Diagnosing Tangled Situations

System archetypesPart V · Pattern and Judgment · Chapter 11 of 12 · the second of two chapters on system archetypes

What if more than one story is true at once? Four more archetypes: two sides answering each other, growth short of capacity, a target that slips, partners who get in each other’s way. One loop structure can be read as two of these stories, and one situation can hold two archetypes, coupled through the variables their loops share.

Print this chapter, with a sheet for each exercise and the answers at the back, from its PDF.

Birthday parties in one class

In one school class, each family holds a birthday party for its child and sets a limit on how many children it invites. Each family measures its party against the others’: a party that looks small grows at the next birthday, and one that looks large makes the next family invite more. Guests over the limit are cut from the next party’s list, a birthday later, or the family raises its limit toward the party it held.

B1 and B2 run through how ours compares, one family’s party against the others’: escalation. B3 and B4 run through guests over our limit, the party’s size less the limit: eroding goals, with its delay on the cut. Both archetypes are declared, and the loops match both.

What the notation expresses

Escalation

Escalation is two sides acting on one comparison, relative position: A’s activity raises it and B’s lowers it. When it falls, A does more (B1); when it rises, B does (B2). Both loops are balancing, and together they trace a figure-8 that reinforces: each side’s response is the other’s reason to respond. The figure-8 passes relative position twice, so it is no loop of its own and carries no label. Kim notes that it takes two sides to escalate and one to stop, and asks whether the relative measure between them can change.

Growth and underinvestment

Growth and underinvestment is limits to growth with a third loop. Growth effort and demand raise each other (reinforcing loop R1, the growth loop) until demand lowers performance (balancing loop B1, the limit loop). Performance below the performance standard raises the perceived need to invest, which adds capacity after a delay and restores performance (B2, the capacity loop). Where the standard is lowered as performance falls, the need to invest falls too, and performance with it (Kim). Kim puts the leverage in judging capacity against demand before performance falls. The performance standard acts from outside the loops, so the template leaves it out.

Eroding goals

In eroding goals, a gap between a target and actual performance closes two ways: corrective action raises actual performance, after a delay (B1), and pressure to lower the target brings the target down (B2). Kim calls eroding goals drifting goals. Lowering the target closes the gap at once and corrective action takes time, so B2 acts first: loop dominance. Unlike chapter 8’s target, here the target moves: the gap lowers it. Kim finds that targets set from outside the situation drift less.

In the parties, the gap is guests over our limit, the party’s size less the limit: named as its opposite, so B3’s and B4’s effects carry the opposite polarities and both loops stay balancing.

Accidental adversaries

In accidental adversaries, two partners each raise the other’s success (R1, partnership). When either is dissatisfied with its own results, it takes corrective measures alone (B2, B3) that, unseen, lower its partner’s success; together the two sets of fixes close R4, the adversaries’ loop. Kemeny’s case, where the archetype was first recognized, is Procter & Gamble’s price promotions and Walmart’s stocking up on promoted goods; she puts the leverage in each partner understanding what the other needs. Each side’s fixes, with the partner’s success on the way back, also match fixes that fail, within accidental adversaries.

One structure, two stories

Escalation and eroding goals have one shape: two balancing loops sharing one variable. Matching reads loops and their kinds, so every match of one is a match of the other, and the lab lists one with Also reads as the other. The stories differ in what the shared variable is: one side’s position relative to another’s, in escalation, or a gap between a target and actual performance, in eroding goals. Matching can’t read that, so the loops alone don’t settle which story holds; the declaration records the drawer’s reading.

Check your understandingTwo street musicians, Ines and Kofi, play on one corner, and each plays louder when the other drowns them out. A diagram shows Ines's volume → Ines against Kofi +, Ines against Kofi → Ines's volume −, Kofi's volume → Ines against Kofi −, and Ines against Kofi → Kofi's volume +. Its loops match escalation and eroding goals. Which story is it, and why?

Escalation: the shared variable sets one musician's volume against the other's

Yes. Both archetypes are two balancing loops sharing one variable, so the loops match both. Ines against Kofi compares one side's position with the other's, as escalation's relative position does. A gap would be a target less actual performance: the volume Ines wants less Ines's volume.

Eroding goals: Kofi's volume is the target Ines's volume closes on

Not quite. Kofi's volume moves with the shared variable, as eroding goals' target moves with the gap, so the loops fit. But nobody chose it as a value to stay near: it is the other side's position, and that makes the story escalation.

Either: the loops match both, so both declarations are equally right

Not quite. The loops match both because matching reads loops and their kinds, not what a variable is. The stories differ in what the shared variable is: here it compares the two musicians, so the declaration is escalation.

