In comparison with epistemic modality, deontic modality reveals not just a matter of degree, but also involves the attitude of the modal source.For instance, permission is deontic possibility with a presupposition of the agent’s willingness to carry out the action, and obligation is deontic necessity with a presupposition of the agent’s unwillingness.Thus, the relationship between epistemic and deontic modality can be listed as follows in Table 2.1.
From Table 2.1, it can be found that permission and obligation are not ordered as stronger or weaker, but as having the opposite values of willingness or unwillingness.If intermediate deontic values like advisability (e.g.should）or suggestion (e.g.could）are taken into account, they may form a zero point relative to which permission and obligation form the positive and negative poles respectively.Consider the following:
(21）She can come here.
(22）She needn’t come here.
The scalar quantity of deontic modality vs.that of epistemic modality in Examples (21）and (22）can be compared and differentiated by the following formalization formula:
≠ deontic (◇p +> ~ □p）= deontic (◇+w p +> ~ □-w p)
In this formula, the following points should be known:
A.can = allowed;
B.needn’t = not obliged;
C.◇p= possible;
D.□p= necessary;
E.~ is a negation sign;
F.“≠ deontic” means “not deontic” or “epistemic”;
G.“= deontic” means “deontic”;
H.“+w” indicates “willingness”;
I.“-w” indicates “non-willingness”;
J.“+>” is the sign of “entail”.
This formula is very complex.The modal senses of the MVs (i.e.can and needn’t）in Examples (21）and (22）can be interpreted respectively in the following way:
A.can means that the subject is allowed to come, but he/she does not necessarily come; in this sense, can denotes the epistemic modality of judgment or prediction.
B.can means the subject is allowed and willing to come, and it is necessary for him/her to come; in this sense, can denotes the deontic modality of permission.
C.needn’t means that the subject is allowed not to come, and it is possible for him/her not to come; in this sense, needn’t denotes the epistemic modality of judgment or prediction.
D.needn’t means that the subject is allowed not to come, and he/she is willing not to come; in this sense, needn’t denotes the deontic modality of volition or permission.
The previous research on epistemic and deontic modality demonstrates that the two types of modality are not differentiated on clearly-cut boundaries, but situated on a continuum or cline if they are quantified in terms of truth conditions.In another word, epistemic modality can be converted to deontic modality when considered from the other angle, and vice versa.Also, in some cases modal values may change if some of the MVs on the same value level are negated, as Examples (17）and (18）show.Thus, the application of scalarization to the study of modality is complex and evasive rather than a matter of going through a set of formalities.This suggests that a single modal device can be pluralistic in functions and a single modal function can be expressed by more than one modal device.
2.1.2.3 The form-function equation method
The third method belonging to the approach of truth-condition correspondence is the form-function equation method.Perkins (1983: 89-93）argues that all MAs primarily express epistemic modality, and that most of the MAs are closely related to their equivalent adjectival forms.Take the following for example:
(23）He’s arguably the most promising newcomer.(= It’s arguable that he’s the most promising newcomer.)
(24）Allegedly, he’s the richest man in the world.(= He’s alleged to be the richest man in the world.)
However, Perkins (1983: 90）is cautious in stating that MAs are equivalent to their adjectival forms:
Modal adverbs differ from modal adjectival expressions and modal nominal expressions in that they are ‘peripheral in clause structure’, i.e., they are specifically sentence adverbs.Their peripherality can be seen in the fact that a modal adverb can appear in a number of different positions within the clause without affecting the meaning relation between the clause and the adverb (although the overall meaning of the resulting sentence may differ).
In stating the effects of the positions of MAs on their adjectival equivalence Perkins (ibid）reminds people that even tone groups in speaking or commas in writing may affect the decoding of MAs by means of adjectives.Consider the following:
(25）Mr.Smith possibly has been trying to phone us.
(26）Possibly Mr.Smith has been trying to phone us.
Example (25）can be paraphrased as: It’s possible that Mr.Smith has been trying to phone us.In contrast, Example (26）(without a comma, i.e.possibly does not have a separate tone group, and with a falling intonation on “Smith”）might be interpreted as: Possibly Mr.Smith, as opposed to Mr.Jones, has been trying to phone us.
Also, in paraphrasing MAs in terms of adjectives some “fine-tuning” is required.Greenbaum (1969: 108）studies the correspondence between MAs and adjectives, and presents the following example based on SEU (the Survey of English Usage):
(27）Somewhat unexpectedly, his prose doesn’t rise as readily to the florid possibilities of North Oxford.
