This study addressed long-term land degradation and regeneration effects on soil organic carbon (SOC) composition. This was done in a context of secondary succession following land abandonment in the Mediterranean region of SE Spain. The effects of land use change and soil erosion on SOC composition were studied by using lignin as a biomarker. To get insight into the evolution of SOC composition along a land use and topographical gradient, differences in lignin contribution to SOC were determined at different soil depths (0–0.1 m, 0.1–0.2 m, 0.2–0.3 m). Three deposition locations, three positions on the hillslope and three zones on top of the hillslope (shoulder) were selected on croplands and fields that were abandoned since 10 and 50 years, respectively. Land use change was identified as a driver for the observed gradients in lignin, SOC and N in these semi-arid ecosystems. Abandoned sites were highest in soil lignin, which could be related to the higher lignin input. For deposition and shoulder positions lignin was less degraded at abandoned sites compared to cultivated sites. On croplands lignin was more degraded at hillslope locations compared to deposition zones. Observed differences in soil lignin quantity were highest for the topsoil (0–0.1 m). For deeper soil, differences are less pronounced. However, no differences were evident for lignin contribution to SOC (mg lignin phenols/g SOC). As modern soil erosion rates are very low in the study area, SOC composition on abandoned fields may be more influenced by present day vegetation and its degradation than by soil erosion. Surprisingly, lignin contribution to SOC was not favored by vegetation recovery either. The higher soil N contents for abandoned fields compared to croplands might explain why lignin is not preferentially preserved on recovered sites.