Coupled archetypes

Coupled archetypes are two matches sharing a variable: coupled loops, read as archetypes, so a change in one archetype’s loops changes the other’s. In the parties, escalation (B1, B2) and eroding goals (B3, B4) share our party’s size: a party grown to keep up opens a gap over the limit, and a party cut back to the limit looks small again. The lab reports them coupled through our party’s size and how ours compares, pairing escalation with the eroding goals that B1 and B3, two balancing loops sharing our party’s size, also match. An external driver acting on both archetypes’ loops doesn’t couple them: it stands in for no variable of any match.

Ranking leverage

Where several leverage points are proposed, which to try first is a ranking, written beside the diagram, in words, each leverage point with its reason. No mark ranks: every leverage point is the same box. For the parties:

how ours compares: both escalation loops pass through it, and Kim asks of escalation whether the comparison between the sides can change. A class that agrees one number of guests for every party leaves nothing to compare.
our limit on guests: held where the family set it, B4 stops, and B3 cuts each party back to it, a birthday late.

A reason resting on a dashed effect, a hypothesis in evidence coding, is a hypothesis too.

Check your understandingA pottery studio's evening classes fill by word of mouth, and its waiting list grows. A diagram of the studio shows two leverage points: the wheels in the studio, and the waiting list the owner treats as normal. Its author judges the waiting list the better place to act. How does the diagram show that?

A ranking written beside it: the waiting list first, the wheels second, each with its reason

Yes. No mark ranks leverage points: each is the same box with its question marks. The order and the reasons go beside the diagram in words, where a reader can test each reason against the loops.

A heavier box around the waiting list

Not quite. Every mark is drawn in pencil in one weight, so a heavier box carries nothing a reader can read. A leverage point's box marks a hypothesis about where to act, not its rank.

Only the waiting list boxed, the wheels left plain

Not quite. That drops the second proposal, and the comparison with it. Both boxes stay, and the ranking, with its reasons, goes beside the diagram.

The bad diagram

What’s wrong with this diagram of the birthday parties?

It shows the comparison alone, declared escalation. Our limit on guests and guests over our limit hold the party’s size down but close no loop: the party’s size never reaches guests over our limit, and the guests over it never move the limit. The lab’s check for two coupled archetypes reports:

1 declared archetype. Brief: at least 2, coupled through a shared variable.

The corrected diagram

The party’s size raises guests over our limit, closing B3, the cut, and the guests over the limit raise the limit, closing B4. Eroding goals is declared beside escalation.

Bad diagram

Corrected diagram

Labels unchanged. Declared archetypes: eroding goals added. Effects: 2 added. Loops: 2 added. Now matched with its delay: eroding goals.

Notation moves

Notation move	When	What it shows	Example
Declare two archetypes on one diagram	Two stories run through one situation	Coupled archetypes, where their matches share a variable	Archetype: escalation and Archetype: eroding goals

Declare escalation on two balancing loops sharing a variable	The shared variable compares two sides	The story the loops alone don’t settle	how ours compares

Draw the gap as a variable between the target and actual performance	The target moves with the gap	Eroding goals’ second loop	guests over our limit

Mark the delay on corrective action’s effect	Correction arrives later than a lowered target	Which loop acts first	guests over our limit → our party’s size

Mark the delay into capacity	Capacity takes time to build	Capacity arriving late	need for more trains → trains in service, in AR 07

Propose several leverage points	There is more than one place to act	Hypotheses about where to act, ranked in words beside the diagram	how ours compares, then our limit on guests

Common misreading

Read one way, a match is a diagnosis: the loops are there, so the situation is that archetype. In the parties, B1 and B2 also match eroding goals, how ours compares as the gap and either family’s party as the target, and B3 and B4 also match escalation. The loops are the same either way; the stories differ in what the variables are. How ours compares sets one family’s party against the others’, so B1 and B2 are escalation; guests over our limit is a party’s size less a number the family chose, so B3 and B4 are eroding goals.

The strongest objection

Archetypes fit any story told loosely enough, so naming one explains nothing. Lane (1998) divides generic structure in system dynamics into three kinds, of which the third, archetypes, involves “merely metaphorical thinking”.

The names are loose: escalation and eroding goals share one shape, and which one a drawing holds is the drawer’s reading, recorded in the declaration, not settled by the loops. The structure isn’t loose. A loop closes or doesn’t, its kind follows the parity rule, a delay sits on one effect and not another, and each is checked on the drawing and can be wrong: the bad diagram’s limit closes no loop, and the lab counts one declared archetype where two coupled ones were asked for.

