Abstract words refer to things and entities in the world that cannot be directly experienced through our senses (e.g., truth, morality). How we learn, represent, and use abstract words is one of the deepest problems in cognitive science today. I investigated this question in two experiments. In Experiment 1, I examined the effects of six semantic richness variables (sensory experience, valence, number of associates, context availability, arousal, semantic neighbourhood, and number of associates) on lexical-semantic processing for 207 abstract nouns. Behavioural tasks were lexical decision (LDT) and semantic categorization (SCT). Results showed that participants were faster to respond in LDT to words high in context availability, and faster to respond in SCT to words high in valence and sensory experience. These results suggest that abstract meanings might be grounded in our experiences with situational contexts and our bodily experiences. To further assess the extent of grounding for abstract word meanings (Experiment 2), I tested thirty-one pairs of undergraduate students in a variation of the Taboo task, in which participants communicated the meaning of the “secret” word (10 concrete, e.g., insect; 20 abstract nouns, e.g., impulse) to a partner in 1 minute or less. Analysis of verbal and gestural data yielded a number of important insights. Most notably, on a majority of trials, participants activated their partner’s concept of self or other (“You”, “Dalai Lama”), then placed that self into a situational context (“in class”) and then described the mental states, feelings, goals, etc. the self may experience. This, together with the results of my quantitative data analyses, suggests extensive recruitment of information derived from our experiences with the social and physical world and that such information is an important aspect of what we know about abstract words.