logo ELT Concourse teacher training
Concourse 2

Reporting verbs in EAP

reporting verbs

This guide is not to do with reported or indirect speech and will not cover the simpler reporting verbs such as say, tell, exclaim, ask, enquire etc.  For a guide to those matters click here for the guide to indirect speech in general English.
What follows assumes that the rules for reporting direct speech are familiar.
The following is most relevant to formal academic writing although the concepts are generalisable.  It is, therefore, of most interest to people teaching EAP (English for Academic Purposes).


why

Why is this important?

Consider this:

Even the most original academic paper integrates facts, ideas, concepts, and theories from other sources by means of quotations, paraphrases, summaries, and brief references.
(Campbell 1990, in Jordan, 1997:171)

It is unlikely, but not impossible, that learners of English for Academic Purposes will be producing the most original academic paper so the ability to insert summaries, paraphrases and citation appropriately, using the right reporting verb, is a key writing skill.  Not least, of course, to avoid any whiff of plagiarism.  Understanding the implications of reporting verbs is also a key reading skill to identify slant and angle.

Reporting verbs can be classified and presented in a number of ways and we can also, as we shall see, combine elements of the analysis to make the area accessible to learners.


attitude

Classifying reporting verbs: attitude

Reporting verbs are subtle but powerful signals of the writer's attitude to the message being sent by a paraphrase or citation.
Compare, for example:
    Guru states that ...
    Guru presumes that ...
    Guru claims that ...
    Guru suggests that ...
    Guru mentions that ...
    Guru hypothesises that ...

One way to classify such verbs is to arrange them on a cline from tentative, through neutral to assertive.  Something like this:

reporting verbs 

There are obvious problems with this, not least that opinions may differ concerning the exact connotation each verb carries.
Such a classification does, however, have some classroom utility because it alerts learners to the connotations that the verbs carry and may help them avoid inappropriate choices.


function

Classifying reporting verbs: function

Reporting verbs are used for a range of communicative functions.  Presenting them from this standpoint, we can get something like:

reporting verbs functions

This sort of presentation and analysis sits well with a communicative approach because it focuses clearly on the communicative value of the verbs.

However, one obvious problem is that some verbs can be synonyms and some can perform multiple functions.  There is not a great deal of difference in meaning between, e.g.:
    Guru proposes that we ...
and
    Guru suggests that we ...
However, suggest is also polysemous in a way that propose is not so we can also have:
    The data suggest that ...
but not
    *The data propose that ...
Another obvious issue is that the diagram ignores the strength of the reporting verbs and may allow learners to believe that all the verbs in each category carry the same connotations.  As we saw above, that is not the case.


form

Classifying reporting verbs: form

The final way to classify these verbs in this guide is by grammatical and lexical form.  We need to look at concepts of collocation and colligation here, especially the latter.  This might result in this kind of analysis:
colligation

but there are problems with that, too, some colligational, some collocational:

  1. Transitivity
    1. Some verbs in the lists are both transitive and intransitive.  We can have
          Guru concedes the point that ...
          Guru concedes that ...
          Guru questions the conclusion
          Guru questions whether ...
          Guru proposes a solution
          Guru proposes that ...
    2. Some verbs are only intransitive.  We can have:
          Guru observes that ...
      but not
          *Guru observes the conclusion that ...
      or
          *Guru theorises a solution
    3. Some verbs are only transitive.  We can have:
          Guru recommends a solution
      and
          Guru recommends that ...
      but not
          *Guru discounts that ...
      or
          *Guru discusses that ...
  2. Appropriacy of subject:
    Some of these verbs will collocate with inanimate subjects, some with animate only and some with both.
    1. Animate subjects can be used for most of them but some can also take inanimate subjects.  We can have, e.g.,
          The study shows ...
          Guru shows ...
          The evidence indicates ...
          Guru indicates
      ...
      etc.
    2. Some may only have animate subjects.  We can have:
          Guru hypothesises ...
          Guru maintains ...

