Key message: Metabarcoding analysis of soil fungal communities in French mountain forests revealed that harvesting intensity, time since last harvest and former land use had no effect on fungal community composition compared to key abiotic factors. Low-intensity management in these uneven-aged mountain forests therefore has limited effects on soil fungal community composition which is mainly driven by elevation and edaphic properties. Context: Past and current human activities are known to affect forest biodiversity. However, the effects of former land use and forest management have been studied much more extensively on higher plants than on fungi. Aims: Our objectives were to assess the effects of harvesting intensity, duration since last harvest and former land use on soil fungal communities in uneven-aged mountain high forests. Methods: On the basis of historical land-use maps drawn between 1862 and 1864 and on historical forest management archives, we selected 62 sites in the French Alps with contrasting land-use histories (ancient forests, which were already forested on historical maps vs recent forests, which have recovered following abandonment of pastures) and different durations since last harvest (from 1 to over 50 years). We carried out soil sampling and assessed fungal diversity by metabarcoding analysis. We analysed soil fungal molecular operational taxonomic units (MOTU) diversity as a whole and for the main lifestyle groups (such as wood saprotrophic or ectomycorrhizal fungi) using multiple linear regressions on Shannon’s diversity index and fungal taxonomic composition using canonical correlation analysis. Results: We found no significant effect of harvesting intensity, time since last harvest or land-use history on total fungal MOTU diversity, fungal lifestyle diversity or taxonomic composition. In contrast, we observed significant effects of elevation, pH, organic carbon and available phosphorus content on the taxonomic and functional composition of soil fungal communities. Conclusions: The structure of soil fungal communities (i.e. diversity and species composition) was mainly determined by elevation and edaphic factors, indicating a high-context dependency, as previously found in similar studies. Our study in mountain forests shows that recent forests established on former pastures had no legacy effect on soil conditions and fungal communities, in contrast to previous results in lowland areas, where recent forests were mainly established on former cropland. Uneven-aged forest management had no effect on fungal diversity, in contrast to previous results observed in even-aged high forests.