WEBVTT 1 00:00:01.280 --> 00:00:05.440 Welcome. This is Marcia for RADIOI, and today I will 2 00:00:05.519 --> 00:00:10.080 be reading National Geographic Magazine dated January twenty twenty six, 3 00:00:11.599 --> 00:00:14.919 which is donated by the publisher. As a reminder, RADIOI 4 00:00:15.039 --> 00:00:17.519 is a reading service intended for people who are blind 5 00:00:17.600 --> 00:00:20.320 or have other disabilities that make it difficult to read 6 00:00:20.359 --> 00:00:24.199 printed material. Please join me now for the continuation of 7 00:00:24.239 --> 00:00:29.640 the article I began last time, entitled Decoding the Lost 8 00:00:29.960 --> 00:00:36.640 Scripts of the Ancient World by Joshua Hammer. Based on 9 00:00:36.799 --> 00:00:39.479 what they knew of other ancient forms of writing, they 10 00:00:39.520 --> 00:00:43.079 also theorized that the script could have been built upon 11 00:00:43.240 --> 00:00:46.880 an underlying language that in some form might still be spoken. 12 00:00:47.280 --> 00:00:49.119 This is one of the great tools used in the 13 00:00:49.159 --> 00:00:53.399 detective work of ancient philology. Figure out the sounds that 14 00:00:53.439 --> 00:00:56.399 the characters make, string them together, and you might conjure 15 00:00:56.439 --> 00:01:00.000 up the meanings as well. By comparing scraps and pieces. 16 00:01:00.240 --> 00:01:04.120 By triangulating what is known with what his mysterious, researchers 17 00:01:04.159 --> 00:01:09.920 can inch their way toward clarity. After some hunting, Mahadevan 18 00:01:10.359 --> 00:01:15.040 and Parpola agreed that the Indus Script was likely built 19 00:01:15.079 --> 00:01:20.480 on Proto Davidian, a nascent form of language that many 20 00:01:20.519 --> 00:01:25.120 philologists believe dominated the Indian sub continent. During the early 21 00:01:25.439 --> 00:01:29.480 Bronze Age, the ancient language was lost, but vestiges remain 22 00:01:29.640 --> 00:01:34.959 in modern Tamil and other Southern Indian tongues. Parpola then 23 00:01:35.120 --> 00:01:37.760 zeroed in on the most prevalent sign in the script, 24 00:01:38.079 --> 00:01:40.680 a fish like character that the professor believed was a 25 00:01:41.079 --> 00:01:46.519 logogram in Tamil. Parpola knew the word for fish is min, 26 00:01:47.079 --> 00:01:51.000 but min has a second meaning star. All early scripts 27 00:01:51.040 --> 00:01:55.760 had the reebis principle. Parpola, now retired and living in Helsinki, 28 00:01:55.840 --> 00:01:58.920 told me using a pictogram or symbol for its sound, 29 00:01:59.000 --> 00:02:02.920 not its meaning. For example, in the world's first writing system, 30 00:02:03.000 --> 00:02:09.199 Sumerian cuneiform scribes combined the pictogram for barley, which has 31 00:02:09.240 --> 00:02:12.680 the phonetic value she, with the symbol for milk, which 32 00:02:12.719 --> 00:02:18.159 connotes she the sound ga to make shiga, which has 33 00:02:18.240 --> 00:02:22.319 nothing to do with barley or milk, but meant pleasing. 34 00:02:23.080 --> 00:02:27.120 Following these principles, Parpola, over several years in the late 35 00:02:27.240 --> 00:02:30.960 nineteen sixties and early seventies, searched for other clues. He 36 00:02:31.039 --> 00:02:34.439 found in the indiscript another symbol that showed a fish 37 00:02:34.639 --> 00:02:37.680 divided down the middle by a vertical line. The line 38 00:02:37.680 --> 00:02:41.800 he proposed stood for passu, the ancient Tamil word for half, 39 00:02:42.240 --> 00:02:46.599 but Passu also means green. If the fish connoted min, 40 00:02:47.199 --> 00:02:51.479 he now had Aribis Pasu min or green star, which 41 00:02:51.520 --> 00:02:55.120 he took to mean the planet Mercury. Pushing further, Parpola 42 00:02:55.240 --> 00:02:58.719 found what he thinks are ribuses for Saturn, Venus and 43 00:02:58.840 --> 00:03:02.319 other stars. He also located what he believes were a 44 00:03:02.360 --> 00:03:06.759 few purely phonetic signs. What did it all signify? Because 45 00:03:06.800 --> 00:03:10.639 the inscriptions are so short, Parpola believes that they contained 46 00:03:10.719 --> 00:03:14.560 no grammar, no full sentences, no elements of real writing. 47 00:03:15.039 --> 00:03:18.719 He posits that they never were intended to communicate messages, 48 00:03:18.800 --> 00:03:21.879 but rather were used as the markers of citizens who 49 00:03:21.919 --> 00:03:26.360 were named after celestial objects, like many rulers of the age, 50 00:03:26.479 --> 00:03:32.840 including those in Assyria and Babylonia. Of course, decoding works 51 00:03:32.919 --> 00:03:38.080 like this that rely on interpretation and speculation prompts disagreement. 