American Scientist

To the Editors:

As an aspiring science-fiction writer, I am always on the lookout for novel science that may presage new technology to incorporate into my stories. The apparatus depicted in Figure 6 of the July-August article "Quantum Erasure" by Stephen P. Walborn et al. appears, with a few simple modifications, to enable Bob to send digital information to Alice at faster-than-light speeds. If Bob could switch between a horizontal and digital polarizer, and Alice could distinguish a fringe from a "blob"—and they had accurate, synchronized clocks—then they ought to be able to send messages using a series of dots and dashes. Since the vast majority of scientists currently believe that superluminal communication is impossible, I wonder if the authors would care to comment on this scheme?

Stuart H. Smith, Jr.
New Orleans, Louisiana

Stephen Walborn replies:

Stuart H. Smith Jr.'s faster-than-light communication scheme based on the quantum erasure experiment was one of several similar emails we received. Unfortunately, none of these schemes will work. Each of Bob's photons has a 50 percent chance of being detected with positive diagonal polarization and a 50 percent chance of being detected with negative diagonal polarization. In the quantum eraser experiment, positive and negative diagonal directions correspond to interference fringes and antifringes, respectively. Alice observes the sum of the two interference patterns—a large "blob" pattern of total photon counts (shown as a black line in Figure 7). Regardless of what Bob does with his photons, Alice will only detect a "blob."

The only way that Alice can observe an interference pattern is through post-measurement "bookkeeping," that is, using Bob’s measurements to separate her own results into subsets that show interference, as in the dialogue at the end of the article. Of course, Alice and Bob do not need to meet face-to-face, but something like a "phone call" is necessary. Since we cannot do this step faster than light, the quantum erasure process will not be superluminal. However, a similar device (supplied with a "phone line") does function as a secure cryptographic system (see Ekert, A. K. 1991. Physical Review Letters 67:661).