§ 145. An alphabet, however, may be sufficient, and yet imperfect. It may err on the score of inconsistency. Let there be in a given language two simple single sounds, (for instance) the p in _pate_, and the f in _fate_. Let these sounds stand in a given relation to each other. Let a given sign, for instance, פ (as is actually the case in Hebrew), stand for the p in _pate_; and let a second sign be required for the f in _fate_. Concerning the nature of this latter sign, two views may be taken. One framer of the alphabet, perceiving that the two sounds are mere modifications of each other, may argue that no new sign (or letter) is at all necessary, but that the sound of f in _fate_ may be expressed by a mere modification of the sign (or letter) פ, and may be written thus פּ, or thus פ′ or פ`, &c.; upon the principle that like sounds should be expressed by like signs. The other framer of the alphabet, contemplating the difference between the two sounds, rather than the likeness, may propose, not a mere modification of the sign פ, but a letter altogether new, such as f, or φ, &c., upon the principle that sounds of a given degree of dissimilitude should be expressed by signs of a different degree of dissimilitude. Hitherto the expression of the sounds in point is a matter of convenience only. No question has been raised as to its consistency or inconsistency. This begins under conditions like the following:--Let there be in the language in point the sounds of the t in _tin_, and of the th in _thin_; which (it may be remembered) are precisely in the same relation to each other as the p in _pate_ and the f in _fate_. Let each of these sounds have a sign or letter expressive of it. Upon the nature of these signs, or letters, will depend the nature of the sign or letter required for the f in _fate_. If the letter expressing the th in _thin_ be a mere modification of the letter expressing the t in _tin_, then must the letter expressive of the f in _fate_ be a mere modification of the letter expressing the p in _pate_, and _vice versâ_. If this be not the case, the alphabet is inconsistent. In the English alphabet we have (amongst others) the following inconsistency:--The sound of the f in _fate_, in a certain relation to the sound of the p in _pate_, is expressed by a totally distinct sign; whereas, the sound of the th in _thin_ (similarly related to the t in _tin_) is expressed by no new sign, but by a mere modification of t; viz., th. § 146. A third element in the faultiness of an alphabet is the fault of erroneous representation. The best illustration of this we get from the Hebrew alphabet, where the sounds of ת and ט, mere _varieties_ of each other, are represented by distinct and dissimilar signs, whilst ת and תּ, sounds _specifically_ distinct, are expressed by a mere modification of the same sign, or letter.