Check your understanding

Check your understandingSam and Jo share a flat and take turns cooking dinner for both, which leaves each of them free evenings. Short of time in the week, each starts cooking a week's lunches on Sunday afternoon. Sam's cooking takes the oven and every pan when Jo comes to cook, and Jo's leaves a sink of washing-up that Sam clears before starting. Which archetype's story is this?

Accidental adversaries

Yes. Their turns at cooking free each other's evenings: the partnership loop. Each Sunday's cooking eases its cook's own week, a balancing loop for each, and gets in the other's way; the two together close the adversaries' loop. Kemeny puts the leverage in each partner learning what the other needs.

Escalation

Not quite. Escalation's two sides act on one comparison, each against the other's lead. Neither flatmate measures their week against the other's: each Sunday's cooking is for its cook's own week, and it gets in the other's way as it goes.

Fixes that fail, once for each flatmate

Not quite. A drawing of the flat does match fixes that fail, one flatmate's Sunday cooking as the symptomatic solution and its side effect running through the other's week. But each such match lies within accidental adversaries, and read alone it leaves out the partnership loop that ties the two weeks together.

The VA’s staffing goal, redrawn

Redrawn from Erin S. Kenzie, Mary Patzel, Erik Nelson, Travis Lovejoy, Sarah Ono and Melinda M. Davis, “Long drives and red tape: mapping rural veteran access to primary care using causal-loop diagramming”, BMC Health Services Research 22, 1075 (2022), Fig. 5. CC BY 4.0.

Its blue lines marked + and red lines marked − drawn alike, each polarity kept; its faded variable, provider turnover, and its line left out; VA dropped from every name; sending veterans to community care renamed as an amount, veterans sent to community care, and three other names reworded; rurality of the facility marked as an external driver; its loops’ labels kept, B9 to B11 numbered as in the authors’ whole model, and their names reworded by what each loop does; eroding goals, which its authors name drifting goals, declared.

Question 1 of 3Which variable is eroding goals’ target, and what brings it down?

Staffing need, lowered by community care.

Not quite. Staffing need is the gap: the staffing goal raises it and provider retention lowers it. Community care reaches the need only by lowering the staffing goal, around B10 and B11.

Staffing goal, lowered by veterans sent to community care (B10) and by the downsizing of rural services that follows (B11).

Yes. A higher staffing need sends more veterans to community care, and each − effect into the staffing goal brings the goal down, closing the gap from the target’s side. Here the target moves, and recruitment and retention, B9, are left less to close.

Provider retention, lowered by provider burnout.

Not quite. Provider retention is actual performance, the side recruitment raises through B9. Burnout lowers it around R1, but no variable sets it as a target is set.

Question 2 of 3Its authors count R1, the burnout spiral, in their drifting goals (eroding goals). With B9, R1 matches another archetype, coupled with eroding goals. Which?

Shifting the burden, with community care as the symptomatic solution.

Not quite. Community care relieves the staffing need only by lowering the staffing goal, eroding goals’ second loop, and R1 doesn’t pass through it. No shifting the burden matches here.

Escalation, between provider recruitment and provider burnout.

Not quite. Escalation is two balancing loops sharing one variable, and R1 is reinforcing. The escalation that matches here is eroding goals’ own shape: B9 with B10 or B11.

Fixes that fail: B9 brings retention back as the need rises, and R1 drives it down through patients per provider and burnout.

Yes. With provider retention named as its opposite, provider loss, as the problem symptom, B9 is the fix and R1 the side effect’s loop, its delay not drawn. B9 is eroding goals’ corrective action too, so the two matches share the staffing need, recruitment and retention: coupled archetypes.

Question 3 of 3Ranked by what Kim finds of eroding goals, which leverage point comes first?

Provider recruitment, pushed harder.

Not quite. Recruitment corrects through B9, after a delay, while a lowered staffing goal closes the gap at once; and R1’s burnout drives out the providers recruited. While the goal can fall, a recruitment push closes the gap no sooner.

Veterans sent to community care, cut back.

Not quite. Community care reaches veterans who would otherwise drive far for care, and it lowers the target only through its effect on the staffing goal. Holding the goal itself stops B10 and B11 with no veteran sent back to the road.

The staffing goal, set from outside the facilities and held there.

Yes. Kim finds that targets set from outside the situation drift less. Held, the staffing goal no longer falls with the community care sent, so B10 and B11 stop, and the gap stays open until B9 closes it.

Exercises

Passing the checks means meeting the brief, not being right.

