4 ms·

I know groupthink has a negative connotation and can be damaging but on the other hand it seems necessary to have some kind of groupthink in order to align an o

I know groupthink has a negative connotation and can be damaging but on the other hand it seems necessary to have some kind of groupthink in order to align an organization on a goal. Especially in areas where there is not a clear cut decision what’s best you have to form a group opinion that assumes certain things. For example in software dev at some point you have to decide on an architecture and live with it even if there are alternatives that are as good. People who constantly reopen that question are not really helpful at some point. The danger is probably for people to stick to that decision for too long even after weaknesses start showing up. But I think especially for a large organization it’s hard to find the point when it’s time to change course. I guess all I want to say is that damaging groupthink is easy to spot in hindsight but that well functioning organizations also rely on groupthink. And that it’s quite hard to distinguish between positive and necessary groupthink and negative groupthink.

ben_w 7y ago
Are you sure that’s groupthink that you’re describing? Groupthink isn’t mere agreement, but specifically agreement contrary to the professional (not personal) options of informed experts. “It’s not my first choice, but it will do” isn’t, I think, in that category.
- Ididntdothis 7y ago
  I think often agreement gets relabeled as groupthink later.
- asdf21 7y ago
  What about when all the informed experts are wrong? Like consuming dietary cholesterol causing high blood cholesterol? Or trans fats being healthier than saturated fats? Or tariffs harming the US economy?
  - lopmotr 7y ago
    How can your group know something that informed experts don't? If you have special knowledge they don't, then they aren't informed. Your group might knowingly take a risk in guessing the experts are wrong, perhaps to test a novel idea, but that's not the same as refusing to recognize their opinion as valid, like in groupthink.
    - asdf21 7y ago
      "Experts" are especially prone to groupthink, this has been shown numerous times. People can be "informed" about something and still need to shut-up about it, when it's politically incorrect / represents a unpopular view, when it puts their income / reputation at risk, etc. Experts have more to lose, so they are especially vulnerable to this sort of pressure.
      - lopmotr 7y ago
        Of course experts can be wrong, but you won't know that they wrong or how in any particular instance, so you won't be justified in agreeing on something contrary and silencing dissenters who bring up the expert opinion. I'm just talking about obvious reasonable behavior here. For instance, experts probably agree that nuclear radiation causes cancer. It would not be sensible to believe that nuclear radiation doesn't cause cancer and then expose people to that radiation, believing it's harmless just because experts are at risk of being corrupted.
- Nasrudith 7y ago
  I think it is more "obviously wrong in face of evidence and refuse to reconsider" than informed experts per se. Doctors being stuck up about handwashing because they heuristically interpretted "a gentleman's hands are alway clean" as a descriptive instead of a directive was certainly groupthink. It is possible to be wrong and not be engaged in groupthink or right and lacking logical basis.
Nasrudith 7y ago
I believe that is cohesion which isn't inherently linked to group think but attempts to promote may lead to it. To use a cliche combat context it is like loyalty effectively. There may be questions of whether to attack from the woods or an open field with tank support and working together is more important than one or another as even the situational "wrong" choice is better than a split effort. However absolute blind loyalty ends only in needless tears from trying to shoot tanks with pistols, charging the no man's land into machine guns, massacre of villages, or military coups. The dangerous aspect is the "delusion" in both.
- bilbo0s 7y ago
  Military is a bad example. I can tell you from experience, it's not a place that runs on "groupthink", or "consensus" views. A decision is made by a decision maker, and that decision is implemented. In the military they even have the "mind guards" that "insulate leadership from criticism" that were mentioned in another comment. Only in the military, we call them "NCO's". You can be diametrically opposed to a decision in the military if you like. Heck, you can think whatever you want in the military, because no one cares what you think.
Jtsummers 7y ago
Consensus and common goals aren't groupthink. Groupthink is specifically about when people stop thinking individually, and take the group's position to be correct-by-default. Consensus and common goals can be achieved without that assumption. Groupthink, in software architecture, happens when we can no longer question the basis for a design or assumption. Perhaps the design made perfect sense once, but now we're trying to scale to 1000+ servers, not just 5-10. Communication overhead is killing us, groupthink promotes (for too long, ultimately to the detriment of the business and its customers) the idea that the architecture is beyond question. Consensus and common goals are different. If the architecture is limiting the business, then that can be an agreed upon idea, but not actionable yet. Maybe we realize that other things need to be addressed first (our build system, CI/CD pipelines, testing systems) in order to make architecture updates even feasible. Groupthink would deny even addressing those concerns because they still won't accept that the architecture is limiting the business.

