Text this: Leveraging Organizational Hierarchy to Simplify Reward Design in Cooperative Multi-agent Reinforcement Learning