Happiness formulae: an update and critique.

GENERAL SEMANTICS THINKERS have developed several formulae for happiness. A recent issue of ETC discussed and reprinted original writings about such formulae by Alfred Korzybski and his followers, including Wendell Johnson, Harry Weinberg, Irving J. Lee, and Robert P. Pula. (1) Since those texts were written, research in psychology and economics on this topic has continued. I want to show where the GS formulations are supported or discredited by modern research.

Modern research shows that happiness depends on many things, including your expectations, the uncertainty of the result, how much control you feel you have over the result, comparison with other results, alternative ways to get a result, how you value the results, your goals, etc. (2) I'm going to concentrate just on the impact that expectations, results, and those things directly related to them, have on happiness.

First I'm going to define a few terms, as I find people misunderstand a lot in this area when I teach it.

Extensional Definitions: Defining things by non-verbal events, lists of individual members or demonstrations, and observable "facts," e.g., the people Korzybski, Einstein, Johnson, Weinberg, Lee, and Pula are examples of "men"; or I could go to a farm and point out "men" as distinct from the "animals."

Intensional Definitions: Defining things in terms of other words, e.g., "'Man' is a featherless biped." The trouble with this sort of definition is that you can say "There goes a 'man,'" when you see a plucked chicken running away from the chopping block.

Extensional minimum: A minimum of an actual list of items, e.g., if someone scored the following exam results: 83, 88, 65, 78, 74, 87 then the extensional minimum is 65. (3)

Intensional maximum: The highest value in a class by verbal definition, e.g., the maximum exam result is 100. (4)

Intensional minimum: The lowest value in a class by verbal definition, e.g., the minimum exam result is 0. (5)

Expectation: what you think will actually happen. Not what you wish, hope, or desire to happen. You can have an expectation about a goal but it is not a goal, although some may use the term that way.

Goal: What you would like to happen. What you aim to achieve. But you may not actually expect to get there. So a goal is like an aspiration, hope, or desire, etc., e.g., when I buy a Lotto ticket my goal is to win $1,000,000, but I don't think that I will always get there.

Result: The outcome that actually did happen, e.g., the Lotto ticket won a prize of $22.

Value: How much the result or goal or expected result means to you. Economists also use the word utility for how much we value something.

Motivation: How much you want to do something or get to a goal.

Research seems to show that, in terms of expectation (E), results (R) which have happened, and how much one values (V) them, a happiness (H) formula would go something like this:

H = V(R) - V(E) + etc

In the above, Happiness = how much you Value the Result minus how much you Valued what you Expected to get, i.e., there is a gap between your evaluations of what you got and what you expected. It is a bit like profit in a business, i.e., Profit = Income - Costs. To complicate things, there is an indirect relationship between our expectations and the results. Let "[right arrow]" stand for "affects" or "is a partial cause of."

This relationship can be shown as: E [right arrow] V [right arrow] M [right arrow] B [right arrow] R. What you Expect (E) affects the Value (V) you have for the expected result. This Value then affects your Motivation (M). Your Motivation affects your Behavior (B) and this affects the actual Result (R). For example, if you think that Lotto is rigged so that only the organiser's relatives get a payout, then you'd expect to get nothing: E = 0. You value nothing as worthless: V = 0. Hence you have no motivation to buy a Lotto ticket: M = 0. So you don't buy a ticket. Your behavior has changed and the result is no win because you have no ticket: R = 0.

Note: the "[right arrow]" or affect doesn't have just one cause in any of the above links. You could value something for more than just what you expect to get from it. You could be motivated by things other than just the value. You could behave out of habit or error as well as motivation. And the result isn't entirely due to your behavior, especially when winning at Lotto.

The relationship between motivation/stress and behavioral performance is not straightforward, as shown by the Yerkes-Dodson principle in the diagram below. (6)

[GRAPHIC OMITTED]

The Yerkes-Dodson principle says that as your stress from motivation increases, your performance/efficiency increases. Hence your result goes up also, but only up to a point. Then your performance/efficiency starts to decrease as the stress gets too high. So your results go down. As the result changes, so might your happiness.