AR 06An arms race between two neighboring countries

Each side's safety read against the other's

Reasoning3 of 5
Polish4 of 5

Two neighboring countries, called here North and South, each judge their safety by how their armed forces compare with the other's. North buys arms until its forces are well ahead of South's: the more arms North buys, the larger North's lead in forces, and the larger that lead, the fewer arms North buys. South buys arms while North is ahead: the more arms South buys, the smaller North's lead, and the larger North's lead, the more arms South buys. Each country pays for its arms from the budget that also pays for its schools and hospitals: the more arms either country buys, the less the two countries spend on schools and hospitals.

The brief

Draw each country's arms purchases closing a balancing loop through one variable that compares the two countries' forces. checked by the lab
The two loops match escalation. checked by the lab
Draw the spending on schools and hospitals, reached from both countries' purchases. checked by the lab
One variable compares the two countries' forces, and both loops pass through it. you check this
The spending on schools and hospitals returns to neither loop. you check this
The declared archetype is escalation, not eroding goals, though the same loops match both. you check this

Your palette

variable
effect
polarity
delay
loop
balancing loop
archetype

To draw in the lab, use a tablet or a larger screen. Everything else reads normally here.

Press an empty spot to add a variable and type its name. Drag a variable to move it. Select a variable, an effect or a loop label to change it. Double-click or double-tap a variable or a loop's label or name to edit its words in place, or select it and use Rename or the fields under the canvas.

Nothing selected. Select a variable, an effect, a loop label or a declared archetype on the canvas, and what you can change shows here.

The tools

Variables V: add a variable (press an empty spot), move one (drag it), rename one (double-click it, or Rename under the canvas), or select anything.
Effects E: draw an effect by dragging from one variable to another.
Loop L: declare a loop by selecting its effects in order, then label and name it.
Archetype K: declare an archetype: press an empty spot, then choose which of the nine.

Keys

R reverse the selected effect
+− the selected effect's polarity
Delete delete what's selected
Ctrl+Z undo
Ctrl+Shift+Z redo
Esc clear the selection
Enter finish the words you're writing
Esc cancel them

Keys act on the lab you last used, or on the open lab view, and never while you type.

Archetypes the structure matches

Read from the loops and their kinds as drawn, after each change. Nothing here is saved.

No archetype matches yet.

0 of 8 checks passed; 8 pending.

Draw each country's arms purchases closing a balancing loop through one variable that compares the two countries' forces. (Not started)No loop declared yet. Brief: exactly 2 balancing loops.
The two loops match escalation. (Not started)No effect has a polarity yet. Brief: escalation.
Draw the spending on schools and hospitals, reached from both countries' purchases. (Not started)No variables yet. Brief: at least 4.
One variable compares the two countries' forces, and both loops pass through it. You check this.
The spending on schools and hospitals returns to neither loop. You check this.
The declared archetype is escalation, not eroding goals, though the same loops match both. You check this.

Always checked

Each loop's kind matches its − effects (Not started)No loop declared yet.
Loop labels are unique and match their kind (Not started)No loop declared yet.
Every effect has a polarity (Not started)No effects yet.
The structure matches each declared archetype (Not started)No archetype declared yet.
No variable's name breaks the more test's three rules (Not started)No variable drawn yet.

Written feedback

Turn on written feedback

Export

One defensible answer

It opens here once you've drawn at least one variable in the lab above. Your attempt saves as you draw.

AR 07A country's railways

Growth held back by trains never ordered

Reasoning4 of 5
Polish4 of 5

A country's railway operator draws passengers with punctual trains and cheap fares: the more passengers, the more fare offers it can pay for, and the more fare offers, the more passengers. Crowded trains wait longer at every station: the more passengers, the lower the punctuality, and the higher the punctuality, the more passengers. The operator reads its need for more trains from how far punctuality falls short of its punctuality target, once a year's figures are in: the lower the punctuality, the greater the need for more trains, after a delay, and the higher the target, the greater the need. A new train takes four years to build: the greater the need, the more trains in service, after a delay, and the more trains in service, the higher the punctuality. When the need is large, the operator lowers its target rather than order trains: the greater the need for more trains, the lower the punctuality target.

The brief

Draw the passengers and the fare offers closing a reinforcing loop. checked by the lab
Draw a balancing loop through the passengers and the punctuality, and one through the need for more trains and the trains in service. checked by the lab
The loops match growth and underinvestment, with its delay where new trains arrive. checked by the lab
Draw a delay where the operator reads its need for more trains, and one where new trains arrive. checked by the lab
The punctuality target acts on the need for more trains, and the greater the need, the lower the target. you check this

Your palette

variable
effect
polarity
delay
loop
reinforcing loop
balancing loop
gapdrawn as an ordinary variable
archetype

To draw in the lab, use a tablet or a larger screen. Everything else reads normally here.

