The oxidation of formic acid and carbon monoxide was studied at a gold electrode by a combination of electrochemistry, in situ surface-enhanced Raman spectroscopy (SERS), differential electrochemical mass spectrometry, and first-principles DFT calculations. Comparison of the SERS results and the (field-dependent) DFT calculations strongly suggests that the relevant surface-bonded intermediate during oxidation of formic acid on gold is formate HCOO. Formate reacts to form carbon dioxide via two pathways: at low potentials, with a nearby water to produce carbon dioxide and a hydronium ion; at higher potentials, with surface-bonded hydroxyl (or oxide) to give carbon dioxide and water. In the former pathway, the rate-determining step is probably related to the reaction of surface-bonded formate with water, as measurements of the reaction order imply a surface almost completely saturated with adsorbate. The potential dependence of the rate of the low-potential pathway is presumably governed by the potential dependence of formate coverage. There is no evidence for CO formation on gold during oxidation of formic acid. The oxidation of carbon monoxide must involve the carboxyhydroxyl intermediate, but SERS measurements do not reveal this intermediate during CO oxidation, most likely because of its low surface coverage, as it is formed after the rate-determining step. Based on inconclusive spectroscopic evidence for the formation of surface-bonded OH at potentials substantially below the surface oxidation region, the question whether surface-bonded carbon monoxide reacts with surface hydroxyl or with water to form carboxyhydroxyl and carbon dioxide remains open. The SERS measurements show the existence of both atop and bridge-bonded CO on gold from two distinguishable low-frequency modes that agree very well with DFT calculations.