I've not seen that video before and it's a beauty. Here Feynman lives up to his reputation as a showman, as that has to the greatest non-answer of all time to that question. It would be interesting to know how much physics the interviewer actually knew, as it's the same question I ask professional physicists whenever I get a chance.<p>The trouble with that question is it's just about simplest, most obvious question a layperson can ask yet it is one of the most profound and difficult in physics to answer at any level. QED gives us some wonderfully useful answers, we've the Master himself along with Schwinger, and Tomonaga to thank for that but it still doesn't tell me <i>exactly</i> what that field actually is to my satisfaction.<p>The matter of static fields, potentials and virtual particles is a vexed and complicated issue and one doesn't have to look much deeper to be at the cutting edge. For instance, I'm still waiting for some genius to give me a full explanation of the Aharonov-Bohm effect (here, it seems we've everything in the mix—charged particles, phase shift, zero electric field, the potentials/fields argument that Feynman mentions, even 'locality' is important—yet it's still not fully resolved to everyone's satisfaction).<p>I would have loved to hear Feynman's answer if the interviewer had raised the candle and the fridge magnet question, as it's one rung up from the magnetic field question. To remind you, <i>'why doesn't the fridge magnet's magnetic 'battery' go flat and it can stay indefinitely where it's put, whereas a candle radiates energy and after some hours it runs out?'</i> It's obvious the candle radiates energy away at speed <i>c</i> and soon there's no energy left, but we're also told magnetic fields also radiate at speed <i>c</i> so what gives?<p>Right, now we're at the crux of the problem—in the thick of it so to speak. We really cannot answer this question without resorting to the depths of magnetostatics—QED and all that that encompasses, near and far fields, static magnetic fields, potentials, virtual particles, permeability, etc. It even raises the question of why <i>c</i>, vacuum permeability, μ, vacuum permittivity, ε, are the values that they are and no one yet knows.<p>Incidentally, I recall a while ago in the pages of <i>New Scientist</i> a similar question being asked of physicist Frank Wilczek about what actually <i>is</i> a field. His reply was to the effect that it's a 'type of matrix'. At the time, I thought surely he could do better than that. The fact is that at the most fundamental level it's not entirely clear.