52 00:03:38.639 --> 00:03:43.599 Mahadavan parted ways with his colleague on many of Parpola's readings, 53 00:03:43.879 --> 00:03:47.919 beginning with the fish sign, which Mahadivan believed had nothing 54 00:03:47.919 --> 00:03:51.039 to do with astronomy. It was he contended the sign 55 00:03:51.120 --> 00:03:54.759 for a sea nymph, a creature prevalent in Indian mythology, 56 00:03:55.199 --> 00:03:58.199 but the brevity of the inscriptions and the uncertainty about 57 00:03:58.240 --> 00:04:01.759 the language meant that neither scholar could say he solved 58 00:04:01.800 --> 00:04:06.319 the riddle. Raja Gopal, for one, thought that Parpolo was 59 00:04:06.360 --> 00:04:08.879 on the right track. A few years ago, he began 60 00:04:08.960 --> 00:04:12.120 to fixate on one seal that consisted of a row 61 00:04:12.199 --> 00:04:16.680 of five rotating swastikas, a sacred symbol in many ancient religions, 62 00:04:17.079 --> 00:04:21.920 followed by two parallel vertical lines. Parpola had claimed that 63 00:04:22.000 --> 00:04:24.680 the double line formed part of a rebus for the 64 00:04:24.759 --> 00:04:28.600 planet Venus. When combined with the fish, he said it 65 00:04:28.720 --> 00:04:33.240 made vinin min or bright star, but the swastikas remained 66 00:04:33.319 --> 00:04:37.360 an enigma. Raja Gopal told me he was consumed by 67 00:04:37.399 --> 00:04:40.519 the mystery of the sequence and thought about it constantly. 68 00:04:40.920 --> 00:04:43.040 I was going down a rabbit hole. One morning in 69 00:04:43.079 --> 00:04:47.240 November twenty twenty, while sitting in his home office in Chennai, 70 00:04:47.800 --> 00:04:51.240 he chanced upon a Nassau website that traced the trajectory 71 00:04:51.360 --> 00:04:55.079 of Venus as it moved across the morning and evening skies. 72 00:04:55.480 --> 00:04:58.519 He stopped and stared at his computer screen. The double 73 00:04:58.600 --> 00:05:02.439 snakelike pattern formed by the planet's path looked just like 74 00:05:02.480 --> 00:05:05.480 the swastika he had been studying. I got the goosebumps, 75 00:05:05.519 --> 00:05:08.959 he says, with that he was often running The inscription 76 00:05:09.160 --> 00:05:13.120 he theorized had to be a celestial omen signifying the 77 00:05:13.160 --> 00:05:16.519 completion of Venus's eight year cycle as it orbited the 78 00:05:16.560 --> 00:05:18.879 Sun and returned to the same position in the sky. 79 00:05:19.639 --> 00:05:24.240 He looked for similar astronomical depictions. Another seal he proposed 80 00:05:24.639 --> 00:05:28.639 showed the alignment of three planets. Half of the seals 81 00:05:28.639 --> 00:05:33.560 he now believes identify celestial events. Raja Gopal theorizes that 82 00:05:33.600 --> 00:05:37.000 they were written so that Indus valley priests could dispatch 83 00:05:37.120 --> 00:05:40.399 these seals to villages to provide guidance for the timing 84 00:05:40.439 --> 00:05:44.680 of crop plantings and harvest festivals. Raja Gopal now claims 85 00:05:44.680 --> 00:05:48.519 to have deciphered with confidence seventy seals out of forty 86 00:05:48.560 --> 00:05:51.720 two hundred. He finds it difficult to pull himself away. 87 00:05:52.160 --> 00:05:55.160 I will take some seal and keep obsessing, and somehow 88 00:05:55.199 --> 00:05:57.319 I get an idea and push it forward. He told 89 00:05:57.360 --> 00:05:59.800 me many times, I hit a dead end and then 90 00:06:00.000 --> 00:06:03.439 I got to spin the next hypothesis and try again. 91 00:06:04.560 --> 00:06:08.720 Not everybody accepts his interpretations and The public attention around 92 00:06:08.879 --> 00:06:14.639 raj Rajagopal's progress has only highlighted how tricky this work 93 00:06:14.680 --> 00:06:19.079 can be. At a twenty twenty three Indescript conference in Chennai, 94 00:06:19.480 --> 00:06:23.519 a fellow decipherer stood up during his lecture and ridiculed him. 95 00:06:23.720 --> 00:06:31.160 Another amateur scholar in Chennai, Sumangali Kidambi ven Katisan, says 96 00:06:31.439 --> 00:06:35.720 that Raja Kobal's celestial theory is misguided and that most 97 00:06:35.759 --> 00:06:39.839 seals were simply shipping instructions bronze age versions of d 98 00:06:40.199 --> 00:06:45.680 h L waybills. Van Katissan says he has found his 99 00:06:45.720 --> 00:06:51.319 own Dravidian based rebuses, which are nothing like Rajagopal's. One 100 00:06:51.360 --> 00:06:55.879 inscription Van Keitasan told me consists of a combination of 101 00:06:55.959 --> 00:07:00.800 logograms and phonetic signs and refers to the highlands of Afghanistan. 