Introduction: This study evaluated nerve recovery through retrospective study of patients with lingual nerve damage. Patients and Methods: The patients who visited Seoul National University Dental Hospital for an injury to the lingual nerve from April 1988 to August 2009 were enrolled in this study (n=41). The relevance of various factors including the causes of damage, age, etc. was analyzed by the subjective improvement based upon questionnaires and the clinical records. The evaluation variants were a subjective assessment and neurosensory examination composed of the direction, contact threshold, two-point discrimination, pin prick, thermal discrimination and current perception threshold. Results: The causes of lingual nerve damage were an extraction of the lower third molar (75.6%), local anesthesia (9.7%), incision and drainage (4.88%), trauma (2.44%). The evaluation of subjective prognosis exhibited no difference in sensory improvement depending on the cause, age and gender. Based upon the subjective evaluation, 44.7% of patients showed sensory improvement. The first hospital visit from injury was shorter in the group showing subjective improvement (3.41 months) than those showing no improvement (5.24 months) (P=0.301). Thirty six out of 41 patients were treated with only conservative therapy and 5 patients were treated by surgical intervention. Neurosensory examinations revealed improvement, although not statistically significant, and the degree was higher in the subjectively improved group. The contact threshold discrimination showed the highest correlation with subjective improvement (P=0.069). Most of the sensory recovery was gained within 12 months and the degree of improvement at the tip of the tongue was higher than that of the dorsum (P<0.001). Conclusion: The damaged lingual nerve improved at a rate of 44.7% and mostly within 12 months after the incident. There was no difference between the subjective prognosis and neurosensory examination depending on the cause of damage, age and gender, whereas the contact threshold discrimination was the best variant that reflected the subjective prognosis statistically.