Recently, one fast radio burst, FRB 200428, was detected from the Galactic magnetar SGR J1935+2154 during one X-ray burst. This suggests that magnetars can make FRBs. On the other hand, the majority of X-ray bursts from SGR J1935+2154 are not associated with FRBs. One possible reason for such rarity of FRB-SGR-burst associations is that the FRB emission is much more narrowly beamed than the SGR burst emission. If such an interpretation is correct, one would expect to detect radio bursts with viewing angles somewhat outside the narrow emission beam. These "slow" radio bursts (SRBs) would have broader widths and lower flux densities due to the smaller Doppler factor involved. We derive two "closure relations" to judge whether a long, less luminous radio burst could be an SRB. The 2.2-s, 308 Jy ms, ∼ 111 MHz radio burst detected from SGR J1935+2154 by the BSA LPI radio telescope may be such an SRB. The ∼ 2-ms, 60 mJy ms faint burst detected by FAST from the same source could be also an SRB if the corresponding FRB has a narrow spectrum. If the FRB beam is narrow, there should be many more SRBs than FRBs from Galactic magnetars. The lack of detection of abundant SRBs from magnetars would disfavor the hypothesis that all SGR bursts are associated with narrow-beam FRBs.