Is SGR 1900+14 a magnetar?

We present Rossi X-Ray Timing Explorer observations of the soft gamma-ray repeater SGR 1900+14 taken 1996 September 4-18, nearly 2 yr before the 1998 active period of the source. The pulsar period (P) of 5.1558199 +/- 0.0000029 s and period derivative ((P) over dot) of (6.0 +/-: 1.0) x 10(-11) s s(-1) measured during the 2 week observation are consistent with the mean(P) over dot of(6.126 +/- 0.006) x 10(-11) s s(-1) over the time up to the commencement of the active period. This (P) over dot is less than half that of (12.77 +/- 0.01) x 10(-11) s s(-1) observed during and after the active period. If magnetic dipole radiation were the primary cause of the pulsar spin-down, the implied neutron star magnetic field would exceed the critical field of approximate to 4.4 x 10(13) G by more than an order of magnitude, and such field estimates for this and other soft gamma repeaters (SGRs) have been offered as evidence that the SGRs are magnetars, in which the neutron star magnetic energy exceeds the rotational energy. The observed doubling of (P) over dot, however, would suggest that the pulsar magnetic field energy increased by more than 100% as the source entered an active phase, which seems very hard to reconcile with models in which the SGR bursts are powered by the release of magnetic energy. Because of this, we suggest that the spin-down of SGR 1900+14 is not driven by magnetic dipole radiation, but by some other process, most likely a relativistic wind. The (P) over dot, therefore, does not provide a measure of the pulsar magnetic field strength, nor evidence for a magnetar.