102 00:07:01.279 --> 00:07:05.800 He translates it to read very clever trapper velapon of 103 00:07:05.839 --> 00:07:08.720 the Triple Mountain sends by boat along the Big River 104 00:07:08.839 --> 00:07:13.040 by care to the tiller of land. The range of 105 00:07:13.040 --> 00:07:16.680 disagreement in the volume of theories underscore the intensity of 106 00:07:16.759 --> 00:07:19.759 the enduring debate, as well as the idea that would 107 00:07:19.759 --> 00:07:23.120 be code breakers may never reach a consensus. Even the 108 00:07:23.199 --> 00:07:27.879 Dravidian hypothesis itself is challenged. Some scholars argue that the 109 00:07:27.959 --> 00:07:32.759 underlying language of the Indus script is Sanskrit, the basis 110 00:07:32.800 --> 00:07:36.920 of Hindi. One decipher claimed to have identified ribuses for 111 00:07:36.959 --> 00:07:41.639 the gods Shiva and Indra mentioned in the ancient Sans 112 00:07:41.680 --> 00:07:47.040 religious texts known as the rig Veda. Like nearly everything 113 00:07:47.040 --> 00:07:51.600 in India, ethnocentrism is fueling the debate. Many Tamils want 114 00:07:51.600 --> 00:07:56.000 to bolster claims that their ancestors, not northern Hindi speakers, 115 00:07:56.399 --> 00:08:00.920 created India's first urban civilization a few hundred miles southwest 116 00:08:00.959 --> 00:08:06.240 of Chennai. Archaeologists recently found glyphs scratched into clay tablets 117 00:08:06.279 --> 00:08:10.480 at Ki Ladi, a twenty six hundred year old site, 118 00:08:10.800 --> 00:08:15.319 that match signs from Mohenjo Daro. This, the insist, is 119 00:08:15.360 --> 00:08:19.040 evidence of a connection between the Dravidian culture of southern 120 00:08:19.079 --> 00:08:22.519 India and the original settlers of the Indus Valley. Of 121 00:08:22.560 --> 00:08:25.959 the fifty different signs, we have twenty nine perfect matches. 122 00:08:26.079 --> 00:08:33.799 Ramesh Masthong, Kikhilandi's chief archaeological officer, told me as he 123 00:08:33.960 --> 00:08:39.000 escorted me around a football size field excavation pit in 124 00:08:39.240 --> 00:08:43.000 a grove of coconut homes in rural Tamil Nadu. But 125 00:08:43.080 --> 00:08:46.360 if Sanskrit is in fact the underlying language beneath the 126 00:08:46.360 --> 00:08:50.399 Indus script, Rajocobal and many others will be sent back 127 00:08:50.399 --> 00:08:53.279 to the drawing board, and that million dollar reward will 128 00:08:53.279 --> 00:08:57.279 seem further away than ever. The great hope among philologists 129 00:08:57.320 --> 00:09:00.200 today is that the breakthroughs lurking right around on the 130 00:09:00.240 --> 00:09:03.759 corner will be ushered in thanks to tools and technologies 131 00:09:03.799 --> 00:09:09.279 that their predecessors never dreamed of. Francesco Parono Katiafoko, a 132 00:09:09.360 --> 00:09:13.840 forty five year old professor of linguistics at chan Jiaotong 133 00:09:14.279 --> 00:09:17.919 Liverpool University in China, has devoted more than half his 134 00:09:18.000 --> 00:09:21.600 life to trying to decipher another Bronze Age script called 135 00:09:21.879 --> 00:09:25.120 linear A. The writing system was used by the Minoan 136 00:09:25.200 --> 00:09:29.480 civilization of Crete, the predecessor of the Mycenaean Greeks, from 137 00:09:29.480 --> 00:09:33.480 about eighteen hundred to fourteen fifty BC. The problem has 138 00:09:33.559 --> 00:09:37.240 been that the Minoans seemingly employed a language unrelated to 139 00:09:37.320 --> 00:09:41.200 any dialect of ancient Greek or apparently any other known 140 00:09:41.279 --> 00:09:44.519 language from that time period. When he began studying the 141 00:09:44.559 --> 00:09:49.519 writing in nineteen ninety nine, Parono Kaciafoko noticed that the 142 00:09:49.600 --> 00:09:52.279 system used many of the same characters as a Cretan 143 00:09:52.720 --> 00:09:58.360 syllabari called linear B deciphered by English scholar Michael Ventris 144 00:09:58.559 --> 00:10:01.639 half a century earlier. But the characters of linear A 145 00:10:01.919 --> 00:10:05.360 formed words that bore no resemblance to those of linear B. 146 00:10:06.000 --> 00:10:09.039 If you try to read the corresponding symbols, it's not Greek, 147 00:10:09.399 --> 00:10:14.919 parono Qutciufoko told me, it's Gibberish. Where parpoulus quest to 148 00:10:15.039 --> 00:10:19.480 understand the Indus script had led him to philological research 149 00:10:19.559 --> 00:10:24.399 into ancient Dravidian perono Kochiufoko took a different path. In 150 00:10:24.440 --> 00:10:29.200 twenty seventeen, he assembled a team of sixteen mathematicians, engineers, 151 00:10:29.240 --> 00:10:32.159 and linguists and designed what they called the linear A 152 00:10:32.759 --> 