Nothing selected. Select a variable, an effect, a loop label or a declared archetype on the canvas, and what you can change shows here.

The tools

Variables V: add a variable (press an empty spot), move one (drag it), rename one (double-click it, or Rename under the canvas), or select anything.
Effects E: draw an effect by dragging from one variable to another.
Loop L: declare a loop by selecting its effects in order, then label and name it.
Archetype K: declare an archetype: press an empty spot, then choose which of the nine.

Keys

R reverse the selected effect
+− the selected effect's polarity
Delete delete what's selected
Ctrl+Z undo
Ctrl+Shift+Z redo
Esc clear the selection
Enter finish the words you're writing
Esc cancel them

Keys act on the lab you last used, or on the open lab view, and never while you type.

Archetypes the structure matches

Read from the loops and their kinds as drawn, after each change. Nothing here is saved.

No archetype matches yet.

0 of 9 checks passed; 9 pending.

Draw the passengers and the fare offers closing a reinforcing loop. (Not started)No loop declared yet. Brief: at least 1 reinforcing loop.
Draw a balancing loop through the passengers and the punctuality, and one through the need for more trains and the trains in service. (Not started)No loop declared yet. Brief: at least 2 balancing loops.
The loops match growth and underinvestment, with its delay where new trains arrive. (Not started)No effect has a polarity yet. Brief: growth and underinvestment, with its delay.
Draw a delay where the operator reads its need for more trains, and one where new trains arrive. (Not started)No effects yet. Brief: at least 2 delays.
The punctuality target acts on the need for more trains, and the greater the need, the lower the target. You check this.

Always checked

Each loop's kind matches its − effects (Not started)No loop declared yet.
Loop labels are unique and match their kind (Not started)No loop declared yet.
Every effect has a polarity (Not started)No effects yet.
The structure matches each declared archetype (Not started)No archetype declared yet.
No variable's name breaks the more test's three rules (Not started)No variable drawn yet.

Written feedback

Turn on written feedback

Export

One defensible answer

It opens here once you've drawn at least one variable in the lab above. Your attempt saves as you draw.

AR 08Knee exercises after an operation

A target that moves toward what was done

Reasoning4 of 5
Polish4 of 5

After an operation on her knee, Priya sets herself a weekly target of fifteen sessions of the exercises her physiotherapist gave her. Each Sunday she counts the sessions done that week, and the gap is the target minus the sessions done: the higher the target, the larger the gap, and the more sessions done, the smaller the gap. When the gap is large she fits extra sessions into the next week, so the gap raises the sessions done a week later: the larger the gap, the more sessions done, after a delay. On busy weeks at work she falls short, and each time she does, she sets next week's target closer to what she managed: the larger the gap, the lower the weekly target.

The brief

Draw two balancing loops through the gap: one through the sessions done, one through the weekly target. checked by the lab
The loops match eroding goals, with its delay on the extra sessions. checked by the lab
The gap is a variable of its own: the weekly target minus the sessions done. you check this
The second loop is named for the lowered target, which closes the gap without a session more. you check this

Your palette

variable
effect
polarity
delay
loop
balancing loop
gapdrawn as an ordinary variable
archetype

To draw in the lab, use a tablet or a larger screen. Everything else reads normally here.

Nothing selected. Select a variable, an effect, a loop label or a declared archetype on the canvas, and what you can change shows here.

The tools

Variables V: add a variable (press an empty spot), move one (drag it), rename one (double-click it, or Rename under the canvas), or select anything.
Effects E: draw an effect by dragging from one variable to another.
Loop L: declare a loop by selecting its effects in order, then label and name it.
Archetype K: declare an archetype: press an empty spot, then choose which of the nine.

Keys

R reverse the selected effect
+− the selected effect's polarity
Delete delete what's selected
Ctrl+Z undo
Ctrl+Shift+Z redo
Esc clear the selection
Enter finish the words you're writing
Esc cancel them

Keys act on the lab you last used, or on the open lab view, and never while you type.

Archetypes the structure matches

Read from the loops and their kinds as drawn, after each change. Nothing here is saved.

No archetype matches yet.

0 of 7 checks passed; 7 pending.

Draw two balancing loops through the gap: one through the sessions done, one through the weekly target. (Not started)No loop declared yet. Brief: exactly 2 balancing loops.
The loops match eroding goals, with its delay on the extra sessions. (Not started)No effect has a polarity yet. Brief: eroding goals, with its delay.
The gap is a variable of its own: the weekly target minus the sessions done. You check this.
The second loop is named for the lowered target, which closes the gap without a session more. You check this.