That this relates to happiness is supported by work reported in the book Flow (3) as shown by a modified diagram below.

[GRAPHIC OMITTED]

If you travel up the cross section line, in the diagram above, (for a set skill level) from low stress/motivation to high, you go from boredom through happiness to anxiety. This follows a similar pattern to the performance-stress curve: happiness first goes up, peaks, and then it comes down as the motivation increases.

Now how does this relate to Korzybski's formula of minimal extensional expectation?

Lets take an example of a students exam results. If students have had the following grades in the last year: C, C, B, B, B, A, then their extensional minimum is a C. Say the student values the grades as per the following table.

Positive values mean the student is happy with the result.

Negative values mean the student is unhappy with the result.

Now let's assume that the student gets the same set of grades the following year and calculate how happy the student would be under several expectations.

Korzybski said to expect an extensional minimum. In this case a C.

Compare this with expecting the intensional maximum. In this case an A.

Finally, if you expect the intensional minimum, which is an F grade, then that would be of no positive value to you. Hence you would not be motivated to study or even sit the exam. Hence you would not get a good result and would be unhappy. This happens via the E [right arrow] V [right arrow] M [right arrow] B [right arrow] R relationship.

So the research backs up Korzybski's formula of minimal extensional expectations. The only caveat I would add is that you may want to modify this expectation a little, based on what you value, so as to keep up your motivation. However, watch out for the danger of only valuing perfection, because if you do, you will end up unhappy again.

The research also supports Johnson's IFD formulation. For example, the more you value something the more you will be upset if you miss out on it.

I want to emphasize that there is a difference between expectations and goals.

[GRAPHIC OMITTED]

The graph above shows the success rate as a % for a goal level. The goal might be how far you can throw a heavy ball. The success rate curve shows that you can throw it 20 feet for certain and 40 feet, 80% of the time, etc. It also shows how you value that goal level, in a typical diminishing returns pattern. From these two bits of information one can calculate the average outcome (result) for a goal (or aspiration) level. The peak average outcome/result is at 35. This is above the best of what you can expect for certain, which is 20. So your best goal is 35, while the best you can expect for certain is 20. The point here is to expect an extensional minimum, which you know you can achieve, and then to set yourself a bit of a challenge with your goal, and go for a goal you may not be able to achieve for certain. The two are different! Then I suggest that you reward yourself for anything you get above your minimal expectation. This will make you happier than punishing yourself for everything that falls below your expectation if it's too high. Note that while the optimal goal stretches you a bit, it does not go totally beyond your capabilities.

Critique of Some Happiness Formulations

While modern research seems to support both Korzybski's minimal extensional happiness formulation and Johnson's IFD formulation, I would like to take to task Weinberg's and Lee's formula for happiness: H = M/E. When E = 0 this formula implies that one would be infinitely happy. And some web sites suggest just this as the way to "Nirvana" (8). But modern research and Korzybski's formulation both say that people with zero expectation tend to be very unhappy. I will admit that Weinberg does rule out having E = 0 as he says "The man with zero expectations is either in complete despair or utterly cynical." Also the formula misrepresents reality when not zero but near it. Would you be twice as happy with an exam result of 50% if you expected to get 10% rather than 20%? Or to make it even more obvious, would you be twice as happy if you expected to get 1% instead of 2%? I can tell you that in the first case, I'd be a bit happier but not twice as happy. And in the second case the difference probably wouldn't make a difference to me. What about for you? But the H = M/E formula predicts that you'd be twice as happy when the expectation is halved. For me, I see this as a case of the map not being structurally similar to the territory.

Another problem is that H = M/E implies that the more motivated you are, the happier you will be. However the Yerkes-Dodson law, that is the relationship between motivational stress and performance/efficiency, shows that there is an optimal level of motivation. When you go beyond that optimum, the results you achieve diminish towards nothing, for example, writer's block. And how happy you are depends on the gap between what you expect and what you get. So if you get over motivated, you tend to get less results. Hence you will become less happy. So again this formula fails to map the territory accurately under some conditions. So while it has some similarity of structure with reality in some contexts, I feel that it is not structurally similar enough to the territory to be used effectively.