00:10:38.080 Decipherment Program. The team fed into the computer the vocabularies 153 00:10:38.080 --> 00:10:43.919 and grammatical structures of deciphered ancient languages across the Mediterranean 154 00:10:43.960 --> 00:10:51.879 area Ancient Egyptian, Luian, Assyro, Babylonian, Aramaic, Amharic, and Hittite, 155 00:10:51.879 --> 00:10:54.440 to name just a few, and asked the program to 156 00:10:54.440 --> 00:10:59.440 compare the words with possible transliterations of linear A. They 157 00:10:59.480 --> 00:11:03.200 hoped that the comparison would find similar etymological roots. But 158 00:11:03.240 --> 00:11:07.000 after three years the machine, as Perno Kacchavoko called it, 159 00:11:07.440 --> 00:11:10.399 had found no matches, and twenty twenty funding for the 160 00:11:10.399 --> 00:11:14.279 project ran dry. The pandemic shuttered the university and the 161 00:11:14.320 --> 00:11:17.639 project was put on hold. Computer programs he had come 162 00:11:17.679 --> 00:11:21.200 to believe aren't totally up to the task. Now, Perno 163 00:11:21.320 --> 00:11:25.080 Kutchafoko's team is using a modified version in the hope 164 00:11:25.120 --> 00:11:31.639 that it will produce better results. Artificial intelligence. Parono Kuchifoko says, 165 00:11:31.639 --> 00:11:35.159 we'll probably never be able to decipher an undeciphered language 166 00:11:35.360 --> 00:11:39.360 because it's not able to produce original thought or intuitive connections. 167 00:11:40.039 --> 00:11:42.080 What it could do, he explains, is speed up the 168 00:11:42.120 --> 00:11:47.559 deciphering process by recognizing patterns and making statistical calculations about 169 00:11:47.600 --> 00:11:52.120 the appearances of certain characters in unknown texts. In twenty 170 00:11:52.159 --> 00:11:56.960 twenty three, researchers from MIT and Google's Deep Mind tested 171 00:11:57.039 --> 00:12:02.240 AI tools on previously deciphered linear and were impressed by 172 00:12:02.279 --> 00:12:05.720 the ease with which the programmes worked. The system guided 173 00:12:05.759 --> 00:12:09.960 about sixty percent right, says Andreas Fools, an engineer at 174 00:12:09.960 --> 00:12:14.399 Berlin's Technical University, one of those attempting to decipher both 175 00:12:14.679 --> 00:12:18.120 the Indus script and the mysterious writing on what's known 176 00:12:18.200 --> 00:12:22.720 as the Feistos disc, an ancient clay saucer found on 177 00:12:22.879 --> 00:12:27.039 crete that contains a spiraling inscription in an unknown writing. 178 00:12:27.799 --> 00:12:30.840 In those experiments, the AI was primed with the knowledge 179 00:12:30.879 --> 00:12:34.200 that Linear b was an ancestor of ancient Greek. It 180 00:12:34.240 --> 00:12:37.720 turns out that AI is excellent at comparing a script 181 00:12:37.799 --> 00:12:41.120 to other members of the same linguistic family, but when 182 00:12:41.159 --> 00:12:44.679 asked to decode a script written in an unknown language 183 00:12:44.879 --> 00:12:50.279 that has no relationship to other tongues, AI still struggles. This, 184 00:12:50.600 --> 00:12:56.000 says Francois des Si, the philologist unraveling Linear Elamite, is 185 00:12:56.039 --> 00:12:59.600 because the technology, at least currently lacks the ability to 186 00:12:59.600 --> 00:13:02.600 think creatively to come up with something out of nowhere. 187 00:13:03.399 --> 00:13:06.039 On a recent morning in Francis Laire Valley, I met 188 00:13:06.399 --> 00:13:11.120 Dsse at his apartment in the city of Ongar's tussl 189 00:13:11.200 --> 00:13:14.440 haired and goateed. He was happy to describe for me 190 00:13:14.519 --> 00:13:17.120 that moment in twenty fifteen when he finally got to 191 00:13:17.159 --> 00:13:24.159 see the Mabubian families prized Kunanki for years, though his 192 00:13:24.240 --> 00:13:27.840 efforts to decipher linear Elamite had been stymied by the 193 00:13:27.840 --> 00:13:33.279 paucity of inscriptions. Unlike hieroglyphs which covered temples and tombs 194 00:13:33.279 --> 00:13:38.159 along the Nile, or Akkadian Kineiform inscribed on palace walls 195 00:13:38.200 --> 00:13:42.960 and clay tablets in ancient Mesopotamian cities, linear elamite was rare. 196 00:13:43.480 --> 00:13:46.000 Writing didn't seem to play the same central role for 197 00:13:46.039 --> 00:13:50.279 the Elamites. Dass says, after much floundering and frustration, he 198 00:13:50.360 --> 00:13:56.360 found himself in Roya Mabubian's elegant North London apartment, examining 199 00:13:56.399 --> 00:14:00.279 an academic gold mine seven hundred and fifty nine linear 200 00:14:00.399 --> 00:14:05.519 elamite signs engraved on ten cups and fragments, plus