal sign with a question mark on it is not too bad, but I'm sure you can do better!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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