AN ENUMERATION OF 77 fE SCIENCES 19

К оглавлению1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 
136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 
153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 
170 171 172 173 174 175 176 177 178 179  181 182 183 184 185 186 
187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 
204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 
221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 
238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 
255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 
272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 
289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 
306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 
323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 
340 

 

it pertains to every other, and no other can be pursued

without it.

 

Nevertheless, another science can advance a claim seem-

ingly as unanswerable in another respect as is the claim of

Mathematics, as just stated. Xo science can claim to be

absolutely primary which has to depend on another science

for explanation and comprehension. Mathematics is a science

of "number"; but what is "number"? Moreover, numbers are

alike or not unlike, and a perception of " likeness and unlike-

ness " was declared, in our introductory chapter, to be at the

base of all the sciences. What, then, it must be further asked,

is " likeness " ? May not the science which can solve these

riddles justly claim to underlie, and be prior to, the science

of Mathematics?

 

The idea of " number " implies comparison, together with a

recognition that the things compared are similar, and yet not

identical. Things which are quite dissimilar such as, e.g.,

"a violet blossom" and "a fall in consols" cannot be said

to be two, unless it be two expressions or two thoughts in

which respects they are alike. But the idea of number, inas-

much as it recognizes things as similar but not identical,

implies many things besides similarity and identity. In every

perception of number there are, and must be, latent the ideas

of " existence," " distinction," " similarity," " unity," and

' truth," as a little reflexion will show. Thus, to say " there

are two sheep," implies that they are not merely imaginary,

but that they actually exist ; that they are not seen double

by some optical delusion, but are really distinct ; that they

are certainly both sheep and not one of them a goat i.e. t that

they are similar, and that they have that unity of nature

which we have just seen to be necessary in order that they

should be susceptible of numeration, and finally the assertion

implies that the thought of the assertion corresponds with

objective reality, that is, it implies truth.

 

 

it pertains to every other, and no other can be pursued

without it.

 

Nevertheless, another science can advance a claim seem-

ingly as unanswerable in another respect as is the claim of

Mathematics, as just stated. Xo science can claim to be

absolutely primary which has to depend on another science

for explanation and comprehension. Mathematics is a science

of "number"; but what is "number"? Moreover, numbers are

alike or not unlike, and a perception of " likeness and unlike-

ness " was declared, in our introductory chapter, to be at the

base of all the sciences. What, then, it must be further asked,

is " likeness " ? May not the science which can solve these

riddles justly claim to underlie, and be prior to, the science

of Mathematics?

 

The idea of " number " implies comparison, together with a

recognition that the things compared are similar, and yet not

identical. Things which are quite dissimilar such as, e.g.,

"a violet blossom" and "a fall in consols" cannot be said

to be two, unless it be two expressions or two thoughts in

which respects they are alike. But the idea of number, inas-

much as it recognizes things as similar but not identical,

implies many things besides similarity and identity. In every

perception of number there are, and must be, latent the ideas

of " existence," " distinction," " similarity," " unity," and

' truth," as a little reflexion will show. Thus, to say " there

are two sheep," implies that they are not merely imaginary,

but that they actually exist ; that they are not seen double

by some optical delusion, but are really distinct ; that they

are certainly both sheep and not one of them a goat i.e. t that

they are similar, and that they have that unity of nature

which we have just seen to be necessary in order that they

should be susceptible of numeration, and finally the assertion

implies that the thought of the assertion corresponds with

objective reality, that is, it implies truth.