We developed particle identification (PID) in a CsI(Tl) crystal using digital pulse shape analysis. We present the details of the experiment, the data analysis, and the results. Four different methods of digital PID were applied to the same data set: (A) the rise-time inspection, (B) the sampling of the ADC waveform at the time of maximum waveform separation, (C) the charge comparison method with a rectangular integration window, and (D) the charge comparison method with a custom weight function. The performance of the methods A and B, which strongly depends on the bandwidth of the signal, is optimal when the bandwidth is reduced to about 0.5 MHz. The performance of the methods C and D can be only slightly improved by restricting the bandwidth. From among the four methods, the method A provides the poorest particle discrimination, while the method D provides the best. The algorithm D compares favorably with the analog charge comparison method, achieving very good proton/α-particle discrimination at energies as low as about 1 MeV, even though the measurements were carried out at the dynamic range of 35 MeV. Algorithms B, C, and D yield the PID index which is almost independent of particle energy, within the energy limits used in this work. Present results clearly show the power of digital electronics in achieving good particle identification for charged particles measured with large dynamic range.