      but not
          *The evidence comments ...
          *The data say ...
          *The evidence states ...
      or
          *The facts allege ...
    3. Some are open to metaphorical use, assigning an action to an inanimate subject normally reserved for people (pathetic fallacy):
          The study argues ...
          The facts imply ...
          The findings argue for ...

      and some are not:
          *The evidence describes ...
          *The findings believe ...
  3. Omitting that
    The theoretical distinction here is between what are called bridge verbs and non-bridge verbs.  Many simple reporting verbs such as say, tell, think, know, write, claim and hear are bridge verbs and it is perfectly in order to omit the word that when they are followed by a clause so we allow both:
        He said that the results are unsatisfactory
        Guru thinks that the solution is to ...
        She claims that the analysis is flawed
    etc. and:
        He said the results are unsatisfactory
        Guru thinks the solution is to ...
        She claims the analysis is flawed
    Many find (that) the sentences without that are more stylistically acceptable.
    However, in academic writing simple verbs like these are often avoided for the sake of style or precision as we have seen above and with the less frequently used verbs, sometimes called non-bridge verbs, omitting that often results in some clumsiness.  All of the following, for example:
        Guru confirms the results are reliable
        He acknowledge the experiment was flawed
        The findings indicate there is a need for ...

        She explains findings are provisional
    are better expressed with that retained as in:
        Guru confirms that the results are reliable
        He acknowledge that the experiment was flawed
        The findings indicate that there is a need for ...
        She explains that findings are provisional

combine

Combining the analyses

We can weave aspects of all three analyses together to produce quite sophisticated analysis.  For example, if we combine attitude with function we can produce something like:

disagree

suggest

and we can do similar things with many of the other functions.
It is also possible to combine functional and formal analyses:

form

It is even possible to go one step further and combine all three analyses but, at that stage, the data start to get impenetrable.


tenses

Tense, aspect and voice

Reporting verbs are frequently used in the present simple so that is not difficult to teach.  Perversely, some lists of reporting verbs put all of them in the present simple 3rd person which is misleading at best.  Two other tenses are also frequently used and multiple authorship is common.  Here are examples of all three:

  1. Present simple
    This is the most frequent form:
        Guru notes that ...
        The data imply ...
        In that paper, Guru and Mentor propose ...
  2. Past simple
    This is frequently used for sources which are older and have become seminal authorities in some way.  For example:
        Guru (1949) identified ...
  3. Present perfect
    This is used a) when the writer needs to emphasise the present relevance of a source or b) when the sources are varied and (sometimes) not individually identified.  It is often used in the passive voice but need not be.  For example:
        It has been noted (Guru, 2016) that ...
        Guru (2010) has discovered that ...
        Guru and Mentor (2000) investigated the structure of these complex substances and have shown that they are ...
        It has often been asserted that ...
  4. Passive clauses and the dummy it
    There are time when the writer may wish to disguise the source of a statement, when the statement needs no reference or when the source is unknown or so ubiquitous in the field that citing individual authors would be too cumbersome.
    In these cases, a passive construction (usually in the present simple or perfect) and a dummy it come to the rescue as in, e.g.:
        It has been established that ...
        It is reckoned that ...
        It has often been noted that ...

    and so on.
    Sometimes, the source is included in such constructions as in, e.g.:
        It has been shown (Guru, 1998, Mentor, 1999 and others) that ...
    In the last case, the by-phrase conventionally used to indicate the agent is unnecessary because it is implied by the citations in brackets.
    The dummy it is often avoided, too, especially when the patient of a passive clause is inanimate so we may encounter, for example:
        The results have been disputed in the literature
        The search for a solution has, until now, been abandoned as ...
        Many reasons are suggested, including, for example, ...
        The law has been criticised as being ...
        This proposal has been rejected as ...

