The mirror that I'm talking of, is situated in Hazarduari Palace in Murshidabad, West Bengal, India. This place was once the capital of Bengal under the Nawab before the British defeated the Nawab in the historic battle of Plassey, 1757.

The history is not that important here, but the mirror is. In fact, this is not a single mirror, but a set of mirrors that adorn the so-called "dressing table" of the Nawab. The importance of this set of mirrors is that, when you look at it, you can see everyone around yourself, but not yourself.

When we had visited this historic place, we also got a chance to see this famous mirror set. Whether you look at it from a distance or very close, you'll never be able to see yourself, but you will clearly be able to see everyone standing next to you.

As it was a museum, photography was prohibited there, due to which I cannot show a photo of the mirror. I searched on Google but couldn't find a photo, perhaps due to the same reasons. So you've to believe my words that something like that really exists.

Now, my question is, can anyone show or explain how such a wonder can be created with plane mirrors? I've never seen it elsewhere. How can you not see yourself but everyone around you?

  • $\begingroup$ Some more details would be essential, some drawing maybe... And how much are you restricted in angle at which you look at that mirror set? $\endgroup$ – OON Nov 9 '17 at 21:26
  • $\begingroup$ The wikipedia article says : There are also two pairs of mirrors in the museum, that are placed at an angle of 90 degrees in such a way that one cannot see one's own face but others can see. $\endgroup$ – sammy gerbil Nov 9 '17 at 22:00
  • $\begingroup$ @sammygerbil did not serve that. I only gave the article first reference for those who don't know about the place. Of course I didn't see it. $\endgroup$ – Wrichik Basu Nov 10 '17 at 4:24
  • $\begingroup$ Wikipedia says for TWO PAIRS OF MIRRORS that is FOUR MIRRORS and not two mirrors as in the till now given answers. $\endgroup$ – Frobenius Dec 31 '18 at 13:06

It's hard to say how they did it based on the fact that you don't provide any information about the geometry of the mirror planes, but I can offer a possibility:

A retroreflector is a configuration of mirror planes such that a ray of light reflecting off of the mirrors always comes out in exactly the opposite direction (parallel). This can be achieved with mirrors in a "corner cube" geometry but, to a lesser extent, the effect can be seen with two vertical mirrors placed at right angles to each other. If you've ever seen this, then you will have noticed that your reflection will always appear at the corner where the two mirrors meet.

So, my answer to your question is that if you were to place two mirrors at right angles to each other but then remove the parts of the mirrors near the corner (or perhaps obstruct it with a column or something) then you will never see your reflection in the mirrors, but you will be able to see the reflections of the space on either side of you. This is only true if you are not "inside" the retroreflector volume (in which case you will be able to see your reflection in both mirrors individually).


The comment by sammy gerbil and the answer by Andy Barrette already hinted at this - but this diagram may explain (this is speculation as we have no pictures of the actual setup):

enter image description here

The person in blue cannot see his/her own reflection (red line goes through the gap between the mirrors) unless they look 45 degrees to the left or right; however, they can see the reflection of people standing next to them...

Note that unlike with a conventional mirror, the person standing to your right will appear in the left hand part of the mirror image, and vice versa.

  • $\begingroup$ ..but the person in blue could see himself looking in a direction normal to any one of the mirrors... $\endgroup$ – Frobenius Dec 31 '18 at 13:08
  • $\begingroup$ @Frobenius if the mirrors are far enough away, and you add some obstacles that block the “normal incidence “ view (imagine these mirrors are some distance away and you are looking “through a window”), this problem can be overcome. $\endgroup$ – Floris Dec 31 '18 at 13:11
  • $\begingroup$ See also the last line in the other answer... $\endgroup$ – Floris Dec 31 '18 at 13:12
  • $\begingroup$ I think that this could be overcome dragging the mirrors parallel to their plane so that their upper ends coincide at least with the feet of the normals. By the way, how many mirrors are used ??? 4 or 2 ??? $\endgroup$ – Frobenius Dec 31 '18 at 13:18
  • $\begingroup$ I think this works with just two mirrors. Although you will see a left/right anomaly (what you see won’t be the usual “mirror image”. You could add another mirror to do the LR flip.) $\endgroup$ – Floris Dec 31 '18 at 13:20

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.