ben_w 7y ago

Are you sure that’s groupthink that you’re describing? Groupthink isn’t mere agreement, but specifically agreement contrary to the professional (not personal) options of informed experts. “It’s not my first choice, but it will do” isn’t, I think, in that category.

Ididntdothis 7y ago
I think often agreement gets relabeled as groupthink later.
asdf21 7y ago
What about when all the informed experts are wrong? Like consuming dietary cholesterol causing high blood cholesterol? Or trans fats being healthier than saturated fats? Or tariffs harming the US economy?
- lopmotr 7y ago
  How can your group know something that informed experts don't? If you have special knowledge they don't, then they aren't informed. Your group might knowingly take a risk in guessing the experts are wrong, perhaps to test a novel idea, but that's not the same as refusing to recognize their opinion as valid, like in groupthink.
  - asdf21 7y ago
    "Experts" are especially prone to groupthink, this has been shown numerous times. People can be "informed" about something and still need to shut-up about it, when it's politically incorrect / represents a unpopular view, when it puts their income / reputation at risk, etc. Experts have more to lose, so they are especially vulnerable to this sort of pressure.
    - lopmotr 7y ago
      Of course experts can be wrong, but you won't know that they wrong or how in any particular instance, so you won't be justified in agreeing on something contrary and silencing dissenters who bring up the expert opinion. I'm just talking about obvious reasonable behavior here. For instance, experts probably agree that nuclear radiation causes cancer. It would not be sensible to believe that nuclear radiation doesn't cause cancer and then expose people to that radiation, believing it's harmless just because experts are at risk of being corrupted.
Nasrudith 7y ago
I think it is more "obviously wrong in face of evidence and refuse to reconsider" than informed experts per se. Doctors being stuck up about handwashing because they heuristically interpretted "a gentleman's hands are alway clean" as a descriptive instead of a directive was certainly groupthink. It is possible to be wrong and not be engaged in groupthink or right and lacking logical basis.

Nasrudith 7y ago

I believe that is cohesion which isn't inherently linked to group think but attempts to promote may lead to it. To use a cliche combat context it is like loyalty effectively. There may be questions of whether to attack from the woods or an open field with tank support and working together is more important than one or another as even the situational "wrong" choice is better than a split effort. However absolute blind loyalty ends only in needless tears from trying to shoot tanks with pistols, charging the no man's land into machine guns, massacre of villages, or military coups. The dangerous aspect is the "delusion" in both.

bilbo0s 7y ago
Military is a bad example. I can tell you from experience, it's not a place that runs on "groupthink", or "consensus" views. A decision is made by a decision maker, and that decision is implemented. In the military they even have the "mind guards" that "insulate leadership from criticism" that were mentioned in another comment. Only in the military, we call them "NCO's". You can be diametrically opposed to a decision in the military if you like. Heck, you can think whatever you want in the military, because no one cares what you think.

Jtsummers 7y ago

Consensus and common goals aren't groupthink. Groupthink is specifically about when people stop thinking individually, and take the group's position to be correct-by-default. Consensus and common goals can be achieved without that assumption. Groupthink, in software architecture, happens when we can no longer question the basis for a design or assumption. Perhaps the design made perfect sense once, but now we're trying to scale to 1000+ servers, not just 5-10. Communication overhead is killing us, groupthink promotes (for too long, ultimately to the detriment of the business and its customers) the idea that the architecture is beyond question. Consensus and common goals are different. If the architecture is limiting the business, then that can be an agreed upon idea, but not actionable yet. Maybe we realize that other things need to be addressed first (our build system, CI/CD pipelines, testing systems) in order to make architecture updates even feasible. Groupthink would deny even addressing those concerns because they still won't accept that the architecture is limiting the business.