Always checked

Each loop's kind matches its − effects (Not started)No loop declared yet.
Loop labels are unique and match their kind (Not started)No loop declared yet.
Every effect has a polarity (Not started)No effects yet.
The structure matches each declared archetype (Not started)No archetype declared yet.
No variable's name breaks the more test's three rules (Not started)No variable drawn yet.

Written feedback

Turn on written feedback

Export

One defensible answer

It opens here once you've drawn at least one variable in the lab above. Your attempt saves as you draw.

AR 09Two parents sharing care after a separation

Partners whose own fixes hold each other back

Reasoning5 of 5
Polish5 of 5

Maya and Tom separated last year and share the care of their two children, a week each. Each helps the other's weeks go well, with lifts to football and the same homework rules in both homes: the better Maya's weeks go, the better Tom's go, and the better Tom's weeks go, the better Maya's go. When a week goes badly, each makes changes alone to ease their own weeks, such as a new bedtime or a swapped weekend: the worse Maya's weeks go, the more changes she makes alone, and the more changes she makes alone, the better her weeks go; the same holds for Tom. Each one's changes reach the other's home a week or two later, with the children: the more changes Maya makes alone, the worse Tom's weeks go, after a delay, and the more changes Tom makes alone, the worse Maya's weeks go, after a delay. Both parents see how settled the children are, which rises the better either parent's weeks go.

The brief

The loops match accidental adversaries. checked by the lab
Draw a delay on each parent's changes' effect on the other's weeks. checked by the lab
Mark one leverage point. checked by the lab
The two delays sit where each parent's changes reach the other's weeks, not on the help each gives the other. you check this
The leverage point is a measure both parents can see. you check this

Your palette

variable
effect
polarity
delay
loop
reinforcing loop
balancing loop
leverage point
archetype

To draw in the lab, use a tablet or a larger screen. Everything else reads normally here.

Nothing selected. Select a variable, an effect, a loop label or a declared archetype on the canvas, and what you can change shows here.

The tools

Variables V: add a variable (press an empty spot), move one (drag it), rename one (double-click it, or Rename under the canvas), or select anything.
Effects E: draw an effect by dragging from one variable to another.
Loop L: declare a loop by selecting its effects in order, then label and name it.
Archetype K: declare an archetype: press an empty spot, then choose which of the nine.

Keys

R reverse the selected effect
+− the selected effect's polarity
Delete delete what's selected
Ctrl+Z undo
Ctrl+Shift+Z redo
Esc clear the selection
Enter finish the words you're writing
Esc cancel them

Keys act on the lab you last used, or on the open lab view, and never while you type.

Archetypes the structure matches

Read from the loops and their kinds as drawn, after each change. Nothing here is saved.

No archetype matches yet.

0 of 8 checks passed; 8 pending.

The loops match accidental adversaries. (Not started)No effect has a polarity yet. Brief: accidental adversaries.
Draw a delay on each parent's changes' effect on the other's weeks. (Not started)No effects yet. Brief: at least 2 delays.
Mark one leverage point. (Not started)No variables yet. Brief: exactly 1 leverage point.
The two delays sit where each parent's changes reach the other's weeks, not on the help each gives the other. You check this.
The leverage point is a measure both parents can see. You check this.

Always checked

Each loop's kind matches its − effects (Not started)No loop declared yet.
Loop labels are unique and match their kind (Not started)No loop declared yet.
Every effect has a polarity (Not started)No effects yet.
The structure matches each declared archetype (Not started)No archetype declared yet.
No variable's name breaks the more test's three rules (Not started)No variable drawn yet.

Written feedback

Turn on written feedback

Export

One defensible answer

It opens here once you've drawn at least one variable in the lab above. Your attempt saves as you draw.

AR 10A coastal town and its sea walls

Two archetypes sharing one variable

Reasoning5 of 5
Polish5 of 5

A coastal town of eight thousand people floods when winter storms push the sea over its seafront: the more winter storms, the more flooding of the seafront. Each flood brings calls for sea walls, and a wall stops the sea at once: the more flooding of the seafront, the more sea walls built, and the more sea walls built, the less flooding. The walls stop the waves from carrying sand along the shore, so the beach in front of them thins, and years later the sea reaches the walls with more force: the more sea walls built, the more sand lost from the beach, and the more sand lost, the more flooding, after a delay. Restoring the dunes behind the beach would protect the coast for decades, though new dunes take years to grow: the more flooding, the more dune restoration, and the more dune restoration, the less flooding, after a delay. Behind each new wall, the town allows homes and shops: the more sea walls built, the more building behind the walls. The town's engineers think the new building makes dune restoration costlier, since the land behind the beach is no longer free, but they have no figures yet: the more building behind the walls, the less dune restoration.