Robert Pula's formula of H = ME + MM, I find better, but I would suggest making the following changes. Many people interpret the ME (Minimal Expectations) as an intensional Minimum Expectation, which we know tends to make one unhappy. It can be a call for mediocrity for some. So I would suggest that we change it to MEE (Minimal Extensional Expectations). However, the MM (Maximum Motivation) has the same problem as Lee's formulation, as it subscribes to too much motivation. I'd suggest changing this to OM (Optimal Motivation). I will admit of course that one can interpret "maximum motivation" as "optimal motivation" or "maximally useful motivation." So Pula's formula then becomes H = MEE + OM which reads "Happiness equals Minimal Extensional Expectations plus Optimal Motivation." Of course this is an abstraction that leaves out a lot of other factors, so we could put in an "etc.," also.

To Sum Up

I think I can now answer the question (1): "Should we 'dream the impossible dream' or 'dream the possible dream'?" For maximum happiness, I think our expectations should stay with the possible, at or near the extensional minimum suggested by Korzybski, with some minor adjustments based on what we value so as to get enough motivation. And if you only value perfection, or value things too much, then I suggest that you may need to change the way you value your results, if you want to become happier.

I think however that our goals should stretch us beyond the 100% possible, but not all the way to the impossible, i.e., take a degree rather than "either/or" orientation here.

NOTES

1. ETC: A Review of General Semantics. Vol. 62, No. 3, p. 284

2. Other factors that are not related to expectation affect happiness:

a. Alternate plans for the result reduce our risk and hence our anxiety about getting nothing.

b. How much control we feel over the result also affects our happiness. The optimum feeling of control is near what an accurate 'map' would show. Feeling significantly less or more control than you actually have causes problems.

c. How we compare the result to other results also affects our happiness, e.g., comparing our grades to those of our friends.

d. Failure tends to have a bigger impact on us than success.

e. Etc. For example, see research by Daniel Kahneman for other structural factors.

3. In Lecture 6 of General Semantics Seminar 1937, given at Olivet College, Korzybski gives his Extensional Theory of Happiness. In this an extensional minimum is his E2.

4. As in (3) Korzybski's E1 is an intensional maximum.

5. As in (3) Korzybski's E3 is an intensional minimum. Typically an unhappy person starts of with an E1, gets frustrated with the results and then ends up with an E3, still unhappy but for different reasons.

6. See The Joy of Stress by Dr Peter Hanson, p. 17 for a diagram on this.

7. For the actual diagram see: Flow: The Psychology of Happiness by Mihaly Csikszentmihalyi, p. 74

8. I found a web site with a related formula to H = M/E:

Happiness = [Desires fulfilled (= Results)]/[Desires entertained (= Expectations)]

The web site text says, "Clearly, the less desires one has the greater the proportion of those entertained that will be fulfilled and the greater the happiness quotient. The other implication of the formula is that if one can have no desires, then one is infinitely happy; the Buddhist call this state 'Nirvana'; the Taoist call this finding the 'Tao,' returning to the great void."

This I see as similar to Korzybski's description of the Aristotelian problem of 'infinity.'

DAVID HEWSON*

* David Hewson, author of other ETC articles, including "Problem Solving with General Semantics" in ETC, Vol. 53, No. 2, has an Institute of General Semantics "Approved Teacher of General Semantics" certificate and teaches GS in AGS seminars. He also runs a web site, www.dh.id.au, where you can search an online index for particular issues of ETC or try out your GS knowledge with some exercises.

Result  Happiness Value

F           -10
D            -5
C             5
B             8
A            10

Expect  Result  V(R) - V(E) = Happiness

C       C        5-5 = 0
C       C        5-5 = 0
C       B        8-5 = 3
C       B        8-5 = 3
C       B        8-5 = 3
C       A       10-5 = 5
        Total         14--very happy

Expect  Result  V(R) - V(E) = Happiness

A        C        5-10 = -5
A        C        5-10 = -5
A        B        8-10 = -2
A        B        8-10 = -2
A        B        8-10 = -2
A        A       10-10 =  0
         Total          -16--very unhappy