A Clear View of Inductive Learning

December 2015

I’m yet to find a really accurate, original, and dare I say it, interesting, explanation of inductive learning so I thought I’d write one. Firstly, it’s necessary to examine inductive and deductive reasoning. Reasoning is how we structure our thinking to arrive at answers and truths, using facts and logic as the basis for higher conclusions.

Politeness and humour are the cement between the bricks.

Inductive reasoning means conjecturing an answer using the information you have as platform for further inference.

Deductive reasoning means finding the answer contained in the information you have been given.

For example:

Out of these five fruits: Banana, Apple, Tomato, Peach, Pear, the tomato is the odd one out because it is the only one eaten or cooked with salt.

The others can be, but in reality almost always aren’t.

Out of these five fruits: Banana, Apple, Tomato, Peach, Pear, the tomato is different as it can also be classed as a vegetable.

This is a basic fact and cannot be falsified.

You can see from the examples that inductive reasoning takes far more imaginative and cognitive work than simple Aristotelian recall.[1] In other words, inductive reasoning is a deeper form of reasoning that involves speculating, calling on a wider field of information, and taking an intellectual risk. While deductive reasoning involves a certain degree of playing it safe.

  • There is usually more room for debate with an inductive argument as it gives a likely answer or outcome, whereas a deductive argument gives the matter of fact answer or outcome.
  • Inductive arguments are measured in plausibility, whereas deductive arguments are either true or false.
  • Inductive reasoning seeks to provide a strong argument, above a valid one. While deductive reasoning seeks to provide a valid argument above a strong one.

A classic example of deductive reasoning.

I feel there is not enough promotion of reasoning in schools and academia generally. We use reasoning every day but formal and deep argument I believe is something of a forgotten skill.[2] Instead, people are encouraged to simply review and reference literature on a subject and use hedged and equivocal language to arrive at a typically tame conclusion.

In the same vein, inductive learning has become less favoured in education than its easier, less original brother, deductive learning. That is, the promotion of knowledge over understanding.[3] Both are required in education and in life, but the two call on very different thought processes.

In Language Learning

Lying at the root of reasoning in language learning are the questions we ask our learners. For instance, a comprehension question about a text is deductive, while asking the reader’s opinion is inductive.

“The wise man doesn’t give the right answers, he poses the right questions.” ~ Claude Levi Strauss

Deductive vocabulary teaching

  1. Teacher shows students some flashcards and then reviews them. — Learners are relying on their short term memory.
  2. Teacher: “P = Q.” ... “What does P mean?” — Explicit teaching and translation.
  3. Teacher: “Environment is spelt e, n, v, ...” — Spoon feeding.

Inductive vocabulary teaching

  1. Teacher shows students pictures and elicits the names, before giving clues and finally teaching them. — Gambling on the learners’ long term memory first and giving guidance second.
  2. Teacher: “R = S, so what do you think P means?” — Using context to elicit.
  3. Teacher: “Try to spell environment.” — Challenging the learner, openness towards error-making.

Deductive grammar teaching

  1. Teaching specific rules of syntax.
  2. PPP — Starting with a rule and practising specific examples.
  3. Top down instruction.

Inductive grammar teaching

  1. Teaching general principles of grammar.
  2. Teacher introduces several examples of a form, students are asked if the can see/generalise the rule.
  3. Bottom up reasoning.

[Don’t get me started on pron. That’s for another post...]

The inductive approach has in many contexts gone out of fashion. It is viewed as a relic from the Direct Method where learners were felt to be overly challenged and often struggled to grasp a concept.[4] Inductive learning is thus considered an uphill approach. Despite the Direct Method being such a natural method; attempting to mirror the L1 acquisition process, it is also natural that teachers are very quick to provide the answers and make learning as easy as possible. Sometimes people pay lip service to thorough practice, but in reality take the easy route of teaching at the students and to the test.[5]

Teacher-centredness and authority really hinder communicative development.

Deductive elicitation does indeed have its limits but inductive elicitation has a lot more mileage in constructing understanding. The basic premise of this is that as language teachers, we are teaching a skill more than we are teaching a knowledge. Ergo, the emphasis ought to be on the skill of communicative risk-taking above the ability to recall declarative facts. The latter relies on tedious repetition, the former on discovery. Knowing a rule or a word is something altogether different than being able to use it in normal interaction. For that, we do need repetition but more so we need independent and somewhat unscaffolded practice. The truth is, people surprise themselves and grow when they try to do something new.[6]

If language is taught explicitly it creates competence, but if taught implicitly it engenders performance.

Inductive reasoning and learning requires people to make generalisations. These are often wrong, but that shouldn’t be an issue in a language classroom. The wrong answer is fodder for further reasoning and talk.[7] Communicative risk taking is so important to develop in learners. When a learner does not know how to say a word she sees on the page, instead of stopping and asking how it is said, which is the default reaction for many. If the learner just attempts the word, fifty percent of the time they get it right! And of the remaining fifty percent, half the time they get it half right! People surprise themselves and grow when they try to do something new. This is how we make quick learners out of people and dispel a deficit mentality in teachers. Inductive approaches like this make better real world communicators out of us too.

Some ways to encourage inductive learning

  • Use wh questions rather than the inherently deductive and reductive yes/no questions
  • Demonstrate, don’t explain
  • Elicit everything — play dumb
  • Don’t facilitate — employ silence
  • Employ self correction — instil an attitude-shift regarding mistakes
  • Task Based Learning & problem solving, as opposed to PPP
  • Encouraging generative language such as paraphrasing & reported speech as well as divergent/unscripted role plays
  • Teach vocabulary implicitly through contextual approaches as opposed to direct translation or dictionaries
  • Use plenty of examples, metaphors, and analogies
  • Make connections between things
  • Play the odd one out game
  • Play family fortunes
  • Avoid multiple choice testing in favour of short-answer and essay testing


1. Aristotle was the first to comprehensively outline deductive reasoning. As well as this, his theories of classification universals all relate to deductive inference.

2. Which is why I’ve written a book on the subject.

3. Genuine understanding only comes from doing

4. For an overview of the Direct Method, its value and its failings, read this.

5. One of these days I’ll do a post on the Silent Way.

6. For example: these assertions on the website of Transparent Language Learning deprecating inductive learning.

7. See the activity Attributes.