    and so on.
    (There are two short exercises in the learners' section of this site on using passive reporting verbs.  Click here to open the index in a new tab.)

class

Teaching and learning issues

Here's an incomplete list of over 150 of the verbs commonly used to report the work of others in academic writing.  It is unclassified by any of the three analyses considered above but a classified version is available from the next link.

accept
acknowledge
add
admit
advise
advocate
affirm
agree
alert
allege
allow
analyse
announce
appraise
argue
articulate
assert
assess
assume
assure
attack
aver
believe
blame
cast doubt on
challenge
characterise
claim
clarify
classify
comment
concede
conclude
concur
confirm
congratulate
consider
contend
contradict
contribute
counter
criticise
critique
debate
decide
declare
defend
define
demonstrate
deny
depict
describe
determine
develop
disagree
disapprove
discard
disclaim
discount
discover
discuss
dismiss
disprove
disregard
doubt
emphasise
encourage
endorse
estimate
evaluate
examine
explain
explore
express
fault
feel
find
forbid
forget
forgive
guarantee
guesshighlight
hold
hope
hypothesise
identify
ignore
illustrate
imagine
imply
indicate
infer
inform
inquire
insist
interpret
intimate
investigate
justify
know
list
maintain
mention
note
object (to)
observe
oppose
outline
persuade
point out
portray
posit
postulate
praise
predict
present
profess
promise
propose
protest
prove
provoke
put forward
query
question
realise
reason
rebuff
recognise
recommend
refer
refute
reject
remark
remind
report
restate
reveal
say
scrutinise
show
speculate
state
stress
study
substantiate
suggest
support
suppose
suspect
take into consideration
take issue with
tell
theorise
think
throw light on
underline
understand
urge

This list is also available as a PDF document for reference.
A list which is categorised by function is available by clicking here.

Clearly, presenting learners with a list like this is not going to be a very productive approach.  Somehow we have to help our learners eat the elephant so a piecemeal approach is the only practical way forward.

Before you can begin, you need to make a selection of the reporting verbs which will form the target of a teaching sequence.  This site can't do that for you because there are some variables to consider first:

approach

Approach

You can approach the verbs from any of the three analyses above by considering attitude, function and form separately.  A combined approach is often productive providing the number of target verbs is limited and carefully selected.

For example:

  1. Start with a shortlist of verbs which function to state what an author believes (say, hold, assert, believe, claim, declare, maintain etc.) and introduce them in context to show the colligations (see above for that).  At lower levels, some focus on tense, aspect and voice is appropriate at this stage.
  2. Move on to presenting them on a cline from tentative, through neutral to assertive (see above).
  3. Then, once the function, form and meaning are clear, learners can proceed to practising them.  A simple way to do that is to present a text which only uses a simple verb like say and get learners to replace the verb with something more appropriate.  It could look something like this:
    Jones (1964:20) says that ... and says it is important to ... but Smith (1990:85) says that this is not true and he says work by Robinson (1990) says that ...
    More recently, Smith now says that what the study said was not fully correct.  He now says that what Jones said is important.

    could become something like:
    Jones (1964:20) asserts that ... and emphasises that it is important to ... but Smith (1990:85) avers that this is not true and he points out that work by Robinson (1990) reveals that ...
    More recently, Smith now
    concedes that what the study showed was not fully correct.  He now allows that what Jones noted is important.
  4. Finally, the learners can move on to writing or improving their own texts using the limited range you have introduced.
  5. The procedure can be duplicated with another target function and set of verbs until the learners are able to deploy a wide range of reporting verbs accurately, with attitudinal awareness and for functions they can clearly identify.


Related guides
reported or indirect speech the general guide to the area
verbal processes for a general guide to what verbs do
verb and clause types for a guide to understanding verbs and their structures
EAP index for links to other guides in the area


Reference:
Campbell, in Jordan, RR, 1997, English for Academic Purposes, Cambridge: Cambridge University Press