The brief

Declare each archetype the structure holds, at least two. checked by the lab
The declared archetypes are coupled: their matches share a variable. checked by the lab
Mark at least two leverage points. checked by the lab
Code the effect you judge less supported as a hypothesis, declared in the legend. checked by the lab
Draw at least three loops. checked by the lab
Beside the diagram, rank the leverage points in a list, each with its reason. you check this
The winter storms sit outside a boundary named for the town's coast, and nothing inside acts on them. you check this
No variable is drawn twice: the sea walls built are one variable, in both archetypes. you check this

Your palette

variable
effect
polarity
delay
loop
reinforcing loop
balancing loop
boundary
external driver
evidence coding
leverage point
behavior note
archetype
symptomatic solutiona role, written above its variable's name
fundamental solutiona role, written above its variable's name
side effecta role, written above its variable's name

To draw in the lab, use a tablet or a larger screen. Everything else reads normally here.

Press an empty spot to add a variable and type its name. Drag a variable to move it. Select a variable, an effect or a loop label to change it. Drag a boundary by its outline or its name to move it, and by the square at its corner to resize it. Drag a behavior note to move it. Double-click or double-tap a variable, a loop's label or name, a boundary's name or a behavior note to edit its words in place, or select it and use Rename or the fields under the canvas.

Nothing selected. Select a variable, an effect, a loop label, a boundary, a behavior note or a declared archetype on the canvas, and what you can change shows here.

The tools

Variables V: add a variable (press an empty spot), move one (drag it), rename one (double-click it, or Rename under the canvas), or select anything.
Effects E: draw an effect by dragging from one variable to another.
Loop L: declare a loop by selecting its effects in order, then label and name it.
Boundary B: drag out a boundary around the system, and name it.
Note N: place a behavior note (press where it goes).
Archetype K: declare an archetype: press an empty spot, then choose which of the nine.

Keys

G legend on or off
R reverse the selected effect
P the selected variable's role, in turn: none, then each role this exercise offers
+− the selected effect's polarity
Delete delete what's selected
Ctrl+Z undo
Ctrl+Shift+Z redo
Esc clear the selection
Enter finish the words you're writing
Esc cancel them

Keys act on the lab you last used, or on the open lab view, and never while you type.

Archetypes the structure matches

Read from the loops and their kinds as drawn, after each change. Nothing here is saved.

No archetype matches yet.

0 of 14 checks passed; 14 pending.

Declare each archetype the structure holds, at least two. (Not started)No variables yet. Brief: at least 2 declared archetypes.
The declared archetypes are coupled: their matches share a variable. (Not started)No archetype declared yet. Brief: at least 2, coupled through a shared variable.
Mark at least two leverage points. (Not started)No variables yet. Brief: at least 2 leverage points.
Code the effect you judge less supported as a hypothesis, declared in the legend. (Not started)No effects yet.
Draw at least three loops. (Not started)No loop declared yet. Brief: at least 3 loops.
Beside the diagram, rank the leverage points in a list, each with its reason. You check this.
The winter storms sit outside a boundary named for the town's coast, and nothing inside acts on them. You check this.
No variable is drawn twice: the sea walls built are one variable, in both archetypes. You check this.

Always checked

Each loop's kind matches its − effects (Not started)No loop declared yet.
Loop labels are unique and match their kind (Not started)No loop declared yet.
No effect runs into an external driver (Not started)No variable is marked as an external driver yet.
External drivers sit outside every boundary (Not started)No variable is marked as an external driver yet.
The legend is on when effects are coded or tagged (Not started)No effect is coded as a hypothesis or tagged with a viewpoint yet, so the legend has nothing to declare.
Every effect has a polarity (Not started)No effects yet.
The structure matches each declared archetype (Not started)No archetype declared yet.
Each role is the place its variable stands in a match (Not started)No variable has a role yet.
No variable's name breaks the more test's three rules (Not started)No variable drawn yet.

Written feedback

Turn on written feedback

Export

One defensible answer

It opens here once you've drawn at least one variable in the lab above. Your attempt saves as you draw.

