Are the magnetic fields of millisecond pulsars ~ 10^8 G?

It is generally assumed that the magnetic fields of millisecond pulsars (MSPs) are $\sim 10^{8}$G. We argue that this may not be true and the fields may be appreciably greater. We present six evidences for this: (1) The $\sim 10^{8}$ G field estimate is based on magnetic dipole emission losses which is shown to be questionable; (2) The MSPs in low mass X-ray binaries (LMXBs) are claimed to have $< 10^{11}$ G on the basis of a Rayleygh-Taylor instability accretion argument. We show that the accretion argument is questionable and the upper limit $10^{11}$ G may be much higher; (3) Low magnetic field neutron stars have difficulty being produced in LMXBs; (4) MSPs may still be accreting indicating a much higher magnetic field; (5) The data that predict $\sim 10^{8}$ G for MSPs also predict ages on the order of, and greater than, ten billion years, which is much greater than normal pulsars. If the predicted ages are wrong, most likely the predicted $\sim 10^{8}$ G fields of MSPs are wrong; (6) When magnetic fields are measured directly with cyclotron lines in X-ray binaries, fields $\gg 10^{8}$ G are indicated. Other scenarios should be investigated. One such scenario is the following. Over 85% of MSPs are confirmed members of a binary. It is possible that all MSPs are in large separation binaries having magnetic fields $> 10^{8}$ G with their magnetic dipole emission being balanced by low level accretion from their companions.