inscriptions 201 00:14:05.519 --> 00:14:09.039 in Cuneiform. I was thinking eureka, he told me. But 202 00:14:09.159 --> 00:14:12.559 I was also skeptical. Everybody in the academic world was 203 00:14:12.600 --> 00:14:15.120 sure it was all fake. But to say, nurse to 204 00:14:15.200 --> 00:14:19.240 conviction that the Kunnki had to be authentic. No forger 205 00:14:19.320 --> 00:14:23.000 could have strung together so many linear elamite signs that 206 00:14:23.120 --> 00:14:27.799 actually contained bits of coherent text. Later tests by a 207 00:14:27.799 --> 00:14:32.639 metallurgists in Italy confirmed that the Kununki were made of 208 00:14:32.679 --> 00:14:36.919 a ninety percent silver ten percent copper alloy, consistent with 209 00:14:36.960 --> 00:14:40.440 other vessels from the period. Even though philologists had long 210 00:14:40.440 --> 00:14:44.519 ago determined that the language spoken in ancient Elam had 211 00:14:44.600 --> 00:14:48.039 no known relatives. To say, did have one tool at 212 00:14:48.039 --> 00:14:51.600 his disposal, something that for centuries had aided those who 213 00:14:51.639 --> 00:14:55.600 study antiquity. Sometimes, when a message was committed to writing 214 00:14:55.639 --> 00:14:58.039 on a templet or a piece of art, the same 215 00:14:58.120 --> 00:15:01.480 or similar information was in described in several languages and 216 00:15:01.519 --> 00:15:05.120 scripts at once. This practice helped the scribes reach the 217 00:15:05.200 --> 00:15:09.440 largest audience possible. Thousands of years later, that habit has 218 00:15:09.519 --> 00:15:13.480 helped philologists who can compare deciphered writing with the adjacent 219 00:15:13.600 --> 00:15:18.879 undeciphered script to say had just such a bilingual inscription 220 00:15:19.039 --> 00:15:21.960 that he hoped could provide clues to some of linear 221 00:15:22.039 --> 00:15:26.440 Elamite's phonetic values. Archaeologists at SUSA in nineteen o three 222 00:15:26.840 --> 00:15:31.279 had turned up a limestone artifact called the tabreu aulon 223 00:15:31.799 --> 00:15:34.879 now on display at the Louver. The kings of Elam 224 00:15:35.320 --> 00:15:39.240 had often inscribed linear Elamite side by side with inscriptions 225 00:15:39.279 --> 00:15:44.279 in Akkadian Cuneiform, the writing of nearby Mesopotamia, which had 226 00:15:44.279 --> 00:15:47.759 been deciphered in the eighteen fifties. In nineteen o five, 227 00:15:47.840 --> 00:15:54.279 the German linguist ber Dinand Bork noticed a sequence of 228 00:15:54.279 --> 00:15:58.559 four linear Elamite characters on the tabre au leon that 229 00:15:58.759 --> 00:16:03.320 appeared twice. The adjacent Acadian text also had four characters 230 00:16:03.320 --> 00:16:07.440 that showed up with the same frequency. These kittiidformed signs 231 00:16:07.480 --> 00:16:13.519 formed part of two names, Puzzuer Sushniak, the last king 232 00:16:13.600 --> 00:16:19.200 of Elam's Awand dynasty, and the Elamite god in Sushnak. 233 00:16:19.840 --> 00:16:23.519 From these parallel sign patterns, Bork deduced that the Elamite 234 00:16:23.639 --> 00:16:28.600 characters must be su Chi, Nah, and K. That gave 235 00:16:28.679 --> 00:16:31.919 him pooh, Zu and R as well, for a total 236 00:16:31.919 --> 00:16:36.080 of seven characters. From there, the decipherment stalled. The trouble 237 00:16:36.240 --> 00:16:39.440 was for decades, there just wasn't that much linear Elamite 238 00:16:39.519 --> 00:16:44.039 in existence. Could de Say's new trove change everything. Early 239 00:16:44.080 --> 00:16:46.919 one morning in spring twenty seventeen, he sat before his 240 00:16:47.000 --> 00:16:51.000 laptop in his apartment near the University of Tehran where 241 00:16:51.000 --> 00:16:54.759 he was teaching. The screen glowed with linear Elamite characters, 242 00:16:55.000 --> 00:16:58.279 which to Say had copied by hand and digitized from 243 00:16:58.360 --> 00:17:01.159 the photos he had taken. His eyes settled on a 244 00:17:01.200 --> 00:17:04.839 four sign pattern from one of the Kunanki. It showed 245 00:17:04.920 --> 00:17:07.839 up repeatedly. He recognized the first sign in the sequence 246 00:17:07.880 --> 00:17:11.680 immediately it was she as Bork had first seen in 247 00:17:11.799 --> 00:17:16.480 Pasur Sushnak. Then came three unknown signs, the last two 248 00:17:16.480 --> 00:17:19.200 of which were identical. In a flash, to Say could 249 00:17:19.240 --> 00:17:22.880 see it. The duplicated sign was ha, which formed the 250 00:17:22.960 --> 00:17:28.640 last syllables of sie ha Ha, an early second millennium 251 00:17:28.640 --> 00:17:32.559 b c elamite king. Minutes later, the Say spotted another 252 00:17:32.599 --> 00:17:37.839 repeated sequence with one recognizable sign R in Kineiform texts. 