Archetype extraction

Jump to navigation Jump to search

List of methods / Narrative methods

Archetype Extraction is the most common application of the Social Construction of Emergent Properties. Archetype Extraction is a method of articulating and visualising common cultural understandings. It is an activity commonly run in parallel, or as part of a larger project. Specifically, Archetype Extraction helps to identify these shared views and re-create them in the form of Archetypal characters with the aid of an on-site cartoonist. The activity results in a set of Archetypes unique to the context of the participants. These Archetypes can then be used in various initiatives such as branding, communications or strategy. Archetypes that are extracted from the anecdotes told naturally in a community resonate: they have bite. Archetypes have a long and honourable tradition in story telling. The Greek and Norse Gods are all archetypes: they represent extreme aspects of human behaviour and stories about them collectively allow humans to reflect on their own condition.

Typical Uses

  • Aid in conflict resolution between different groups with opposing views
  • An effective way to induct new recruits into a community, team or organisation
  • Sensitise team for boundary transitions, to improve anticipatory awareness – tools which alert us to changing circumstances in our environment and organisations
  • Organising training materials around tacit knowledge and skills
  • An alternative approach to understanding customers and their needs (product design, user requirements)


Collect anecdotes from people through Anecdote Circles or other ways. Ideally all of those involved in the sharing of the narratives should be present in this Archetype Extraction workshop.

Things You'll Need

  • A collection of anecdotes
  • Per participant: hexie pads (at least 4 different colours) and a marker
  • A large surface (table or wall) where hexies can be clustered, and space for participants to move around


Task Comment
FIRST STAGE: In a workshop environment, a representative sample of the target group is brought together. They take some time to familiarize themselves with the collected narratives, and are then split randomly into several sub-groups by the facilitator. The facilitator should work out in advance how to maximize the diversity amongst the various sub-groups.

A recommended minimum number of participants would be 8, while a maximum would depend on the facilitator’s ability.

Each participant should be provided with a hexie pad and a marker.

Every participant is now asked to identify all the characters that they see in the narratives. Each character identified by a participant is to be:

  • Written as an adjective-noun combination (ie “bumbling goalkeeper”)
  • Placed as one item per piece of hexie
  • Have all the hexies placed on a surface (usually a wall, large whiteboard or a table).
Ask participants to randomly mix their hexies on the surface. During this process of identification of characters, participants work individually within their sub-groups and do not see the results of the other sub-groups.
Have participants group all the hexies into clusters on a like-for-like basis (ie similar characters should be clustered together). Ensure that there are no less than 6 or more than 10 clusters.

Give participants a few more minutes if necessary, to ensure that the number of clusters are in the above range.

When the clusters of hexies have been finalized, have participants agree on a name that would be suitable for each cluster. As a facilitator, you may wish to suggest that an “adjective-noun” combination be used for the cluster names.
SECOND STAGE: Studying the clusters which have been created and named in the first part of the exercise, each participant now works to identify the positive and negative attributes of each cluster.

Each attribute is an adjective (at least one word, up to four or five is fine) and is written on a single hexie. This hexie should then be placed around its corresponding existing cluster. Each cluster should have a balance of positive and negative attributes.

If there had been sub-groups earlier, it is up to the facilitator to decide whether to mix the sub-groups or use the same sub-groups.

For a cluster of character items, the facilitator asks the participants to identify: “What would the character’s best friend say about this character?” and “What would the character’s worst enemy say about this character?” Facilitator needs to emphasize that digital photos can be taken at this point so that participants may be able to track back, if they want to.

Now, remove the titles and original items from each cluster and place the removed items in separate stacks away from the board.

There will now only be the attributes left. Participants will completely randomize the attributes.

The original items are kept in separate stacks, so that the group may track the evolution of items toward the final output.
Have participants form clusters with the hexies again.

This time, each cluster should have a balance of positive and negative attributes.

Facilitator to ensure again that there are no less than 6, or more than 10, clusters of hexies.
Once again, have participants give each cluster a name. Once again, facilitator to suggest that an “adjective-noun” combination may be most effective.
CARTOONIST : Now, the cartoonist will start drawing visual representations of the various Archetypes which have been extracted. As they are drawing, the participants should actively participate in giving their opinions on various components of what the Archetype should look like – ie from gender, attire, environment to facial expression. It is important to use a cartoonist who draws a faithful representation of the Archetypes described. The cartoonist should not impose any artistic/other notions of what an Archetype should look like.

Do's and Dont's

  • Why is there a minimum and maximum number of groups? To keep participants focused and the exercise useful and manageable. From experience, 6-10 people seem to work best considering we usually have approximately 100-200 narratives to work with for each exercise
  • How do I manage someone who seems to be dominating the discussion? Rotate the person through a different group, so that they do not have too much of an influence on the formation of a group of items or attributes.
  • How long will this method take? Depending on the size of the group and the number of narratives, it may run from an hour to several hours. With a group of 30 people handling 120 narratives, the method takes an hour.
  • Tell me more about names for the groups. Don't let people give names to archetypes that refer to real people. Use animal names that connote metaphorical characteristics (badger, lemming), or pop-culture names (superman, queen), or generic role names (recluse, he-man). If they insist on naming a real person, ask what it is about that person that makes them unique, then ask them to rename it to that. Explain the differences between archetypes and stereotypes, but only do this when the people are doing final clustering and naming, and only if they are ending up with stereotypical names.


Kurtz, C. & Snowden, D. (2006) Bramble Bushes in a Thicket. Snowden, D. (2005) Archetypes as an Instrument of Narrative Patterning. Snowden, D. (2005). Simple but not Simplistic: the Art and Science of Story. Snowden, D. (2005). “Stories from the Frontier”. E:CO Issue Vol.7 Nos. 3-4 2005 pp. 155-165.