However, management types can discourage you from doing this as it is an acceptance that there is some liability for whatever it is that has gone wrong. There is a fear of unintended consequences from the instinctive apology.
The same applies with a road traffic accident - to apologise to the other party might be what you want to do (even if you were not entirely at fault) however the insurance company will not be best pleased.
As a consequence we are conditioned to only apologise under duress or for things that are inconsequential. Maybe it is this that affects how we react when someone randomly apologises - it is quite an unusual gift worthy of reciprocation.
Just letting someone know that you understand that they're stressed and annoyed or frustrated, and that you'll try to help them makes a big difference. At least, it does when people do it with me.
To many, apologizing has become a automatic reaction and they now apologize for everything. They don't say the person is thinking of others, nonsensical apologies mean that the speaker isn't thinking anything about you and just wants to get on with their day.
You did it in the next paragraph. You're not sorry, you probably don't give a damn about them, you just need to move on to the next person.
I recommend the book "Everything Is Obvious(^1): How Common Sense Fails Us"
^1 Once you know the answer.
It talks about the human tendency of arranging observations and beliefs to fit what we're told, solidifying our belief that this is just common sense. I had just read it so apologies if I shoehorn off of your post to reference it.
It also paints a very negative brush about most sociology/behavior type experiments, making me immediately skeptical of this one. I wonder if it's reproducible, and what specific regional cultural issues play into it.
As a personal note, I get very wary of anyone who apologies for things outside of their control.
From a fellow reader of Everything is Obvious, same here. It seems obvious (heh) to me that it's not the _apology_ that caused an increase in trust, but the addition of neutral or better contact.
In all of the experiments, the control group did one action (e.g. asked for a cell phone or played their move) and the test group did something over and above that action which was a neutral or better action (a superfluous apology).
I see no reason to think that the same results could not be reproduced if the control said "you have nice shoes" or "that was a nice move". That is, the conclusion should instead be that an additional amount of neutral or better contact with a person causes them to trust you more than they would without the statement--the contact does not have to necessarily be an apology.
In fact, the article admits as much in second to last paragraph:
>How trustworthy are these results? The accumulated findings from several experiments help build a convincing case, but unfortunately the field study - which had the potential to provide the most persuasive evidence - is seriously flawed. The actor apologised (sic) for the rain then asked to borrow a phone, or in the comparison condition he just asked to borrow the phone. There was no proper control condition. This means we don't know if the impact of the apology was specific to making an apology or merely an effect of uttering any kind of ice-breaker.
There's a rule I have followed for many years when doing customer service - if the person is upset, apologize. It doesn't matter if you can't give them what they're asking for, or if they made a silly mistake. Just - make them feel better, then do what you sensibly can. It really works well... I don't get many angry complaints anyway, but this tends to defuse them immediately.