253 00:17:37.920 --> 00:17:41.839 He knew Silkha was often mentioned with who is most 254 00:17:41.920 --> 00:17:46.200 likely his father? E PARTI. In fifteen minutes, to Say 255 00:17:46.200 --> 00:17:49.519 had acquired the values of five signs. Thrilled, he plugged 256 00:17:49.519 --> 00:17:53.160 the sounds into the texts and spotted other names, leading 257 00:17:53.200 --> 00:17:57.079 to as identification of more signs. For the next few years, 258 00:17:57.079 --> 00:17:59.920 to Say continued making progress. He had assumed from the 259 00:18:00.119 --> 00:18:03.359 start of his studies that linear elamite, like all other 260 00:18:03.440 --> 00:18:06.799 writing systems from the early Bronze Age, was mixed. I 261 00:18:06.880 --> 00:18:10.599 was looking for logograms, saying where the hell are they? 262 00:18:10.640 --> 00:18:13.640 He said, But as is knowledge of the signs deepened, 263 00:18:13.880 --> 00:18:17.519 to Say took what he calls a creative leap. Linear Elamite, 264 00:18:17.559 --> 00:18:21.319 he now perceived was a purely phonetic script with seventy 265 00:18:21.359 --> 00:18:26.680 seven signs, including five vowels and twelve consonants. Until now, 266 00:18:26.799 --> 00:18:30.400 is widely accepted among linguists that the oldest purely phonetic 267 00:18:30.440 --> 00:18:36.880 alphabet was protor proto Siniactic, a Middle Bronze Age script 268 00:18:36.920 --> 00:18:40.279 from the Sinai Peninsula which appeared five hundred years after 269 00:18:40.359 --> 00:18:46.960 linear Elamite. Desay's analysis, if verified by other philologists, could 270 00:18:47.000 --> 00:18:49.640 force a radical reconstruction of the history of writing and 271 00:18:49.720 --> 00:18:53.559 human progress. It would reorder the chronology of phonetic writing, 272 00:18:53.880 --> 00:18:57.880 shifting the focus away from the Levant to the Iranian plateau. 273 00:18:58.319 --> 00:19:02.119 It would also elevate the previously overlooked Elamite kingdom to 274 00:19:02.319 --> 00:19:06.640 a primary place in human intellectual development. Seven years after 275 00:19:06.680 --> 00:19:10.799 that encounter in London, in twenty twenty two, to Say 276 00:19:10.839 --> 00:19:15.359 and for colleagues authored an article in Zeitschrift for a 277 00:19:15.519 --> 00:19:23.160 Syriologui and for Denayitious Archaeology Journal for a Syriology and 278 00:19:23.319 --> 00:19:27.880 Near Eastern Archaeology, a respected journal published in Berlin. The 279 00:19:27.880 --> 00:19:31.960 phonetic values of every linear Elamite character they declared had 280 00:19:32.000 --> 00:19:35.519 been deciphered at last. The claim was remarkable. If true, 281 00:19:35.519 --> 00:19:38.240 it represents the first time that all the sounds of 282 00:19:38.279 --> 00:19:41.240 an ancient script have been figured out in decades, but 283 00:19:41.359 --> 00:19:44.440 to say in his team, he admits heavn't accomplished a 284 00:19:44.519 --> 00:19:48.160 one hundred percent decipherment of the writing system. According to 285 00:19:48.200 --> 00:19:52.559 most philologists, real decipherment occurs only when a script's sounds 286 00:19:52.559 --> 00:19:57.519 and its meaning are both understood. Champollion had determined the 287 00:19:57.720 --> 00:20:01.079 meaning of hieroglyphs after sounding out some wordsords and recognizing 288 00:20:01.160 --> 00:20:05.519 that the language must be a direct ancestor of Coptic Egyptian. 289 00:20:06.000 --> 00:20:10.240 Hinks and Rawlinson were able to understand Akkadian Cuneiform after 290 00:20:10.319 --> 00:20:14.559 perceiving that the underlying language closely resembled Hebrew, but the 291 00:20:14.640 --> 00:20:19.599 language of linear Elamite, known simply as Elamite, remains largely 292 00:20:19.640 --> 00:20:23.839 a mystery. Thanks to the inscriptions on the Kunangi, Dissey 293 00:20:23.960 --> 00:20:26.519 has managed to tease out not just the names of 294 00:20:26.559 --> 00:20:30.160 places and kings, but also a smattering of titles, epithets, 295 00:20:30.519 --> 00:20:34.799 common nouns, adjectives, and verbs. One silverbeaker, he suggests, was 296 00:20:34.839 --> 00:20:37.640 an offering given by a ruler to the Elamite supreme 297 00:20:37.680 --> 00:20:42.440 god E pala Ishan mighty Lord. The inscription reads, in part, 298 00:20:42.880 --> 00:20:47.519 I am the servant of napiaetia. Dissey found the words kare, 299 00:20:47.880 --> 00:20:52.599 which he translates as devotion or worship, and zeeni a 300 00:20:52.640 --> 00:20:55.960 divine blessing bestowed by a god on his royal subject. 