Earlier this year, Oregon football player LeGarrette Blount was suspended for the entire season for sucker punching an opposing player in the face. Given that this was his senior year, LeGarrette's college football career is over. This is particularly relevant because he is/was considered a decent professional prospect, and his lack of playing time this year could drastically impact scouts' impressions of his abilities.
Apparently LeGarrette has shown extensive remorse and a deep desire to return to the team. So much so that the head coach is considering reinstating him before season's end. From a football standpoint it makes perfect sense, as few doubt that he improves the team. It also gives him an opportunity to showcase his talents to pro scouts. From a psychology standpoint, however, this is a mistake. Why?
It is well-documented that reward systems are superior to punishment models for changing behaviors. If you want your child to do household chores, you are much more likely to get results by offering extra time with the Playstation than you are by hitting him or even withholding dinner. If you want better results on your next exam, you will increase the likelihood of that happening by buying yourself that new album you've been wanting for positive results than if you socially isolate as a penalty for poor performance. People generally respond better when there is something to be gained rather than to be lost.
There are times, however, that punishment is a necessary tool. In clinical settings, you'll see this employed when therapists work with autistic children. When the children are banging or screaming, clinicians will often look away and refuse to acknowledge the child's presence. This withholding of attention (punishment) can sometimes serve to engage in the child in more productive interactions.*
Punishment was required in the LeGarrette case. As bizarre as it sounds, you will statistically get better outcomes if you reward players for each game that they don't get into a physical altercation than you will for punishing those who do. Most scoff at the idea of rewarding what is considered a social norm (e.g., not hitting people) or the way people should behave. However, we've seen that just because society dictates behavior, doesn't mean it occurs.
So under the assumption that punishment is sometimes necessary for behavior modification, there are two ingredients to make it work effectively: immediacy and severity. The coach's action met both criteria. He didn't wait very long at all to lay down the suspension and, barring the athlete's removal from campus altogether, the punishment was pretty intense. We're told that the impact on his professional draft status from not playing could cost him millions of dollars in contract money. That would be enough to stop me from punching people, and I have that urge on an almost daily basis.
The problem is that the coach is considering not following through on what was an iron-clad ruling. There wasn't a "you can get out of this for good behavior" addendum on this punishment. It was fixed. And when you open the door on what you've claimed to be closed, you're begging for trouble.
Good shrinks invariably tell their clients that behavior modification requires consistency. If you need your child to stop screaming in the grocery store and threaten to take away his toy at home (punishment), but then don't do it, you're done. It's over. Your 3 year-old now owns you. Everything you say after that will be an empty threat until you consistently demonstrate otherwise. And if a toddler knows this, there's no doubt the athletic world does too.**
The coach's punishment, however, wasn't simply to change LeGarrette's behavior. In fact, the negative media attention alone might serve as enough punishment to prevent recidivism on this one player's part. No, part of the penalty was to create vicarious learning; not just at the University of Oregon, but throughout college athletics. We know that there were a boatload of players who are prone to violence on the field who said, "Out for the year? That's a long time. Millions of dollars? I better keep my cool." If LeGarrette comes back early, however, they'll know. They'll know that the message has changed from 'you will lose your rights to play this year,' to 'you may lose your rights to play this year.' And that small question mark, that bit of doubt, that one moment where someone says, "I could possibly get away with this," is when the behavior re-emerges.
There are plenty of people who are set free from punishment after demonstrating good behavior, sometimes with excellent results. But for maximum effect you want to make that possibility crystal clear before the official punishment is handed down. Behavior change works best when all parties are fully cognizant of the rules and all grey areas are eliminated. This didn't happen here, and because LeGarrette may not serve his full sentence, vicarious learning is compromised. This is not to say that suddenly we will see sucker punching running amok throughout college football. No, far from it. It's not a common offense anyway. But if the goal of the suspension was to impact other players from engaging in behaviors that even resemble what LeGarrette did, then from a statistical, scientific standpoint, the coach will have failed.
* There are strong arguments that punishment, while effective in altering certain behaviors, doesn't necessarily eliminate them altogether. It simply suppresses them in certain milieu and time frames. In the above example with autistic children, better results are obtained if rewards (e.g., a huge smile and praise when the child stops screaming) are paired with the punishment.
** Yes yes, I know one could make a joke here about a football player's intelligence level as compared to that of a child, but that's not very nice and you should be ashamed of yourself for even thinking it.