BYU, nope not confused; a dipole only needs a balun if their is a chance of extrainious mutual coupling with feedline from other antennas or something nearby...
...A dipole is probaly the only antenna out there that when fed by an unbalanced feed like coax will not exhibit unwanted stuff due to the natural closeness in ohms to the feedline.
Dear me, more confused than I had thought.
Let me suggest you read W7EL's article on Baluns: What They Do and How They Do It
Common mode current has nothing to do with reflected power or the SWR on the feedline.
The currents flowing INSIDE the coax are totally independent from those flowing OUTSIDE on the
shield - the latter are what we call common mode current, and the purpose of a balun is to
reduce these currents.
If you take a theoretical dipole and connect a long wire to one side of the feedpoint that
drops down to ground level, the dipole is no longer balanced, and that added wire will have
some amount of current flow that depends on the length of the wire.
This is exactly the situation you have when you feed a dipole with coax cable: the outside
of the coax acts like the added long wire connected to one side of the dipole, unless you use
a balun to decouple it.
The comments about grounding have to do with whether or not the end of this added wire is
grounded - since it is part of the antenna, that will change the current distributions.
A perfectly symmetric dipole certainly can have common mode current (and no longer be
symmetric) once a coax feeder is connected. This has happened to me several times, where
the SWR changes when I rearrange the extra coax, or the SWR measured with an SWR analyzer
doesn't match that with the rig connected (due to the added length of the power cable being
added to the antenna), etc. As I tried to make it clear above, that doesn't mean that such a
dipole can't work, because "work" is a relative term. I don't use baluns in my wire dipoles for
the reasons you suggest, but I also know how to recognize and deal with the problems that
I encounter due to not doing so. Often the coax length is such that little common mode current
flows and the operator doesn't notice it, but a seemingly insignificant change in the station
configuration can change that.
In the immortal words of Mark Twain, "It ain't what you don't know that gets you into trouble.
It's what you know for sure that just ain't so.