301 00:20:56.480 --> 00:21:00.359 Other words including zempt for king, hurt for people, and 302 00:21:00.480 --> 00:21:04.599 shock for sun also became clear in context. It shines 303 00:21:04.640 --> 00:21:07.960 a little light on this long vanished place, he says, today, 304 00:21:08.400 --> 00:21:12.440 of the eighteen hundred sixty three linear Elamite signs that 305 00:21:12.519 --> 00:21:15.599 exist in the Corpus des Asis, he is able to 306 00:21:15.680 --> 00:21:18.880 sound out eighteen hundred ten of them. The rest have 307 00:21:18.960 --> 00:21:23.759 eroded into illegibility, but he acknowledges he can make sense 308 00:21:23.759 --> 00:21:25.960 of only a few words. I am still facing a 309 00:21:26.000 --> 00:21:29.720 lot of problems with the translation, he told me. He 310 00:21:29.759 --> 00:21:34.160 faces critics too. Jak Abdahl and Oxford professor considered one 311 00:21:34.160 --> 00:21:40.200 of the world's foremost scholars of Mesopotamia, dispute Desay's assertion 312 00:21:40.359 --> 00:21:44.440 that linear Elamite is a purely syllabic script, that part 313 00:21:44.519 --> 00:21:47.119 of the decipherment is certainly not correct. He told me, 314 00:21:47.559 --> 00:21:51.160 I would suspect there were logograms as well. He also 315 00:21:51.200 --> 00:21:53.839 says that so much about the script remains unknown. The 316 00:21:53.920 --> 00:21:56.720 meaning of many words, the grammar the values of certain 317 00:21:56.799 --> 00:22:01.119 signs that Dessay's claims of victory are wildly premature. I 318 00:22:01.119 --> 00:22:03.839 have little patience for des say. Dal told me much 319 00:22:03.880 --> 00:22:07.799 of it is complete rubbish. Dessay has tried to shrug 320 00:22:07.839 --> 00:22:11.960 off the criticism since the announcement of the linear Elamite decipherment. 321 00:22:12.359 --> 00:22:14.759 He told me, escorting me to the door of his apartment, 322 00:22:15.119 --> 00:22:17.759 I have found new friends, and I have found new enemies. 323 00:22:18.359 --> 00:22:21.920 His next undertaking is an attempt to decipher proto elamite, 324 00:22:22.279 --> 00:22:26.880 a precursor to linear elamite that first appeared in the 325 00:22:26.880 --> 00:22:31.640 fourth millennium BC and represents the earliest stage of civilization 326 00:22:31.839 --> 00:22:35.599 in Iran. The writing consists of four hundred to eight 327 00:22:35.680 --> 00:22:40.240 hundred characters, the hall mark of a lago graphic syllabic system. 328 00:22:40.599 --> 00:22:42.880 It is a total mystery, he says, just the kind 329 00:22:42.920 --> 00:22:47.480 of challenge that code breakers love, bringing dead words to life. 330 00:22:47.720 --> 00:22:50.960 Archaeologists excavating the ancient city of Susa in nineteen o 331 00:22:51.079 --> 00:22:55.000 three uncovered the Tabra Leon, which had a four thousand 332 00:22:55.119 --> 00:22:58.960 year old bilingual inscription in known a Kadian cuneiform and 333 00:22:59.119 --> 00:23:03.359 undeciphered linear elamite. Though this was a promising first step, 334 00:23:03.400 --> 00:23:07.319 the dearth of signs stalled decipherment until twenty fifteen, when 335 00:23:07.400 --> 00:23:11.039 Francois de Say gained rare access to silver beakers with 336 00:23:11.200 --> 00:23:14.200 hundreds of new L E signs and piece together what 337 00:23:14.319 --> 00:23:17.839 could be the world's first phonetic writing system. Here's how 338 00:23:17.839 --> 00:23:22.400 he did it. Identifying proper nouns. Early on, linguists determined 339 00:23:22.400 --> 00:23:25.759 that the table leon bore the same names in L 340 00:23:25.839 --> 00:23:30.559 E and a kate Acadian kineiform king Pussur Sashnak and 341 00:23:30.680 --> 00:23:36.559 the Elamite king Ishnek. The tablet's linguistic patterns were a 342 00:23:36.640 --> 00:23:39.240 jumping off point for de Say to later identify other 343 00:23:39.319 --> 00:23:44.519 names and royal titles decoding phrases. When to Say secured 344 00:23:44.559 --> 00:23:47.319 access to a private collection of beakers with both a 345 00:23:47.440 --> 00:23:50.519 Kadian kineiform and L E signs that had never been 346 00:23:50.519 --> 00:23:53.960 seen publicly, had more pieces to the puzzle. This allowed 347 00:23:54.039 --> 00:23:57.240 him to string together one of the first sentenced fragments, 348 00:23:57.240 --> 00:24:01.279 a huge breakthrough in cracking L E. Whence the script 349 00:24:01.359 --> 00:24:03.759 was deciphered to Say, says, it was possible for him 350 00:24:03.759 --> 00:24:06.759 to actually read lines of L E like the text 351 00:24:07.000 --> 00:24:11.359 carved into stone blocks from Suza, offering an ode to 352 