CS 11Kelsham Primary School: the town school's pull and its rising class size

Two archetypes coupled through one share

Reasoning5 of 5
Polish5 of 5

Brayford Primary School teaches each year in classes of its own, and Kelsham's families know it. The larger Brayford's share of the villages' starters, the more Kelsham children at Brayford, and the more Kelsham children at Brayford, the larger the share, as brothers, sisters and friends follow them there. The larger Brayford's share, the fewer pupils on roll at Kelsham; the fewer pupils on roll, the more mixed-age classes at Kelsham; and the more mixed-age classes, the larger Brayford's share. Helen Marsh, Brayford's head teacher, aimed for classes of 26. The more Kelsham children at Brayford, the larger its average class size, and the larger the average class size, the more pupils a class above Brayford's aim. Above its aim, Brayford offers fewer places to children from outside the town the next September: the more pupils a class above the aim, the smaller Brayford's share, a year later. Yet each year its classes run above the aim, Brayford's governors raise it, toward the most they allow, 30: the more pupils a class above the aim, the higher the class size Brayford aims for, and the higher the class size Brayford aims for, the fewer pupils a class above it.

The brief

Draw Brayford's share of the villages' starters on two reinforcing loops that match success to the successful. checked by the lab
Draw the gap between Brayford's class size and its aim on two balancing loops that match eroding goals, with its delay on the places held back. checked by the lab
Declare both archetypes. checked by the lab
The declared archetypes are coupled: their matches share a variable. checked by the lab
Mark one delay: the September before places held back change the share. checked by the lab
Brayford's share of the villages' starters is drawn once, on both archetypes' loops. you check this
Beside the diagram, one sentence says what each rise in Brayford's aim does to the pupils on roll at Kelsham. you check this

Your palette

variable
effect
polarity
delay
loop
reinforcing loop
balancing loop
gapdrawn as an ordinary variable
archetype

To draw in the lab, use a tablet or a larger screen. Everything else reads normally here.

Nothing selected. Select a variable, an effect, a loop label or a declared archetype on the canvas, and what you can change shows here.

The tools

Variables V: add a variable (press an empty spot), move one (drag it), rename one (double-click it, or Rename under the canvas), or select anything.
Effects E: draw an effect by dragging from one variable to another.
Loop L: declare a loop by selecting its effects in order, then label and name it.
Archetype K: declare an archetype: press an empty spot, then choose which of the nine.

Keys

R reverse the selected effect
+− the selected effect's polarity
Delete delete what's selected
Ctrl+Z undo
Ctrl+Shift+Z redo
Esc clear the selection
Enter finish the words you're writing
Esc cancel them

Keys act on the lab you last used, or on the open lab view, and never while you type.

Archetypes the structure matches

Read from the loops and their kinds as drawn, after each change. Nothing here is saved.

No archetype matches yet.

0 of 10 checks passed; 10 pending.

Draw Brayford's share of the villages' starters on two reinforcing loops that match success to the successful. (Not started)No effect has a polarity yet. Brief: success to the successful.
Draw the gap between Brayford's class size and its aim on two balancing loops that match eroding goals, with its delay on the places held back. (Not started)No effect has a polarity yet. Brief: eroding goals, with its delay.
Declare both archetypes. (Not started)No variables yet. Brief: exactly 2 declared archetypes.
The declared archetypes are coupled: their matches share a variable. (Not started)No archetype declared yet. Brief: at least 2, coupled through a shared variable.
Mark one delay: the September before places held back change the share. (Not started)No effects yet. Brief: exactly 1 delay.
Brayford's share of the villages' starters is drawn once, on both archetypes' loops. You check this.
Beside the diagram, one sentence says what each rise in Brayford's aim does to the pupils on roll at Kelsham. You check this.

Always checked

Each loop's kind matches its − effects (Not started)No loop declared yet.
Loop labels are unique and match their kind (Not started)No loop declared yet.
Every effect has a polarity (Not started)No effects yet.
The structure matches each declared archetype (Not started)No archetype declared yet.
No variable's name breaks the more test's three rules (Not started)No variable drawn yet.

Written feedback

Turn on written feedback

Export

One defensible answer

It opens here once you've drawn at least one variable in the lab above. Your attempt saves as you draw.

Sources

Daniel H. Kim, Systems Archetypes I: Diagnosing Systemic Issues and Designing High-Leverage Interventions (Pegasus Communications, 1992, 2000): Kim’s name for eroding goals, why lowering a target closes a gap sooner than corrective action, and targets set from outside, pp. 12–13; escalation, its figure-8, and the relative measure between the sides, pp. 14–15; growth and underinvestment, standards lowered as performance falls, and capacity judged against demand, pp. 6 and 18–19.
Jennifer Kemeny, “‘Accidental Adversaries’: When Friends Become Foes”, The Systems Thinker 5, no. 1 (February 1994): 5–6, on partners whose corrective measures obstruct each other, Procter & Gamble and Walmart, and where the leverage lies.
David C. Lane, “Can We Have Confidence in Generic Structures?”, Journal of the Operational Research Society 49, no. 9 (1998): 936–947, on archetypes as metaphorical thinking (from its abstract).