00:24:11.559 --> 00:24:17.200 Pusar Sushnak and a warning to his enemies. Next the 353 00:24:17.319 --> 00:24:21.720 industrial waste site that glitters like a glacier in northwest India, 354 00:24:21.759 --> 00:24:25.079 A field full of marble debris has another worldly allure 355 00:24:25.759 --> 00:24:32.720 by Puja Chianghwai Waila. The first time Atul Kharasya saw 356 00:24:32.799 --> 00:24:36.680 a YouTube video of the dumping grounds in Kishargach, he 357 00:24:37.039 --> 00:24:39.279 knew he had to visit the place, a pearly white 358 00:24:39.359 --> 00:24:42.799 landscape shining like a snow field atop a hot dry 359 00:24:42.839 --> 00:24:45.640 plain in northwest India. It was a six hour train 360 00:24:45.759 --> 00:24:49.440 ride from his home in Delhi, But for Charisiya, a 361 00:24:49.599 --> 00:24:54.440 travel influencer who makes scenic the social media media villages 362 00:24:55.240 --> 00:24:57.799 videos for a living, the trip paid off with an 363 00:24:57.880 --> 00:25:02.720 Instagram reel that racked up a quarter million likes. Kishagar's 364 00:25:03.359 --> 00:25:07.440 peculiar attraction is built on a byproduct from its marble 365 00:25:07.519 --> 00:25:10.920 processing industry, marble slurry, a mix of water and fine 366 00:25:10.960 --> 00:25:15.720 particles of calcium, carbonate and other minerals. Tanker trucks deliver 367 00:25:15.839 --> 00:25:18.680 loads of it each day to the city's outskirts hauled 368 00:25:18.720 --> 00:25:21.720 in from more than twelve hundred facilities that cut, grind 369 00:25:22.039 --> 00:25:25.599 and polished stone from the quarries that proliferate in the 370 00:25:25.640 --> 00:25:30.000 surrounding state of Rajasthan. As water evaporates from the slurry 371 00:25:30.079 --> 00:25:34.319 in the semi arid climate, the mineral particles recrystallized to 372 00:25:34.359 --> 00:25:38.240 form bone white mounds and terraces. Tourists started showing up 373 00:25:38.240 --> 00:25:41.400 to admire them about a decade ago after a popular 374 00:25:41.720 --> 00:25:46.279 comedian shot a movie scene at the site, says Sampat 375 00:25:46.519 --> 00:25:53.680 Rai Sharma, CEO of the Kishkar Marble Association KMA, which 376 00:25:53.920 --> 00:25:58.359 manages the yard. Then came Instagrammers and YouTubers, engaged couples 377 00:25:58.400 --> 00:26:02.880 with hired photographers, even Bollywood crews. As the crowds grew, 378 00:26:03.000 --> 00:26:08.599 the KMA rebranded the spot the Kishnagarch Snowyard. Soon vendors 379 00:26:08.640 --> 00:26:11.599 were offering everything from snacks to horse rides to a 380 00:26:11.680 --> 00:26:15.200 dinghy for plying the electric blue waters of the artificial lake. 381 00:26:15.720 --> 00:26:21.759 But even as hashtag dumping yard Kishangar has trended among 382 00:26:21.799 --> 00:26:25.000 the travelers, some researchers have raised questions over the site's 383 00:26:25.039 --> 00:26:31.000 health and ecological impacts. Environmental scientist Lakshmi Kant Sharma of 384 00:26:31.079 --> 00:26:34.960 the Central University of Rajasthan says a study he led 385 00:26:34.960 --> 00:26:38.880 in twenty twenty three and twenty four, currently under peer review, 386 00:26:39.279 --> 00:26:44.200 revealed worrying effects in areas surrounding the dump site. Air 387 00:26:44.319 --> 00:26:48.279 quality tests, he says, registered levels of particulate matter beyond 388 00:26:48.319 --> 00:26:52.839 the World Health organizations recommended limits, while soil samples showed 389 00:26:52.920 --> 00:26:58.400 high salinity and heavy metal contamination and groundwater samples exceeded 390 00:26:58.400 --> 00:27:02.119 limits for chemicals like chlor ride and iron. A different 391 00:27:02.160 --> 00:27:06.519 study published in February twenty twenty five found naturally occurring 392 00:27:06.519 --> 00:27:11.880 asbestos fibers in marble dust from another rasash San slurry yard. 393 00:27:11.960 --> 00:27:15.000 For his part, the K m as Sharma disputes such 394 00:27:15.039 --> 00:27:19.000 studies validity and says his organization has received no health 395 00:27:19.039 --> 00:27:24.000 related complaints and while several Indian newspapers have reported the 396 00:27:24.200 --> 00:27:29.000 Kuraje study preliminary findings, at least one vendor suggests it 397 00:27:29.039 --> 00:27:35.160 has an affected visitation. Gurad Yain manages snow Moments, which 398 00:27:35.200 --> 00:27:37.400 opened to kiosk at the yard two years ago, to 399 00:27:37.440 --> 00:27:41.319 rent photo back drops and props. We do thirty to 400 00:27:41.359 --> 00:27:44.599 thirty five shoots a month. Jane says the demand is 401 00:27:44.720 --> 00:27:49.200